Save up to 40%

When Buying Hearthstone Packs!

Limited Time Offer from Amazon!

SAVE NOW!
Rating  68

Contributed by

Smashthings

Guide Type

Last Updated

May 30, 2015

Table of Contents

How to Play Turns Two & Three in Arena

Introduction


Hello guys, welcome to my second Hearthstone Players Guide. If you have read my first article then you have a good idea what this guide is going to be like. For everyone not familiar, let me start by saying its going to be a long ride with a mixture  theoretical abstraction and practical examples. The focus of this guide is how to play  turns two and three of an Arena game. With that said however, many of the points I make can be generalized to other turns of a game. And also, there is plenty of stuff that constructed players may find useful.

Intended Audience: This guide is aimed at intermediate and advanced players looking to ‘go infinite’ (i.e. average about 7 wins). While I sincerely hope beginners read this guide and learn a lot I fear that the content maybe a bit too difficult for such a reader. If I’m describing you and you wish to persevere don’t say I didn’t warn you!

Guide Contents: Let me quickly describe the journey we are going to take;  I start the guide with an explanation as to why I think Turns two and three in Arena games are important enough to justify writing an entire article about them. After thats done I show you a method for systematically approaching these critical turns. After a little bit more theory I end the guide by providing you four detailed examples of how to approach turn two. These final examples are in a ‘pop-quiz format‘, which should give you a chance to practice using some of the ideas I detailed in previous sections. Once that’s done, all you have left to read is the comments section. On the topic of comments, please feel free to ask questions and don’t forget to tell me how awesome I am. :)

Anyway, lets get cracking!

Introductory Example


View post on imgur.com

If you were to play the position in the picture what would you play here?  Have a quick think about it.

This image is take from a game Adwtca played in his  ‘Arena Coop‘ series (he also has written Arena guides published on this website). The duo instantly reach for the coin because they know that they want to play a 2-drop but then the critical question looms; What two drop do we play?  Recombobulator or Shielded Minibot?

The duo discussed the merits of each play; the advantage of playing the minibot first is that if you can trade into something (losing the divine shield) and then cast recombulator then the battlecry ought to generate a little bit of value (since most 2-drops are better than a 2/2).

But the position is a little bit more complicated than that; if the enemy Mage plays a 2/3 minion and then you try to transform the minibot you are likely to get a minion worse than a 2/2 with divine shield. Moreoever, their 2/3 can just trade into the 3/2 recombobulator.

Lets listen to their discussion (you can stop watching the clip after the Paladin plays Avenge, which is at (1:13:28) ) :

I show you this clip because it should help you understand what this guide is all about. By the end of this guide I hope that each and every one of you will be capable of thinking about turns two and three in this much detail. If you are able to achieve this, then you will be one step closer toward becoming infinite.

**Update (22/01/15):  You might be wondering the following question; Was Adwtca right to play the Recombobulator first?  In the comments section, Mr.Shine and I discuss what we think the correct play is. Update (02/02/15): Adwtca has also chimed in. **

At the very end of the guide I have four examples that are like this one be sure to check them out. In the meantime however, we need to start the guide with some more basic introductory material. For example, I should probably try to explain why I feel turns two and three are of critical importance. This will be the next section.

Why Turns Two & Three?


If you have played Arena (especially since GvG) you will have probably realised that most games are won and lost based on who has board-control.  There are a number of reasons for why this is the case, I’ll offer three reasons here:

(1) Arena has worse ‘Board-Clear’:

In Arena, it is generally the case that you are usually less able to clear an enemies board than you are in constructed.

For example, ‘Paladin Control’ (in constructed) is almost without rival when it comes to board clear; they can run combos like ‘Equality+ Wild Pyromancer‘ which means that if they do start to fall behind they have an excellent chance to stabilise for a mere four-mana. Moreover, the Paladin can have two copies of each card (which makes the combo reliable).  In Arena however, not only will you often fail to draft such a combo, even when you do draft it chances are you won’t draft enough copies of either card to yield consistent results.

Classes like Warrior usually do okay for board-clear in constructed, they just run one — sometimes two– copies of Brawl. But in Arena, brawl — owing to its ‘epic’ status– is not a card you will often draft. In Arena, Warrior is usually forced to rely on neutral minions (e.g. Abomination) to get the job done.

As for Priest, in constructed the usual trick to use Circle of Healing alongside Auchenai Soulpriest. In Arena however the problem is twofold; firstly, the soulpriest is a ‘rare’. Secondly, Circle is usually a bad card by itself, making it a questionable pick in most cases (for these reasons Priest most frequently relies on Holy Nova).

For Rogue, Fan of Knives doesn’t usually deal enough damage and Vanish only solves the present problem for a turn. Therefore, this class would prefer for its AoE to come from Blade Flurry, which is a rare. Just like priest, Rogue now has to draft the relevant supporting cards (i.e. Weapon-damage stuff) or else he/she is left with an inferior Arcane Explosion.

As for Druid, well, even in constructed the class usually at board clear :).

The above paragraphs seemingly makes the case then that most classes struggle with board clear due to problems concerning card rarity (e.g. Warrior) and/or the difficulty is due to the requirement of drafting multiple cards to make the combos possible (e.g Rogue, Paladin). Worse still, for some classes (e.g. Priest), the combos require the drafting of otherwise bad cards!

But none of this is true of Mage! When a Mage drafts a few copies of Flamestrike they don’t tend to have any problems clearing the enemies board.  And thats exactly the point!  Mage is widely regarded as the ‘Queen of Arena’ in large part due the power — and ‘common’ rarity– of the card ‘Flamestrike’.

Basically, a general lack of (effective) board clears in Arena typically means that overextending is less likely to be punished in Arena than it is to be punished in Constructed (except against Mage!).  On a related point, in Arena Mage and Warlock (due to cards like Shadowflame, Hellfire) are typically the only classes that regularly deal more than 2 AoE damage (without Spell power). This often means that it is easy to play around a lot of Arena board clear (e.g. just drop 3-health minions). In constructed though, many decks can squeeze extra juice out of the cards (making them harder to play-around). Warrior, for example, can combine Whirlwind or Death's Bite and Baron Geddon for a total of 3 AoE damage. Priest can likewise squeeze more juice out of cards like Holy Nova by dropping a W.Pyromancer first.

(2) Arena decks typically have fewer ‘swing cards’ than constructed decks:

In constructed there are a number of cards (usually epics and Legendaries) that you can put in your deck to provide you with those critical swing turns. For example, you could always top deck The Black Knight to deal with the enemies 5/10 Ancient of War. Such a swing can be devastating enough to bring you right back in the game. Whereas in Arena, it will be a very rare scenario to have the Knight in your deck. Moreover, you may not have even drafted a silence either! Thus, it is quite often the case that your only option is try to punch through a 5/10 brick wall!

Talking of big minions the counter (in constructed) is running a copy of  Big Game Hunter. But in Arena you can’t rely on this card since its an  ‘epic’. As a consequence of this I usually find that massive minions, such as an 8/8 Sea Giant, are likely to live a turn or two longer in an Arena game than it typically would do in a constructed game. And even if it is taken down, chances are the Arena solution is less efficient than the Constructed solution (e.g BGH costs three-mana and puts a 4/2 on board. Whereas Assassinate  is two additional mana and doesn’t add anything to your side of the board).

Long story short, there are fewer ‘big swing’ turns in Arena. Moreover dealing with threats (such as an 8/8 giant)  is often more difficult in Arena than it is in constructed, which therefore means once a threat(s) is/are on the board it is much harder to simultaneously counter and play ones own threats.

(3) Arena decks have less ‘identity’ than Constructed decks. 

**By “identity” I simply mean that constructed decks typically follow a much tighter ‘theme’ (e.g. undertaker decks will often narrowly focus on getting deathrattles into the deck) than Arena decks.**

When was the last time you saw a Murloc deck in Arena?  When was the last time you saw Freeze Mage in Arena?  You typically won’t see either of these archetypes in Arena essentially for the same reason. Namely, the inherent randomness involved in drafting typically means that decks that try to do specific things (in this case swarm with Murlocs, or win with freeze mage) will often fail to achieve a critical mass of staple cards and/or will fail to draft key cards altogether.

The result is that it is hard to draft specialized decks with unique win conditions. Therefore, it is regrettably the case that that most decks default to being either ‘Aggro board-control‘ or ‘Mid-range/Control board-control‘ decks.

This difficulty is ultimately compounded by the drafting process. If you check out this draft it is funny to note just how many Mind Blast‘s ‘adwcta’ & ‘Mr. Shine’ were offered. If they could redo the draft completely (knowing what each future choice was going to be) there is a reasonable chance that they may have picked up every single mind blast offered and thereby create an insane ‘to the face deck‘ (there was even a few other choice cards like Arcane Golem on offer that could have complimented that tactic nicely). But the two Players (quite rightly, I should add) did not draft this deck. When the first mind blast comes up early in a Priest draft you have no idea if you are going to get enough copies of other key cards to make a face deck viable. And since Mind Blast is pretty weak in standard board-control decks the risk of the mind blast pick backfiring is pretty high. Only in hindsight can we definitively say that the face deck was viable.

In short; its rare to see specialized decks in Arena because you are unlikely to draft that much needed critical mass of key cards. Far sadder still, most of the time you won’t draft the first key card because of the risk of not getting the rest of the cards you need to — complete the strategy– is high.

So that’s the three reasons why I think Arena typically ends up being all about board-control. The first two reasons suggest that an advantage on the board can quickly snowball into a win due to the difficulty involved in playing from behind. The third reason suggests that other sorts of decks will not often be drafted owing to the risks involved in making them. But all this analysis does not (as yet) explain why turns two and three are of critical importance.  I’ll quickly offer two reasons in support of that claim:

(1) The Card rarities are not equally distributed across all mana costs:

If you search the entire roaster of Hearthstone cards you will find that there seems to be a link between mana cost and card rarity.  If you look at the eight (or more)-mana minions for example, there are more legendaries than there are rares, commons and epics combined. And now if you look at the lower-end of curve you will see the opposite (i.e. very few legendary cards relative to the number of commons). Given that most of the time you will be drafting commons, it naturally follows that you will be offered more two or three-mana cards than you will be offered eight(or more)-mana cards. Ergo, due to card rarity, decks tend to be pushed toward the lower-end of the mana curve. Druid is one of the few classes you tend to see playing late-game minions; Why is that the case? Well, it probably has something to do with the fact that Druid has access to the biggest common-rarity minion in the game, Ironbark Protector. (Innervate helps as well of course :) )

Basically, if most Arena decks are going to be full of lower-cost cards, then it follows that there is going to be a lot of minion fighting early on as both players jockey for position.

It is worth pointing out however that this is not a general claim about Hearthstone, rather, it is merely a claim about all the cards released so far (for those of you reading in the very distant future; GvG is the current set). Future sets may address this issue.

(2) If Board Control ‘snowballs’ games, it makes sense to try and claim it for ourselves as soon as possible:

If you recall, the previous three points explaining the importance of board-control essentially made the point that once a Player falls behind it is often difficult to bounce back. This is because; (1) there is less and/or weaker board-clear, (2) fewer swing cards, and (3) its hard to build build a deck that doesn’t care about the board. It logically follows from these points then that having board-control makes us more likely to win the game.  And since we want to gain control of the board before our opponent does, it makes sense for use to start the fight ‘as soon as‘ we possibly can.

Well, I say ‘as soon as possible’ but I tend to find turn-one unimportant. So by ‘as soon as’ I actually mean turn-two and turn-three.  Why I don’t consider turn-one important shall be the contents of the next section.

What about Turn-One?


Okay, so the previous section tried to making the point that establishing board presence early in the game is often the key to victory. You can’t get much ‘earlier’ than turn one, so that begs the question; Why do I not consider turn one as important as turns two and three?

The answer is actually pretty simple; the overwhelming majority of one-drops suck. And in Arena, even some of good one-drops suck (e.g Undertaker)!  Here are three reasons why I feel this is the case:

 (1) Hero Powers:

In Hearthstone, it is just all too easy  to kill a Leper Gnome with your face. While you can take a decent chunk of damage doing so, the fact of the matter is that the card advantage gained usually makes it worthwhile.  Meanwhile, killing your typical two-drop with Hero Power takes twice as much time, twice as much mana AND you will often take twice as much damage in the process. It is simply the case then that the 2-drops represent the first minions who must be combated with cards in order to be effectively defeated.

Moreover, suppose for a second that your Druid opponent has a terrible hand. In this case, he was going to play his/her Hero Power on turn-two regardless of what we did. If we skip turn-one this hero power ends hitting our face for  a negligible amount of  damage. Whereas if we play a one-drop then we make his/her use of hero power purposeful.  In short, we inadvertently gave our opponent a good play (he/she now gets +1 card advantage, as opposed to getting a mere +1 face damage). Their hand may remain terrible for a while, but the fact is the +1 card advantage could end up being a game-winning advantage if he/she is somehow able to stabilise later on.

(2) Most one-drops have terrible abilities (for Arena):

Compared with decent two-drops, one-drops typically have terrible abilities (notable exception; Northshire Cleric), which therefore tends to make the one-drops less useful and less flexible.  Compare, for example, the usefulness of Acidic Swamp Ooze with Bloodsail Corsair.  If the average weapon has two durability, then we think of Ooze’s ability as being twice as good in most situations (but thats okay since Ooze is twice the mana!).

(3) Lots of  One-drops don’t want to be played on Turn-One .

This point is pretty simple. The are only a small handful of one-drops that are actually good turn-one plays. A Card like Abusive Sergeant is an okay Arena pick since the ability can be useful later on in the game, but to play it on turn one would be a sub-optimal use of the card (since the battlecry fails to trigger). Young Priestess is exactly the same. Personally I quite like drafting Southsea Deckhand in Rogue, but I hardly ever play it on turn one (I prefer to wait until I have a weapon equipped and/or I need to utilise the combo mechanic).

Basically, of all the one-drops you (or your opponent) might draft very few are actually strong turn one plays and so therefore most of the time you are more likely to see a coined two-drop on turn one than an actual one-drop played on the first turn of the game. **Update 02/02/15: The fact that you can coin a two-drop in roughly half your games gives you yet another reason to not draft 1-drops in Arena.**

To sum up;  most of the one drops currently in the game are rubbish.  This means most people (who know what they are doing) won’t draft a lot of 1-drops (if any at all). Meanwhile, 2-drops are frequently drafted. The result of all of this is that most of the early-game fighting is done on turns two and three.

This is not to say all one-drops are bad however, there are a few that stand out above the rest.  Broadly speaking they fall into a few different categories:

Category

Example Card(s)

The ‘I-got-big-stats’ group.  Zombie Chow, Flame Imp
The ‘who-cares-if-I-die?’ group.  Webspinner, Clockwork Gnome
The ‘Traders’ group.      Worgen Infiltrator
The ‘My-card-text-is-actually-useful’ group.  Northshire Cleric, Southsea Deckhand
The ‘Feed-Me!’ group.    Mana Wyrm

The ‘I-can-survive-longer-than-a-single-turn’ group.

 Argent Squire

 

Although my categories are obviously tongue-in-cheek the table nonetheless lists pretty much every one-drop currently in the game worth drafting. Since this guide is not supposed to be about specifically telling you what cards you ought to draft, I won’t go into any more detail than I already have.

A quick note on Mulligans


And by ‘quick note’ I really mean that.  In this guide I don’t really say much about Mulligans. In the next two sections (‘thinking about turn two’, ‘thinking about turn three’) I try to teach you what you need to think about on those turns. Well, you should also be thinking about that stuff during the mugilan too!  Having an explosive turn two/three ultimately requires you to have the pre-requisite cards in hand. And obviously, one of the main ways to get those key cards in the hand is to make those keep/throw mulligan decisions wisely.

With the introduction now over,  lets actually get to the meat of this guide.

Thinking about Turn Two


During turn two what should we be thinking about? Well, here’s a handy list of questions you need to ask yourself:

  • What happens if they use Hero Power?
  • What happens if they play a 3/2 minion?
  • What happens if they play a 2/3 minion?
  • What happens if they play a minion with some other stat distribution (e.g. a 1/3)
  • What happens if they play a special minion?  (I’ll define the term “special minion” in a moment)
  • What spells, secrets, or weapons could they play?
  • If there are already minions on the board, what attacks/trades can they make.
  • If they have The Coin, what could they do with it.
  • What class specific cards/strategies might they employ?  (will not be covered in this guide).

You can see that I crossed  the last item off the list. This is because this guide is not intended to be a “how to beat every class” guide. Rather this guide is meant to be generalised and not specific to any given meta. With this said, I will point out that when playing against a Paladin (especially at high wins) you ought to expect a turn-two Shielded Minibot. I shall say no more than that.

So we now know what we need to think about on turn-two; for every action (be it playing a card/hero power) you are thinking about playing you need to go through this list. Lets start with a simple example:

Imagine that your hand has two cards in it; Loot Hoarder and Bloodfen Raptor.

It is turn two and you are interested in figuring out the best play. So what do we do?  Well, we simply go through the list of questions!

  • If we play Loot Hoarder:

…and they respond with Hero power.   If it is a Rogue/Mage/Druid then our minion dies. Usually, losing a minion this easily would be a bad thing (see section about one-drops), but as it so happens loot hoarder has a deathrattle that draws us a card. This means that our Opponent has not gained card advantage by killing Loot H. Ergo, against these classes, our position is okay  Against other classes (such as Shaman, Priest, Warrior, warlock) we are in a good position (unless the shaman rolls the 1/1 totem) since these Hero Powers do not threaten Loot H .

…What if they respond not with Hero Power, but with a 3/2 minion?   Well, this is fantastic news for us! Loot H will get good value. Killing a bigger minion with a smaller one is usually a good idea. Moreover, if the deathrattle resolves then we gain +1 card advantage over our opponent.

…What about a 2/3 minion?  Well this is pretty bad, we technically don’t lose card advantage due to the Loot H’s deathrattle but in this case the 2/1 body achieved little.

…What about if they play a minion with some other stat distribution?  Well it depends what the stats are but I’m sure given what I have already written above you can figure is out :)

…What about special minions?  Lets ignore this question for now, I’ll explain it later.

…What about weapons, secrets and spells?   Well to simplify the analysis we will say that Frostbolt on Loot H is good for us since that generated card advantage (remember its not a 1:1 card trade because of the deathrattle!). A weapon hit is fairly neutral since we neither gained nor lost a card. and if they play a secret then we are probably trading 1:1 or possibly better.

…What about minions already on the board?  What about the coin?   Well, to keep this introductory example as simple as possible lets assume that we play Loot H on an empty board and the coin has already been used.

Okay, that gives us a broad overview as to what might happen if we play Loot H.  It is worth pointing out however, I only evaluated Loot H in terms of card advantage. There are other assessments I could have made, for example,  instead of examining Loot H’s + Opponents Response in terms of card advantage I could also have evaluated the resulting positions in terms of ‘board-control’, ‘tempo’, ‘mana-efficiency’, and so on.  I have not done so here mostly because introductory examples are not supposed to overwhelm you with material!  But with that said, I do want to highlight that if we evaluate Loot H by other metrics then our evaluations as what is a ‘good’ or ‘bad’ outcome is may change.

For example, I claimed that if we play Loot H and they respond with a secret we get a 1:1 trade or better.  And this claim is broadly true; if we use Loot H to attack into a Noble Sacrifice then it is the case that we get +1 card advantage.  But, it is also simultaneously the case that our opponent has dealt with Loot H mana-efficiently (i.e. they spent 1 mana to kill a 2 mana minion). And when we deal with threats mana-efficiently that often entails a tempo advantage that can be converted into some some other lead (e.g. board-control). To demonstrate that point; If our opponent has two mana and they can deal with our minion at the cost of one mana (i.e.noble sacrifice) then that leaves one mana remaining  to play their own threat at the same time, such as a Leper Gnome.  And this position is a very dynamic one;  we end up ahead in terms of card advantage but at the cost of being behind tempo and board control.

But I digress!  Lets now look at what happens if we play Bloodfin Raptor:

  • We play Bloodfen Raptor:

…and they respond with Hero Power.  No class in the game has a Hero Power that can take out a 3/2 in a single strike, but some classes (e.g. Mage, Rogue, etc) are capable of taking it out in two strikes. Against these classes, Blood R is good. Against classes that can’t even threaten a two strike kill (e.g. Priest, Warrior) then this is a fantastic outcome.

…What if they respond not with Hero Power, but with a 3/2 minion?   Well in this case we have the option of a simple 1:1 trade. An even exchange.

…What about a 2/3 minion?  Just as above, this is likely to lead to an even exchange.

…What about other stat distributions?  Well it depends what they are but I’m sure given the above to lines you can figure is out :)

…What about weapons, secrets and spells?   A 2+ attack weapon with 1+ durability is bad for us since in this case our opponent is likely to generate card advantage (on the next swing). Against secrets we will typically get a 1:1 trade (only in rare circumstances will we do better/worse than that), this is also the case with most single target removal spells as well.

 So by comparing these two plays we now know when Loot H is better than Blood R and vice versa. Since we should always assume smart opponents, we now know that if we play Loot  H our opponent is more likely to play  a 2/3 minion (or Hero power) than he/she is to play a 3/2 minion.  So that gives you some insight as to what you need to be thinking about on turn two, lets now move our attention to turn three.

Thinking about Turn Three


What do you need to think about on turn three?   Well basically you ask the same questions as you did in the above section, but this time you adjust your expectations accordingly. So, for example, instead of asking yourself how you are going to respond to a 2/3 or 3/2 minion, the question becomes;  “how do I respond to a 3/3, 3/4, or 4/3 minion?”

Ideally, you should be thinking about turn three when you played turn two.  For example, if you play a 2/3 on turn three and your opponent responds with a 3/3 he/she threatens to win the trade. You should have seen this line of play before dropping the 2/3. And if you did see such a line then–if possible– you should have also prepared the counter (e.g. a Fiery War Axe).

I cannot stress this enough; the ability to play two-to-three turns ahead can yield a significant advantage in Arena (and in constructed, for that matter). The rest of this guide is heavily focused on this idea; I’ll will introduce this idea at the ‘theory level‘ first but the four in-depth examples at the end of this guide should give you a better practical understanding as to how we can ‘link’ turns two and three (and beyond) together. But before that there are few more introductory points one must make. For starters, I have yet to define what a ‘Special minion‘ is. Lets do that now…

What is a “Special Minion” ?


**note that while I talk about “Special minions” there is absolutely no reason why this idea couldn’t also be applied to spells, secrets, weapons, etc **

So what is a special minion?  Basically, it is a minion that usually has an ability that, when played, drastically alters the ‘flow of combat‘.  So, we are looking at minions that can change the combat math, and/or whose abilities can punish unwary players.  This definition is a bit vague, so let me give you a (non-exhaustive) list of ‘special minions’, and hopefully by thinking about what this minions have in common will help you to understand the sorts of minions I am talking about:

Dire Wolf Alpha Flametongue Totem Raging Worgen
Shielded Minibot Aldor Peacekeeper Abusive Sergeant
Blood Knight Ironbeak Owl Mad Bomber
Shattered Sun Cleric  Explosive Sheep  Shrinkmeister


I’m sure plenty of you reading this have stories about one or two of these minions being dropped onto the board to devastating effect. And thats basically the theme that unites the members of this list!  If you (or your opponent) are unprepared for minions like these then the game can very quickly quickly snowball out of control.   Let’s try to understand this snowballing effect by thinking about a few simple examples:

Simple Example #1:

  • I play Loot Hoarder:

    …The response is a 2/3

    ….I play Raging Worgen

In this situation I initially got caught out, our opponent did the smart thing by playing a 2/3 against a 2/1 minion. I used this exact example earlier in the guide (see ‘thinking about turn two’ section). But this time we are looking just one move deeper; the follow up play (on turn three) is to drop a 3/3 with a powerful ability. Two attack is not sufficient to kill the Worgen, moreover at 3 health my opponent is also unable to suicide the 2/3 into Loot H.   So in this case, either our opponent can deal that additional point of damage somehow, or he will find himself fighting against an enraged Worgen.

Simple Example #2:

  • I play Loot Hoarder:

    …The response is mad bomber.  (The bombs kill Loot H).

This example is very simple indeed, we play a weak 2/1 and the opponent responds with a minions whose battlecry has a chance to outright kill our minion (RNG permitting). In such cases as this, the side that played Mad Bomber can end up significantly ahead in terms of card advantage, and/or tempo, and/or board-control.

Simple Example #3:

In this example, I have killed a 3/2 because the dire wolf buffed the creeper with an additional point of attack. Moreover, those new 1/1 spiders are also going to get buffed. Meaning that I am likely to get even more value than I normally would from those minions (e.g. next turn we could trade two 1/1 spiders for a 4 health minion)

Let me share with you a quick video clip. Its a priest mirror match in Arena, and my deck is amazing — without a doubt the best Priest draft I’ve ever had. The deck is so good grandma could have gone 12 wins with it. Its a bit of a pity that I played the deck then, since I only went 2-3. :( :( :(  But anyway, in the clip you will get to see how special minions (in this case its Questing Adventurer), with just a little help from supporting cards can just snowball the game.  If you are at work (or are a young child who has never heard ‘naughty words’ before) probably best to skip the clip and watch it later. During the game I was more Salty than a really salty thing.

So hopefully these examples has made it clear to you that ‘special minions’ are powerful. So powerful in fact, failure to play around them can result in devastating losses in board-control, tempo, life, card advantage, etc.  The next two sections look at how we play around such devastating cards.

Picking your Poison


Let’s refer back to simple example #3 from the above section. To save you from scrolling back up, I will reprint it here:

Simple Example #3:

We know that this is a terrible line for the second player, Given this, we can ask ourselves a very simple question; was it better for Player 2 to play a 2/3 minion instead? The (unfortunate) answer to that question is; “no, not necessarily”.  To see why, all one has to do is tweak the example a little bit:

Simple Example #4:

In this example all I did was change the ‘special minion’ (from the ‘Wolf’ to ‘Sergeant’) and poor old Player Two is still in a bad spot. The point I am making here is that it is rarely possible to play around every possible card your opponent may have. A lot of the time you just have to ‘pick your poison‘, so to speak.

So the lesson here is that special minions are powerful and we should try to play around them, but that is far easier said than done; being strong against one sort of attack often comes at the cost of being weak to another. So, what can we do? Well, here is two simple ideas:

(1) Play around cards based on (a) how devastating they are and (b) how likely it is that your opponent drafted/kept them during their mulligan.

Here are a few considerations:

(*) Warrior and Mage tend to draft Raging Worgen a bit more frequently than other classes.
(*) Flamestrike is more likely than Blizzard due to card rarity (‘common’ vs ‘rare’).
(*) Since abusive sergeant is a one-drop (and one drops suck), its more likely the opponent drafted dire wolf alpha than the former card.
(*) Since dire wolf alpha buffs the 1/1’s that pop out of the spider, this is more devastating than the abusive play.
 

Notice here that the last two bullet points suggest that not only is dire wolf more likely, but that it is also the more devastating play. Thus, when deciding whether to play the 3/2 or the 2/3 we now have a theoretical framework to base our decision on. The theory claims that — in the situations where we must choose– it is better to make ourselves strong against good–and probable– play instead of making ourselves strong against the weak(er)– and less likely– play. In this specific case, we ought to drop the 2/3 and hope for the best.

But if you were expecting this guide to be full of simply little rules you can unthinkingly follow in you games I am now going to disappoint you. The above analysis suggests that the 2/3 is better since it nicely plays around a dire wolf.  But here’s the catch; 3/2 minions counter most 3-drops, whereas 2/3’s don’t.  Allow me to illustrate:

Simple Example #5:

So in this time we are not looking at special minions and are instead asking ourselves what happens if we simply play a decent body on curve.  The 2/3 in this case can’t trade with the creeper nor with the 3/3 so the minion ends up stranded with nothing to do.  If Player Two had played a 3/2 instead then they would have a decent trade to make.  What we are learning then, is no matter what we play there is a counter.

But anyway, lets move unto practical tip number two:

(2) Concentrate on attack; while we cannot possibly know for certain what our opponent is holding onto, we do know our own hands and decks. So, if we have special minions in hand its probably a good idea to play in such a way that maximises the chances of getting good value from our own stuff (as opposed playing in fear of what our opponent may do).

So now we move onward to the next section, which is about trying to exploit the propensities of our opponents.

The ‘counter-counter effect’


Suppose for a second that you are playing a game in constructed as a Warlock. Its your turn and you play Cairne Bloodhoof. You then Life Tap and draw a Mortal Coil. Is it weird for me to say that having Cairne in play “buffs” this card?   Indeed, why would that be the case, there is after all, no obvious connection/synergy between these two cards.  In what sense then, is coil ‘buffed’ by the Cairne?

Before answering that question, let me start by explaining the power of cards is never static. Rather, the power of all cards are in constant flux; the value of any given card rises and falls on a annual basis (‘set releases‘), on a daily basis (‘the meta‘), on a ‘per game’ basis (i.e. ‘class match-ups‘), and also on a ‘turn-by-turn’ basis (e.g. the value of having a Hex in hand is low when there are poor targets, decent when there is a good target and excellent when there is a good target AND you can use your remaining mana to good effect).

I could also say that Ancient of Lore and Big Game Hunter ‘buff’ each other. How so?  Well, the ability to drop BGH and kill a big target nicely mitigates one of the weakness of playing lore (i.e. that Lore’s small body makes him an anti-tempo play). Basically, the two cards seem to ‘compliment‘ each other, and this is all I really mean when I say ‘buff’.

But lets go back to my Cairne example. The answer to my question cannot be found by simply studying the cards in a vacuum. Rather, to understand we have to broaden our minds and think about the big picture.  To put you out of your misery, the point is that when we we find themselves confronting Cairne plenty of us would quite like to silence it. And in the current meta one of the most popular silence cards is Ironbeak Owl. And what do we notice about the owl?  Well, its one health makes it a pretty good target for mortal coil. If the meta was to suddenly change and everyone replaced Owls with Spellbreaker‘s then it would no longer be true to say “Cairne ‘buffs’ coil”. But in this case, all what has happened is ‘the buff’ Cairne gives has simply relocated to other cards such as Darkbomb. In short; Coil counters the counter to Cairne Bloodhoof.

How do you guys feel about a pop quiz?

Question 1:  Suppose you are playing (in Constructed) as a Druid with The Black Knight in hand. The enemy Warlock is at 15 life and has 10 mana. On his turn, the Warlock decided to play Dread Infernal. My question to you is; ‘how does this demon (in this particular situation) ‘buff’ black knight?‘  Have a think about it.

Here is my famous (patent pending) dotted line technology that stops you from seeing the answer accidentally!

last chance to stop scrolling!

Answer: Okay, so in order to solve this puzzle you had to figure out that it has almost nothing to do with the minion itself and everything to do with the Warlocks life total. He was at 15 before playing the demon and at 14 after (14 being of course the magic number vs Druid). Thus, the Warlock brings himself in range of Force of Nature + Savage Roar combo. In general, there are only three defences against the combo; the first is Loatheb (which we can rule out since the Warlock only has 4 mana left), the second defence is heal, and the third defence is to Taunt up. The third of which plays directly into the Knight’s hands. Assuming the Warlock is smart, before he played the demon he ought to have known about  this possibility; ergo, he is likely to follow up with a taunt (or heal) with the remaining four mana (that or, the Warlock has decided to ignore the possibility of combo and ‘play for the win’ instead ).  

Basically, the point is that Black Knights value is in constant flux; its value is high when probability of a taunt being played is high, and the value is low when the probability our opponent playing a taunt is low. It is worth noting however, that ‘the buff’ only works because the Warlock expects the druid to have combo. If the Warlock doesn’t feel obliged to taunt up then black knight fails to get the buff.  In short; Black Knight counters the counter to Force + Roar combo.

While this latter sort of combo is less applicable in Arena (because people are less likely to play around such combo’s), I thought that  it was a bit too interesting to leave out of the guide. :)  And in any case, the concept does have some application in Arena; a simple example is that most players fear Flamestrike, and as such they may try to play around that card possibly by playing big (i.e. 5+ health) minions on turns 5/6/7 to pre-empt the strike.  Such a choice however, ‘buffs’ cards like Fireball and Polymorph (note also that this is a good example of ‘picking poisons‘ too; small minion = death by flamestrike, big minion = death by fireball).  In short; Fireball counters the counter to Flamestrike.

So I think the above gives you a solid understanding of what I mean when I say the counter-counter effect. Lets now get back to talking about the key focus of the guide (turns two & three in arena), and see if we can use this idea to improve our play.

Let’s recall a previous example:

Simple Example #1:

  • I play Loot Hoarder

    …The response is a 2/3

    …I play Raging Worgen

So in this example, the second player tries to exploit our Loot H play by dropping a 2/3. To which we counter by playing a special 3/3 minion (in practice though, any 3/3 is a good response to a 2/3).  To use the terminology of the above paragraphs; an enemy 2/3 on the board essentially ‘buffs’ all your minions capable of trading with it, and ‘nerfs’ all of those that cannot trade efficiently.

Lets now look at this example from the second players perspective. Imagine that their hand consists of three cards: (1) vanilla 3/2, (2) vanilla 2/3 and (3)  3-attack weapon.

When Player One played Loot H the obvious response was to play the 2/3 (since the 3/2 merely trades). And now Player Two (being the clever cookie that he is) realises that there is a risk that Player One may just respond with his own 3/3 minion, but he notes that he can counter that play via his weapon and so he is happy to go down this line of play. Basically, the 2/1 is countered by a 2/3, the 2/3 is countered by the 3/3, and the 3/3 is countered by the weapon. And thats the counter-counter effect in action!

So okay, lets change the example slightly, lets suppose that Player Two’s weapon has two (not three, as it is above) attack. What happens now?  Well, in this situation you still might opt for the above line of play, realising that you can trade your 2/3 and a weapon hit for the 3/3. But, this is not that efficient, and so instead you might prefer to use the weapon on turn two against the Loot H, and if they respond with by playing a 3/3 on curve you can drop the 3/2 to try and trade-up.

While some of you may find this reasoning a bit tricky, it is not rocket science. The ‘counter-counter effect’ is simply a style of thinking whereby you anticipate your opponents most likely response to your moves. And then armed with that prediction you try to think of ways to counter-that-counter. It will of course not always be possible, and sometimes your opponent will do something completely unexpected. But nonetheless, this concept can be pretty useful.

…But the Rope is Burning!


So at this point in the guide you may start to be questioning the practical use of what I am teaching. After all, I have spoken about so much for so long it may not be clear to you how it possible to think about all this stuff in a real game; surely if we were to ask every question I posed in the ‘thinking about turn two’ section above for every play we had wouldn’t the rope just burn out before we could calculate all of the possibilities?

The short answer to that question is yes; trying to think of everything is one sure-fire way to run out of time. The solution to this problem is, in essence, to learn to think less.

There is a wealth of scientific literature that suggests that gaining proficiency at tasks usually means you expend less energy, not more. In other words, over time your brain develops a rather sophisticated auto-pilot for dealing with the task(s) at hand. This ‘auto-pilot’ usually goes by another name; ‘intuition‘. The next few paragraphs are going to talk about Chess (specifically intuition in chess), but don’t worry, you don’t need to be a good Chess player in order to understand the comparison/analogy I am going to make.

Here is a chess article written about Magnus Carlsen (the current reigning World Chess Champion) and its all about intuition. The author looks at Carlsen’s ability to simply understand a complex chess position. When Asked why he picks a certain moves over another the World Champion does not respond with reams of analysis, rather he tends to trust his instincts. To quote a passage from the article (note: for those not familiar with the practices of academia, comments in square brackets [] in the passage below are my own. I’ve added them to help you better understand what you are reading):

“…Carlsen thought that the best chance to complicate matters for white in the following position would be to play h4 [In chess, number+letters are used to tell us what peices has moved to what square. ‘h5’ means that a pawn has moved the square ‘h5’] adding “I can go Nd6 h5 Ne4.”. Here [Luke McShane (Carlsen’s opponent)] asks him a bit surprised – “Not Nf5?” On which Carlsen replies: Ne4 looks better for me somehow, I don’t know […after some moves…] I don’t really see a follow up[…]”

[…] Indeed, computer [analysis] supports Carlsen’s move of Ne4 being much stronger than Nf5. […] How does one choose Ne4 over Nf5? I am pretty sure it is intuitive decision as there are no concrete positional flags, neither specific lines. I bet for an average GM [‘GM’ is short for ‘Grandmaster’, a rank which awarded to exceptionally strong players], Nf5 looks as good as Ne4.

In other words, we know from computer analysis that Carlsen correctly assessed a complex position. Many a Grandmaster, the article argues, would not have understood that Carlsens move (Ne4) is better than (Nf5). But the interesting point is not that Carlsen was correct, rather, what’s interesting is that when asked about the move his awnser is a bit vague and fuzzy; “…[it] looks better […] somehow…”  Indeed, the vagueness of his response seems to suggest that Carlsen himself  does not  know exactly why Ne4 is better. What he does know however, is that his gut likes the move and Carlsen see’s no reason not to trust such a feeling.  In an interview for a magazine, a journalist asked Carlsen his thoughts on him being called an “intuitive player”, his response:

“The fact that I can be considered an intuitive chess player, I think, partly comes from my early experiences (as a child), where I put all those hours with myself on the chess board and tried out things. It meant that I eventually got a feel for chess, an understanding of the game. General good players use more long-term memory than short-term memory during a chess game. You use past experiences. It is the intuition that is largely based on the past experiences. So it is your experience that gives you a different impression of the new situations before you (on the chess board) and then you have to consider what impression you can use. You must be able to continuously make up your mind about which past experience that can be used […].” 

Carlsen’s first language is not English, so we ought to forgive the clumsiness of his answer. Nonetheless, the salient point shines through; his intuition (which has so far been a great asset to his chess career) comes as a result of plenty of hard work over a number of years.

Okay so what does all this have to with Hearthstone?  Well, my contention is that thinking about your moves makes you a better player. But the problem is that answering all the questions you need to answer in the allotted in-game time is unrealistic. But, should we develop strong instincts then we can reduce the time it takes to process a turn without adversely affecting the quality of our decisions.

Let me quickly reprint the list of Turn two questions:

  • What happens if they use Hero Power?
  • What happens if they play a 3/2 minion?
  • What happens if they play a 2/3 minion?
  • What happens if they play a minion with some other stat distribution (e.g. a 1/3)
  • What happens if they play a special minion?  (I’ll define the term “special minion” in a moment)
  • What spells, secrets, or weapons could they play?
  • If there are already minions on the board, what attacks/trades can they make.
  • If they have The Coin, what could they do with it.

Do I ask myself every one of this questions on turn two?   The answer is a simple (and resounding) no. Usually I will ask myself maybe 2-3 of these questions. How do I know what are the right questions to pick? Intuition.

In other words, as you get better and better at the game you will develop an ability to quickly realise what the important questions are and what questions need not be asked. How do we develop such instincts? Well, Carlsen’s answer seems to suggest that developing intuition requires a hefty dose of experience.  If you want to get good at the game there is no short-cut or magic fix–its all about the man-hours, I’m afraid.  Guides such as this can only help to illuminate the path one must walk down; I cannot take the footsteps for you.

Speaking of experience, here’s my quest log (as of 17/01/15):

View post on imgur.com

I’m close to five-thousand constructed wins and one-and-a-half-thousand Arena wins. So, if you read my articles and wonder how I got to understand the game as well as I do these numbers probably have something to do with it!  (5k constructed wins and I’m not even playing at pro-level…yet!)

Anyway, lets get back to Arena-talk.

The Four Examples: Introductory Remarks


The next few sections are going to be detailed examples of how to think about a turn (in pop quiz format). But before we begin I want to make a few introductory comments:

  • Doing the Quiz:

If you would like to do the quiz/homework then remember that whenever you see three dotted-likes like these:

Try not to scroll up or down the page to quickly. These lines are simply there to stop you from accidentally seeing the answer.

  • Notation:

Believe it or not, one of hardest problems I encountered writing this guide is trying create a readable format for the reams of analysis that is to follow.  The best I could come up with is bracket notation {}.  The brackets represent board states, e.g. { 1/1 vs 1/1 } means both players have a 1/1 on the board. Our side of the board is always the left hand side of the ‘vs’ symbol.  So, if we both play 4/5 minions and I choose to double-attack his 4/5 I would represent that exchange as:

{ 1/1 + 4/5 vs 1/1 + 4/5 }  (before the trades)

{ 4/1 vs 1/1 } (after the trades)

Hopefully you will find this notation easy to follow. :)

  • The possibility of Errors:

While writing this guide I have worked hard to avoid any and all errors. However, the guide is about to get extremely complex and it is an unfortunate– but inevitable– fact the deeper you delve into the nuance and complexity of a any subject the more prone you are to making the odd error (or two). It should be obvious that fixing a plane is harder than fixing a bike!

So yes, expect errors. If you happen to see one let me know in the comments (and I’ll try and correct it). But until I publish the correction try to look past the mistake and understand the general principle I was trying to teach.

  • Once the clip ends, please pause the video

Okay. I think we are now ready…

In depth Example One


Okay so lets stare at a pretty picture (I’m sorry ladies, the picture it isn’t of my face on this occasion):

Okay, so its Turn Two and the Paladin has coined a Nerubian Egg. The options:

  1. Play Shrinkmeister
  2. Hero Power / Do nothing

Your homework; choose between options (1) &  (2) and explain why.

I’m going to start giving my answers  in a few lines from now. So this is your last chance to look away and do your homework!

Wait what!?  Your Dog did that?  The poor thing must have been so terribly hungry.

Okay, so lets start answering these questions:

The answer?  I did nothing!

Think about it logically; if your opponent is going to coin out a egg that ought to be a big tell than they have something in the hand that combo’s with the egg. In game, I realised that playing the 3/2 gives the Paladin a chance to sac the egg for the 4/4 nerubain, and that dude will effortlessly take out my turn three Dark Cultist.  Playing nothing is (rather unfortunately) my only way to avoiding the egg proc.  For various interested parties the following clip is of me getting completely crushed:

***UPDATE (20/1/14): After thinking about this position some more I’m becoming unsure as to whether or not my analysis is correct. Please see the discussion in the comments section for an explanation. ***

 

Okay, so that was just a simple one to help you get used to the format. The next three examples will analysed to a much higher degree of detail.

In depth Example Two


The image above is taken from an Arena game.  We can clearly see that our Druid opponent has decided to play turn one The Coin + Faerie Dragon.

And now it is our turn, we play Paladin. Our options seem to be:

  1. Play our hero power ( Reinforce ).
  2. Play Murloc Tidehunter
  3. Play Mad Scientist
  4. Do Absolutely nothing.

It is also important to note that the deck contains two secrets; Avengex1 and Noble Sacrificex1

Now I have a few questions I want you to think about:

  1. Out of these four options, what is the best play?
  2. Suppose for a second that deck contains no secrets. Does this change your answer to question (1)?
  3. On a scale of 1-10 (10 being significantly better, 0 being almost equal), how much stronger is the best play compared to the second best play?
  4. [For bonus credit:] Can you think a situation where option (4) is preferable to option (1)?

Before reading on, I really do want you to stop and think about it. Getting the correct answer to the these questions is reasonably straightforward, but in Hearthstone (and in life) the logic behind a decision is often more important than the actual decision itself  (getting the correct answer but for the wrong reasons wins you one game. Whereas getting the decision wrong for the right reasons means you lose this game but win countless others).  So, with this in mind, your homework assignment is to write down exactly why you think ‘play X’ is the best of the four options.

An Englishman, an Irishman and a Scotsman all go into a pub…

…And the Barman says “What’s this, a joke?”

Answers:

  1. Mad Scientist
  2. No, even if the deck had no secrets we should still play mad scientist
  3. The number you picked doesn’t matter, I simply wanted you to start thinking about how important it is to be right.
  4. If our opponent played turn-one Innervate + Raging Worgen then doing nothing (option 4) would be preferable to playing hero power (option 1) since this way the Worgen does not enrage.

Okay so why is the Scientist the better play?

To answer this question we need to do a bit of analysis. We need to look at each of our options and then consider what our opponents likely best response is.  Without further ado:

If we play either minion, our opponent is likely to have four basic options;

  • use a removal spell, such as Wrath
  • Play a minion (e.g. a 3/2)
  • Use his Hero Power.
  • Innervate + something expensive. (this possibility shall not be looked at)

We will start by looking at what happens if we play tidehunter, and then we will consider what happens if we play mad scientist. By comparing the two we can evaluate what the best play is.

Possible line of play #1:  Tidehunter —> They play a 3/2 minion, use F.Dragon to attack our 2/1.

  • The position starts out as: { 2/1 + 1/1 vs 3/2 }.
  • …And then the Dragon attacks (our 2/1), and they then drop a new 3/2.  { 1/1 vs 3/2 }
  • …On our turn we can play Mind Control Tech as a 3/3 and then go face with the 1/1 minion. { 3/3 + 1/1 vs. 3/2 }
  • …The Druid could then trade the 3/2 for the 3/3 and play a new threat, or maybe he could Swipe the board. In either case, our opponent does have the lead in initiative but the position is okay for us here.  If 3-drop: { 1/1 vs 3/3 }, If Swipe: { nothing vs 3/2 }

Possible line of Play #2: Tidehunter —> They play a 3/2, use F.Dragon to kill our 1/1.

  • The position starts out as: { 2/1 + 1/1 vs 3/2 }
  • …And then the dragon attacks the 1/1, and they then drop the new 3/2. { 2/1 vs 3/2 + 3/1 }
  • …On our turn we can play MTC as a 3/3. We also trade the 2/1 into the 3/2. { 3/3 vs 3/1 }
  • …And now (just as before) the Druid could Swipe or ‘Trade + Play threat’.  In either case, the position is okay for us, but notice that (unlike line #1) we don’t have the 1/1. This means that this line of play is slightly worse for us than the line #1 above.  If 3-drop + Trade: { nothing vs 3/3 },  If Swipe: { nothing vs 3/1 }

Okay, so what the above analysis suggests is that if our opponent is going to play a 3/2 then he/she ought to kill the 1/1 with the dragon since this leads to a better position for them. But now lets consider what happens if our opponent plays a 2/3 minion instead:

Possible line of Play  #3: Tidehunter —> They play a 2/3, use F.Dragon to kill our 2/1

  • The position starts out as: {2/1 + 1/1 vs 3/2 }
  • …And then the dragon attacks (our 2/1), and they then drop the 2/3. { 1/1 vs 2/3 }
  • …On our turn, we can play the MTC as a 3/3 and go face with the 1/1. { 3/3 + 1/1 vs 2/3 }.
  • …And now if the Druid has Swipe then the position is more or less the same as it was before. But if on the other hand they play a minion, then we are in good position since our 3/3 favourably trades with their 2/3.  If 3-drop + attack 1/1: { 3/3 vs 2/2 + 3/3 }, If Swipe: { nothing vs 2/3 }

*** Note: Remember in these examples who is to play. For example { 3/3 vs 2/2 + 3/3 } is a good position for us because it is us to make on move. If we trade into the 2/2 and play a Yeti the position suddenly looks like: { 3/1 + 4/5 vs 3/3 }  *** 

Possible line of Play #4: Tidehunter —> They play a 2/3, use F.Dragon to kill our 1/1

  • The position starts out as: { 2/1 + 1/1 vs 3/2 }
  • …And then the dragon attacks (our 1/1), and they then drop the 2/3. { 2/1 vs 2/3 + 3/1 }
  • …On our turn we can now kill the 3/1 dragon with the 2/1 murloc and drop MTC as a vanilla 3/3. { 3/3 vs 2/3 }.  And this position is still pretty good for us, even though we don’t have the 1/1 this time round.

So what have we learnt so far?  Well, it seems that regardless of whether the Druid plays a 3/2 or 2/3 he is always better off killing our 1/1 with the Dragon. Moreover, in most cases playing the 3/2 is more effective (thats because the 3/2 is able to trade-up with MTC). But we’re not done yet! we now need to look at what happens if they use Hero Power instead of playing a minion:

*** Note: I’m not going to analyse the effect of wrath, the reason being is that in terms of board-control wrath is equivalent as Hero Power in this particular case. ***

Possible line of Play #5: Tidehunter —> They play Hero Power (targeting the 1/1).

  • The position starts out as: { 2/1 + 1/1 vs 3/2 }
  • …They power up and kill the 1/1. Go face with the Dragon (note; trading dragon with the 1/1 transposes to the same position). { 2/1 vs 3/2 }
  • …On our turn we can trade in and develop our own threat (e.g. our 3/3 MTC).  { 3/3 vs Nothing }.  This position is fantastic for us.

Possible line of Play #6: Tidehunter —> They play Hero Power (targeting the 2/1).

  • The position starts out as: { 2/1 + 1/1 vs 3/2 }
  • …They power up and kill the 2/1 with face and kill the 1/1 with the dragon { Nothing vs 3/1 }
  • …And now if we Hero power they could Hero power in response, thus its not easy to trade our 1/1 recruit for their 3/1 dragon. Another option is to play the 3/3 but then they can trade the 3/1 and play a new threat . { Nothing vs 3/3 }.  Notice that in both cases we are in a bad position; The druid has maintained the initiative and has gained +1 card advantage by killing both parts of the murloc (without losing a card of his own).

So we can clearly see then, that killing the 2/1 with Hero Power is superior to killing the 1/1.  Its also worth noting that line #6 seems superior to lines #1-4 as well!  So now we are able to say something interesting about the position; EVEN IF the Druid has a two-drop, he/she is still going to play Hero Power!!

Our conclusion then, looks something like this:  “If we play Murloc Tidehunter, their best play is to kill the 2/1 with Hero power and the 1/1 with Faerie Dragon. This play leaves them ahead on board and with a small lead in card advantage.”  So now we have a benchmark; if Mad Scientist can do better job than the Murloc Tidehunter, then we ought to play that card instead. Lets see, shall we?

Possible line of Play #7: Mad Scientist —> Wrath. Go face with F.dragon.

  • The position starts out as: { 2/2 vs 3/2 }
  • ….now they Wrath Scientist, and hit face for three. { Nothing vs 3/2 }
  • ….At this point we need to consider the secrets. if the secret is Noble Sacrifice, then the secret can challenges the dragon. The Druid could trigger it with his Hero Power next turn but that forgoes the option of playing a minion on curve. If we get avenge, then maybe we can just play Tidehunter which in this case will probably allow us to trade the Dragon for our murloc(s). Both of these possible outcomes are okay for us.

Possible line of Play #8: Mad Scientist —> Play a 3/2. Go face with F.Dragon.

  • The position starts out as: { 2/2 vs 3/2 }
  • …They now drop a 3/2 and go Face { 2/2 vs 3/2 + 3/2 }
  • …And we now kill a 3/2, activating the secret.   The resulting position is basically the same as the position in line #7 (i.e. the secret squares up against a 3/2)

Possible line of Play #9: Mad Scientist—> Play 2/3. Go Face with F.Dragon.

  • The position starts out as: { 2/2 vs 3/2 }
  • …They now drop a 2/3 and go Face { 2/2 vs 3/2 + 2/3 }
  • …We now trade into the 3/2 with Scientist. And either play Tidehunter or MTC. In this case, Noble sacrifice is slightly weaker than it was before since the Druid’s minion survives. However, Avenge just got better since 1+2 = 3 health. This means that the Druid cannot use his 2/3 to kill off the minion buffed by Avenge.   In short, this position is okay for us.

Okay, so thats nine lines we have now looked at. But in reality, all we have to do is compare the best outcome (for our opponent) when we play Mad Scientist (line #7) versus the best outcome (for our opponent) vs Murloc Tidehunter (line #6). Which ever one of these lines is least painful for us is variation the game ought to follow (If you don’t understand this logic, you may try reading up on some game theory).  Ergo, line #7 is the likely game continuation.

I asked you earlier if Mad Scientist was still the best play even if the deck had no secrets. You might think that this would require a lot of complex analysis similar to what we have seen above. But thats not necessarily the case. You may remember much earlier in the guide (‘What to think about on Turn Two‘) I wrote that we can evaluate positions based on a variety of criteria; when analysing ‘Loot Hoarder vs Bloodfen Raptor‘ I evaluated by looking at the impact each card had on card advantage. In the ‘Mad Scientist vs Murloc Tidehunter’ analysis above I evaluated the position based on who had Board-control. So, if we evaluate in terms of card advantage we can quickly see that Tidehunter into enemy Hero Power (line #6) quickly drops a card. Whereas, even without secrets, if Mad Scientist gets wrath’d its a 1:1 trade and if Mad S fights F.Dragon the outcome is also a 1:1 trade. Clearly then, the Scientist does a better job than tidehunter (when the criteria for evaluation is card advantage).

Now I urge you to wrestle your homework off of the dog and compare your thoughts to mine.  Also, have a quick look at the number you gave about how important this decision was. We now have some qualitative data that allows us to understand how much better one option is over the other (i,e. the second best option is likely to lead to card disadvantage, whereas the best option gives at least card equality).  Hopefully this example teaches a lot of things, but the lesson I want to stress here is that picking the right minion to play on turn two (or any other turn, for that matter) can really make a big difference.

For those interested, the video clip below is of me playing this position. The audio commentary you can here  is ‘live’  (the audio quality isn’t that great but hopefully you can tolerate it for a minute or two) :) :

Now that you have read– and heard my live thoughts — we can probably revisit some of the idea’s contained in my intuition section (…But the rope is burning!). Did you hear me evaluate nine possible lines of play in a minute?   Of course you didn’t!   But what you did hear is me quickly latching unto the key idea of the position.

Oh, one last thing I’d like to point out is that notice that the Druid played Hero Power on Turn two and it did almost nothing, If I had played Tidehunter I would have inadvertently given my opponent a good play (I discussed this idea in the ‘What about Turn One?‘ section of the guide), I did try to highlight this when I was talking (about 1:31 mark) but unfortunately I did a terrible job!  Hence I’m explaining what I meant to say in this short paragraph. :)

On that note, I think I will end my discussion of this example.  Let’s  move on…

In depth Example Three


This time your homework is going to be a tiny bit different. Hopefully my amazing dotted line technology is preventing from seeing the next pretty picture. The last point I made in example two  was to suggest that I intuitively latched onto the key idea in a given position. Lets see if you can do the same!  Your homework task is to give yourself 10 seconds to try and figure out what the key idea may be.

Are you ready?

Okay then, here we go!

In this example we are a Warrior up against a Mage with nothing on Board. We have several options:

  1. Play Mad Bomber
  2. Play Coin + Scarlet Crusader
  3. Play Coin + Earthen Ring Farseer
  4. Play Coin + Injured Blademaster
  5. Play Hero Power ( Armor Up! )
  6. Do Absolutely nothing

Unless you are a super-fast reader, your 10 seconds is up. Before moving on, I’d suggest you write those thoughts down. Once you have done that we can move onto part II of your homework assignment; which is the deep think. I want you to spend five or so minutes thinking about  each of the six options and jot down a few strengths and weaknesses of each. Once again I want to remind you that getting the correct answer is not that difficult; the logic behind your answer is FAR MORE IMPORTANT than the actual answer itself.

Last chance, the Dog has been fed now, so I don’t want to hear that excuse again!

Wait What!?  A Seagull flew in through the kitchen window and snatched your homework, leaving only bird droppings in its wake?  Okay, hand it in tomorrow, oh else I’ll give you detention!

  • Play (1): Mad Bomber:

This play is somewhat reasonable. If you play him now you are likely to get almost no value from the battlecry (which is unfortunate since this card is one of the most frequently found cards in 12 win decks**). Basically he is merely a Bloodfen Raptor, but that’s okay since (a) Raptor is a decent Arena card anyway, (b) we are still playing sizeable threats on curve (c) we have plenty of options for next turn, (d) we keep the coin.

**{CLICK HERE} for the evidence of that statement.

  • Play (2): Coin + Crusader

Crusader is a great 3-drop in Arena. Unfortunately with that said the card is a lot worse when up against classes whose Hero Power can threaten it (e.g. Mage).  If we play this now then we give the Mage a simple option of either using Hero Power over two turns to kill it (the result here is a trade off between tempo and card advantage), or the Mage can ping it once and  then throw out a minion to trade with it. A third possibility is that the Mage plays a minion that threatens a 1:1 trade without the help of Hero Power (for example, Elven Archer). In all of these cases, the Mage player is likely to deal with the divine shield easily and thus it is not likely that this card will do any better than get a 1:1 trade (technically we won’t even get a 1:1 trade since we spent the coin to play it!). Another (albeit minor) point is that playing this minion limits our ability to effectively deploy mad bomber next turn.

  • Play (3): Coin + Farseer

If we coin out this guy, the best we get is a 3/3 in play and a heal for– literally– one point.  If we wait, chances are we could get better value than this later on in the game.  This play is only better than Crusader in a few marginal cases (e.g. versus Elven Archer).

  • Play (4): Coin + Blademaster

With this play, we put a big minion on the field and we also give ourselves a pretty decent follow threat of making a 4/6 with the help of Farseer.  The downside is that this play is weaker against removal cards like Frostbolt, than Mad Bomber is (e.g we lose a 4/3 when we could have lost a 3/2).

  • Play (5): Hero Power

With this play we do nothing to control the board. Considering the wealth of other options, this doesn’t look like a strong play.

  • Play (6): Just pass

There is no sense in this. Play (5) is better in almost every respect.

To my mind, the blademaster is the best play. The critical point to realise is that making a 4/6 with Farseer next turn not only allows us to play the a new threat on curve the additional three healing means that we are likely to be able to punish a perfectly reasonable looking play from our opponent:

Possible line of Play #1: Blademaster —> They play a 3/3. Pass:

  • The position starts out as: { 4/3 vs nothing }
  • …They respond by playing a 3/3, and then pass. { 4/3 vs 3/3 }
  • …We now play Farseer, healing Blademaster to 4/6. We then attack the 3/3.  { 4/3 + 3/3 vs nothing }.  This is a great position. We have a strong board and +1 card advantage.
  • At 4 mana one of the best plays a Mage could have is Water Elemental. In response we could trade both minions (4/3 & 3/3) and play a new threat.  In this case we lose the +1 card advantage we got from killing the 3/3, but once we play a new threat we regain board control and have maintained the initiative.

What this analysis demonstrates then, is that if the Mage makes a perfectly reasonable play on turn three (i.e. play a 3/3) we can punish such a play so hard that we are likely to still be ahead even if Mage has an excellent turn four!

Of course, I also mentioned this play being weak to frostbolt (or similar cards). Lets quickly look at that now:

Possible line of Play #2: Blademaster —> They Frostbolt, killing Blademaster:

  • The position starts out as: { 4/3 vs nothing }
  • …They respond with Frostbolt on the 4/3.  { nothing vs nothing }
  • …We now drop a new threat.   In this case, we remain equal in terms of card advantage, but are slightly ahead in terms initiative and board-control. In short; this is a perfectly good spot for us to be in.

The weakness of playing Blademaster can been seen in Line #2. While the position is good the fact of the matter is that it is not the optimal solution. To understand why, let us imagine that we can see into a crystal ball and know the future; We know for a fact that the Mage is going to frostbolt any minion we play.  Given these assumptions, it is clearly the case that mad bomber would have been better since we would have lost a 3/2 instead of losing a 4/3.

So now we need to weigh up the scales. In my mind, the Blademaster play has the potential for a significant upside (strong board + card advantage), at the cost of minimal risk (losing a 4/3 to removal instead of losing a 3/2 to removal).  What what my thought process during the game?  Well, why not see for yourself:

Okay, lets move on to another example…

In depth Example Four


Okay, so just as before, lets list the options:

  1. Do Absolutely nothing
  2. Use Hero Power ( Lesser Heal )
  3. Play Bluegill Warrior
  4. Holy Smite face. (Dennis FTW!!)

And now — yes you’ve guess it — your homework is to write down what the best play is AND WHY.  Remember once again I will remind you that the ‘and why’ part is more important that the answer itself.

Ah, my lovely dotted lines! I suggest you cherish them, since this is the last time you are going to seeing them,

You better be waving the dots bye-bye. You Don’t want to make the dots angry.

This time I’m going to start the analysis with a Youtube clip. For this guide you only need to watch the clip until the Paladin plays the Shattered Sun Cleric, which is about one and half minutes from the start point..

So the first thing you might notice is thats not my voice nor is it my ridiculously cute face in the corner. :)  The credit for this clip goes to a Player named ‘Ratsmah‘ (who, if you haven’t heard of him before,  is an ‘infinite arenaTwitch streamer). He had a lot to say about this turn, but I think he struggled to express himself a bit; explaining a complicated thought process in real time is not easy. Just like the other three examples, I’m going to try and explain why skipping the turn is the right idea as best I can.  Let’s begin!

Easy stuff first; Since we are playing Priest, options (1) and (2) are in this case equivalent. Priest is the only class in the game where the Hero power sometimes does nothing and this is one of those situations. If we were any other class then doing something (i.e. Hero Power) would be preferable to doing nothing. This means that there are only actually two lines of play we need consider; play bluegil or play nothing.

Possible line of Play #1: Bluegil —> Hero Power —> Power word: Shield.

  • Starting position:  { nothing vs nothing }
  • …We Play Bluegil Warrior and on the Paladins turn he/she uses Hero Power. { 2/1 vs 1/1 }
  • …We can buff bluegil with Power Word: Shield. At this point we could get lucky and draw a 2-drop and play that as well. If however, we don’t get lucky then the best we can do is heal bluegil back up to a 2/3.   { 2/3 vs nothing }
  • ….But now lets imagine the Paladin plays a 3/3, this threatens the Bluegil. Our best play is probably just Dark Cultist.  { 2/3 + 3/4 vs 3/3 }
  • …And now it the Paladin’s turn four, if he has a good card like Truesilver Champion then he can clear the board.       { nothing vs 3/1 + 4/1(w) }
  • If we were to change the cards the Paladin had then our evaluations may change (given how many wins Ratsmah has in this clip it is reasonable to expect the Paladin to have drafted good cards). But the salient point I want you to focus on is the idea that once we play Bluegil we constantly have to defend it. To let die is to give the opponent card advantage but it keeping it alive requires us commit resources (e.g. PW: Shield) and play minions off curve.

It is quite ironic that the aggressive play in this position forces us down a defensive path…Or does it? What if we just dropped the Cultist on turn three (instead of protecting Bluegil with PW:Shield) ?

Possible line of Play #2: Bluegil —> Hero Power —> Dark Cultist.

  • Starting position:  { nothing vs nothing }
  • …We Play Bluegil Warrior and on the Paladins turn he/she uses Hero Power. { 2/1 vs 1/1 }
  • …And now we go face with Bluegil and drop the Cultist { 2/1 + 3/4 vs 1/1 }
  • …The Paladin now trade the 1/1 for bluegil, and then makes a power play; something like coin + Chillwind Yeti  { 3/4 vs 4/5 }
  • …And now for arguments sake lets imagine our best play is drop a Loot Hoarder (a card I’m imagining we just topdecked**) and to go smite + cultist into Yeti.  { 2/4 vs nothing } Such a play clears the board, but at the cost of giving the Paladin +2 card advantage.
  • We are however, starting to get sidetracked. The only thing pertinent to the current discussion is how much value we are getting out of Bluegill. And in this line, all Bluegil manages to do is kill off a 1/1.

**The reason why I imagined a Loot Horder top-deck is because this card has the same stats as Bluegil. Which helps us make a fair comparison with the alternate line below **

As we can see then, playing Bluegil doesn’t seem to do well against Hero Power. The important question however is the following; does doing nothing lead to a better position?

Possible line of Play #3: Do nothing —> Hero Power —> Dark Cultist.

  • Starting position:  { nothing vs nothing }
  • …We Play nothing and on the Paladins turn he/she uses Hero Power. { nothing vs 1/1 }
  • …And now we drop the Cultist. Lets imagine that the Paladin makes a power play; something like coin + Chillwind Yeti.  { 3/4 vs 1/1 + 4/5 }
  • …And now our turn four can be Bluegil + Holy Smite (targeting Yeti), and then Cultist into Yeti and Bluegil into the 1/1.  { 2/3 vs nothing } Notice here that the position is very similar to line #1 but this time round we have cleared the board at the cost of +1 (not +2) card advantage to the Paladin.

I could go on and on. But I feel that this guide is long enough as it is. If you want to know happens if the Paladin does something else on turn two, by now you should have the skills to evaluate it yourself!

Conclusion


Congratulations, you made it!   Invisible internet beers on me. I really hope that you guys find this guide useful, don’t forget to comment if you need something explaining.

The note I would like to leave you on was something I said much earlier on; this guide ought to of illuminated the path you must walk down. But remember,  the walking bit, thats totally up to you. :)

Enjoyed this article?



Hi. I like to hit various things with blunt implements. 3x Legend, Infinite Arena. You can find on Youtube as well. :)

Learn and Improve Your Game
Join Premium and Become Legend!

Over 400,000 people each month use Hearthstone Players to improve their Hearthstone skills.

FROM JUST $2.95 / MONTH

34 Comments

Leave a Reply

  1. Anonymous says:

    ratsman is a fag. Why are you advertising him?

    • Stonekeep HSP says:

      Probably because he’s a great Arena player and this is an article about Arena, so Ratsmah’s play can be a good example?

  2. Anonymous says:

    *quote*
    …The Paladin now trade the 1/1 for bluegil, and then makes a power play; something like coin + Chillwind Yeti { 3/4 vs 4/5 }
    …And now for arguments sake lets imagine our best play is drop a Loot Hoarder (a card I’m imagining we just topdecked**) and to go smite + cultist into Yeti. { 2/4 vs nothing } Such a play clears the board, but at the cost of giving the Paladin +2 card advantage…
    *end quote*

    The Paladin only has +1 card advantage; (s)he played the coin in this particular situation.

  3. Anonymous says:

    I’ve been playing for less than 2 weeks and have 250+ combined wins (albeit I have a strong understanding of MtG). After reading this article and a couple of others like it, I managed to pull off a 7 win arena last night! Up from my previous best of 6. To the good sir who wrote this article, hat’s off, it’s really upped my game. Thank you.

    • Smashthings says:

      I’m pleased to here that.

      Helping people to improve is the entire idea behind this website :)

  4. patrissimo says:

    BTW, I think Example 3 is really excellent – you have 4 different drops to play, there are arguments for each and benefits to each, to a 3-win player they probably seem pretty similar or the wrong one seems best (ie scarlet), yet in the full analysis it is clear that IBM is modestly but definitely the best choice.

    I’m a 5-win player so I can’t speak to infinite play, but to me this is the kind of decision that gets you from 4 to 5 wins – it is small but significant and consistently finding this degree of edge will definitely impact your winrate.

  5. patrissimo says:

    This is great. I know you said it was getting too detailed at the end, but I would love more of these examples, with in-depth analysis like you did here. For each example, list our possible plays, then for each of those, list opponent’s likely responses, and what our best turn 3 would likely be. Then summarize: for this play, we end up with position X if opp does Y, position Z if opp does Q, etc – for only the most likely plays. Maybe note if a special card wrecks us (“But we get wrecked if opp has a shattered sun”).

    Then a final entry comparing the options. It would be great to build up a library of these over time, analogous to tsumego in go.

  6. Sam says:

    Great read! In-depth and useful analysis.

  7. Jhuderis says:

    What a great read! It really helped me to see where I’m going wrong in a lot of my choices. I’ll be putting the thought process into practice asap.

  8. Anonymous says:

    Great guide

  9. Rekt says:

    Wow, really helpful.. Been searching for guides like this for awhile. Thanks

    • Smashthings says:

      Glad you found it useful.

      Also, while I have you here… are there other sorts of Arena Guides you would like to read? (I ask because as a writer I’m always looking to new topics :) )

      • Photoshop says:

        I’d love to read a guide just like this one, with the concepts of drafting in mind… found this article really really useful, thank you very much for your time, Smashthings ! +1 invisible internet beet to you ! :)

        • Smashthings says:

          I’ve wondered about writing a detailed guide on drafting before. But in the end I decided not to on account of the fact that there is sites like Hearth Arena; The Algorithms they use are pretty good (pros typically differ on maybe 5 picks a draft).

          So my thought process has been something along the lines of “..,If beginners can use such tools and make decent decks repeatedly, what can I offer them with a guide?”.

          Anyway thank-you for the invisible internet beer. After a few pints I shall be blind internet drunk. :)

  10. MrShine says:

    Great article! You spent a lot of time analyzing different situations, and as you mention it’s not always possible to play around everything but it’s often worth trying to squeeze just a little more value out of cards on the critical T2 & T3 turns.

    One thing I noticed is I differed from the Arena Coop on their minibot vs recombobulator decision. My thoughts:
    shielded minibot is almost guaranteed to survive until next turn. Mage ping is a win for us because we recombobulate it and the 2/2 will almost always become a little better. If they play a 2/3 we can run into the 2/3 with the minibot and recombobulate it, which again will almost always be better than a 2/2. If they play a 3/2 we win with a successful trade, and once again recombobulator will oftentimes help our remaining vanilla 2/2.

    Playing the recombobulator pretty much means it’s going to trade 1 for 1 with almost anything that’s played (+ is weaker against mad bomber) and then we’re left with the minibot to deal with whatever he plays on T3. T3 minions more often have 3 hp, so it’s less likely our minibot is going to clear whatever the enemy plays. We also miss out on the recombobulator’s battlecry if we play it on an empty board.

    I know they ended up going with the recombobulator, but I actually think it was the wrong play. Funny how there can be so much behind seemingly simple turns.

    • Smashthings says:

      Yes, it is amazing how such seemingly simple turns can be studied in such tremendous detail. In Chess, sometimes you can feel bewildered and awestruck at how complex things can get. There is a certain beauty in that, I find. It would be nice to think that my article has inspired that sort of wonder in HS players.

      Anyway, Allow me to offer up a counter point to your analysis:

      Possible line of play #2 (you wrote of the first possibility)
      Start = {nothing vs nothing}
      The line is: 2/2 —> 2/3 —-> hit it. This leaves: { 2/2 (no shield) vs 2/1 }

      We then drop a recombob. Lets for arguments sake suppose that it transforms the 2/2 into a 2/3.
      {3/2 + 2/3 vs 2/1}

      The Mage attacks the 3/2, and lets suppose he plays a 3/3 on turn three.
      { 2/3 vs 3/3 }

      And on our turn three, lets assume that we don’t have a play better than Hero power (given the current state of the hand, we would have to top-deck a 2 or 3-drop to have a better play than this):
      {2/3 + 1/1 vs 3/3}

      Its now turn four for them. Lets say they also use Hero power and drop a 3/2. They then make the trades:
      {nothing vs 3/1 + 3/2}

      What this line demonstrates then, is that by playing minibot we can quickly fall behind if:
      (A) The transformation does not give us something that can deal with a 3/3 on turn three.
      (B) If the Mage can efficiently* keep the 3/1 alive on turn four.

      *(“Efficiently” basically means; Kill minion plus play own minion)

      Long story short; Its complex and not entirely clear what the ‘best play’ is. Both plays have their merits but also weaknesses. With this said, I think you might be correct Mr. Shine; to “prove” that turn 2 minibot was bad we had to make a number of assumptions, and if any one of these steps go wrong the Mage doesn’t get ahead. IMO, minibot first has a better upside, and is only weak a small percentage of the time.

      For these reasons, this should have been the play.

      • My thoughts were the same as Mr. Shine. One fairly big upside to playing that playing minibot offers that Recombobulator doesn’t is that the minibot will most likely prevent the play of most minions that have 2 toughness. Unless your opponent has some really great master plan, it would be fairly wasteful to play something that dies to the minibot only to effectively remove its shield. They could get exactly the same effect by simply using their hero power. But if you play Recombobulator, they’re now more likely to play anything that has 2+ power for the potential trade. To put it simply: minibot offers a much higher chance for a timewalk.

      • adwcta says:

        The problem obviously is that you can’t assume you’d get a 2/3.
        Here’s the % for Recombobulator. http://redd.it/2pjvdg

        The average card is a 1.8/2, and includes a slew of useless cards. Much like how unless you are priest or have good backup plans, it is best to play Deathlord later than turn 3, it is better to play Recombobulator for its ability later than turn 2. RNG can absolutely wreck you, and the earlier in the game this happens, the bigger the impact relative to your available mana to deal with it.

        As a good player, with a good deck, would you prefer to trust your play, or to let the dice determine the game?

        Our Recombobulator play will result in the Shielded Minibot having initiative when he drops his 3 drop (assume he does not use his 3 mana on turn 3 to ping, which would be awesome for us). This means that besides Harvest Golem, it will kill before the end of our turn 4 anything he plays on turn 3. There is very little that can go wrong with this play, and the upside remains if he misses any of his drops.

        On the other hand, a Shielded Minibot play will prevent a 3/2 from being dropped, and you end up with a RNG recombobulated minion + Recombobulator to opponent’s 3-drop, which has a high chance of trading into both of your minions anyway (the top 4 3-mana plays all do this). But, if it is a 2/3 being dropped, the Shielded MInibot play will go into your opponent’s turn 3 with a recombobulated minion (a 1.8/2 minion with no shield, dealing on average 1.8 damage on the board; as opposed to shielded minibot, dealing 4 damage on the board). This is how boards are lost and games are lost. There is a slightly higher chance to see a 3/2 than a 2/3, (about 60/40 in this meta). That’s not worth losing the board and likely the game for a small edge which will be wiped away by the 4 most commonly seen 3-drops.

        You cannot assume that 1) If he pings the minibot on turn 2, you’re in the clear. You’re not. There are more turns after that where he’ll have a high chance to overtake your board. 2) The Recombobulator will get you a desirable minion. It’ll get you a good minion more than half the time, but I cannot stress how bad the other times are when played this early in the match. Playing Recombobulator second is giving your opponent the chance to win without doing anything. It’s a desperation play when you know your deck is bad, or your skills are outmatched. Neither were the case here. 3) A Shielded Minibot will not clear your opponent’s 3-drop. Unlike the re-combobulated minion, a Shielded Minibot WILL clear your opponent’s 3-drop in all cases except for 1 neutral common card.

        Avoiding the Recombobulator’s ability here was 100% the correct play and we’ll stand by it.

        Best,
        ADWCTA

  11. glasnost says:

    Although it could probably be shorter, I should mention again that I think this is a good article to demonstrate how complex HS actually is, and well done.

  12. glasnost says:

    Example four is really good; I’m still thinking about it.

    Example 3, I’m not sure about. Scarlet is also a very strong play, and you will have the farseer combo later. If he pings SC Turn 3, that’s an awful turn where he doesn’t develop the board. You follow with BM.. you lose a card, but keep the initiative, even if he goes ping + bolt on T4. He will be able to ping Scarlet whenever you play it, but the tempo cost for him to do it is never higher than now.

    By contrast, Injured BM countered by bolt leaves you playing SC T4, to which yeti is a good response ( at least,if not for your arathi weaponsmith). It’s more competitve position for him than to be in than in the other scenario, where you have an empty board into which you play your 5 drop.

    In general, if you are using your hero power early at all, you are likely to lose the game, and card advantage matters a lot less.

    • Smashthings says:

      In both of your comments I think you make reasonable arguments. And I think that the fact there is a debate to be had about my examples speaks volumes about the complexity of hearthstone.

      Now that I’ve had some sleep, I’ll try and address your points analytically.

      In my first example one of the questions I had was “On a scale of 1-10 (10 being significantly better, 0 being almost equal), how much stronger is the best play compared to the second best play?” I think this idea is relevant to the ScarlettC (SC) vs BladeM (BM) debate of example 3. Personally, I think BM is better by a small (but significant) margin.

      Also relevant to the current discussion is the “picking poisons” section; it is undoubtedly true that there are some situations where SC is better than BM. For starters, I would agree with you, forcing hero power on turn 3 is the best time to do so. Moreover, SC is clearly better vs frostbolt.

      And yes, normally a turn 4 SC is not great since it could go up against a 4/5, But, as you are aware, in this particular case the weaponsmith takes care of this eventuality. Even though I never made this point in the original analysis I would argue that this adds more weight to the BM play: (see: counter-counter effect).

      Anyway, to reiterate my argument; BM on turn 3 has a huge potential upside that merely requires our opponent to make a reasonable move (i.e. play a 3/3). If we play him on turn 4 instead and we go up against a 3/6 or a 4/5 then we don’t get the same value out of the Earthern Ring heal.

      ~~~~

      On a less related point, I would take issue with your claim that early use of hero power likely losses the game. I just don’t think this is true. In example 1 the Analysis suggests hero power to be preferable to playing a 3/2.

      Furthermore, Mage in particular is a class that can often afford to play behind on tempo and initiative; typically they can reclaim the board with a turn 7 flamestrike. In short; a mage can often “get away” with going for the greedy 2-turn ping against SC in arena.

      • glasnost says:

        Re SC vs IBM, I agree it’s a tough call, maybe the least clear winner from your examples. But if you can hold the combo until T6, there’s no longer even a possibility of losing the combo value to opponent reaction. You get 7/10 of stats for 6 mana, guaranteed.

        Your argument is reasonable. I’m not sure how I feel. I think IBM is a fairly high priority removal target vs. Warrior.

        Re hero power, I’m not sure that I agree with your take. It depends to a significant extent on your deck and the oppo’s deck, I agree. There are some decks where hero power a 2/1 on T2 is probably fine.

        But in general, I am of the opinion that when you use hero power on a 2/1 on your T2, or on your T1 with coin, you have gained +1 card advantage, but your opponent has gained the ability to be the first player to put a 2 drop on the board, i.e he has the initiative. If he had a 2/drop in his hand, he handed you that initiative.

        I think card advantage matters a lot less often than initiative in arena these days. I have lost entire games – several – where I drew nothing but minions early, oppo ignored all trades, got on the board first, curved out, went face all day / every day, and just won, all thanks to his 2/1 or coined 3/2 T1, allowing him to have a spare attack to my face every turn while I was forced to make trades every turn.

        OTOH, unless the 1-drop player is holding a buff, a 2/3 is a good counter, and much better response than using a hero power, because your board is not empty.

        If you watch ratsmah, you should know that 1/drops have a power way beyond their bad stats and inherent card disadvantage, and the reason is that oppo can’t generate enough attacks to kill them fast enough.

        I’m losing nearly a third of my games in arena to perfect aggro curves with significant 1-drop usage, including ‘bad’ 1-drops. The reason for this is related to the advantage gained by Player 1 when Player 2 pings a 2/1 T2 instead of developing board (and this is even worse between Turns 3 and 5)

        • Making a less optimal play now (Crusader) at a chance to squeeze out a more optimal play a lot later (instantly larger IBM) will more often than not result in a worse results over time. I see a lot of people get caught in this trap where they play for a bigger effect latter than a tempo effect for the moment.

        • Smashthings says:

          Remember that if you do go for IMB heal on turn 6 you guarantee value out of the heal but you don’t guarantee value out of the minions.

          Eg. if they play a bolderfist ogre on turn 6 if you trade both in not only are you at card disadvantage the heal also failed to achieve anything (since post-heal IMB still dies).

          As for the value of Hero power; this is a huge topic to try and cover. However, I think the subject is somewhat moot; people are only going to play HP on turn two when the alternative play sucks.

          So for example. If you are Mage and the opponent plays a 2/1 . If the choice is between a 3/2 (not a 2/3!) and HP what do you pick? I’d be shocked if you said play the 3/2.

        • patrissimo says:

          The idea of holding the IBM/ERF combo until turn 6 just seems like really dangerous planning/thinking to me. Arena is not about turn 6 two-card combos that put out a 4 attack and 3 attack minion, it is about board control, as the rest of your story demonstrates powerfully.

          It is 100x more important what is the best play on Turn 2 than what a possible combo is on Turn 6, by which time you will have drawn 4+ more cards and the board state will be completely different. Especially with this “combo”, which is way short of something like equality/consecration, or unleash/cult master. Any combo less than that is rarely worth saving for 4 turns in the future.

          Not trying to be critical here – I am far from an expert at this game – I just think that an analysis process for T2 that includes “maybe I should save my two 3 drops for a so-so T6 combo” is probably not taking the right approach to the game.

          It’s not quite as bad here because you could skip those two cards and play on curve for a couple turns by playing T2 bomber, T3 coin arathi. But it greatly limits your options to try to hold 2 of your 3 3-drops when it sure looks like 3/3/4/3+2 should be your next 4 plays.

  13. glasnost says:

    A commendable effort on this writeup, good work.

    However, I have to say that I didn’t agree much with your very first ‘simple’ example, which kind of turns me off for the following examples.

    Shrinkmeister was definitely the right call on T2 vs the coined egg given the info available, regardless of what he has. To see that, simply compare the Pass vs. Shrinkmeister outcomes.

    When you passed, you let him further develop the board, and he got the combo later anyway, which was inevitably going to happen. And you can’t wait to avoid him getting value out of it, because of the huge lead he will obtain.

    Specifically, if he had gone Might, egg into the Shrinkmeister, you could have followed with Harvest golem, which trades evenly with the Neurbian. Well, except for the elven archer, which still would have worked. But b/c of your line of play, he had another way to keep the Nerubian alive – the minibot played last turn.

    And if the Nerubian eats the front of Harvest golem and he follows with Archer / Minibot on 3, you can play spider tank, and it doesn’t die to bot + Archer. At least he then has to sack the Nerubian in to clear the tank. You’re down a 1/1 and a minibot, which is bad, but less bad. Or you play soulpriest and he has to sack the Nerubian + the 1/1, even better.

    Point is, avoiding the combo trigger just lets him use it on Turn 3 to equally devastating effect, and with a larger board lead. Sometimes playing to deny enemy value is wrong and the best you can do is feed resources into the maw as fast as possible.

    • Smashthings says:

      Given what we know about the Paladins hand (and mine) I think it is safe to say that there wasn’t an easy way out.

      But anyway, If I am to be honest the line I was considering shrink—> might —> Cultist. I didn’t even consider the possibility of Harvest Golem trading with the 4/4 over two turns.

      And for what its worth I do think you make a reasonable point; the decision is not as ‘simple’ as I had made out it out to be.

      Its 4am here, so bedtime for me. I’ll mull it all over later :)

      • Smashthings says:

        Okay, so here’s my current feeling on matter; I think I can make a strong case for why SC is better than BM (see my other comment response to you) but I have my doubts as to whether I can defend playing nothing in the shrink example.

        line #1: Egg —> Shrink —> blessing —-> H.Golem.

        So this line of play is clearly better than the game continuation. I admitted earlier (comment above) that I failed to consider the ‘over two turns’ Golem trade, I was thinking Dark Cultist. Which is clearly inferior to your suggestion.

        However, just because the Paladin can pop the egg on turn 2 does not mean he will do so:

        line #2 Egg —> Shrink —-> minibot.

        And now we are once again in a tough spot; hitting the egg ourselves seems bad, and to go face is also bad (since blessing + Hero power) is mana efficient and pops the egg.

        So it looks like we want to kill off our 3/2, which in this particular case means we only get a divine shield for our 3/2. This is not good (((but if the Paladin played a 2/3 or 3/2 instead then this play would have been reasonable))).

        In response we have either Dark Cultist (which survives the egg + blessing) or H golem. But in either case, the position is likely to be a { nothing vs 4/4 }. Which, while bad, is admittedly better than the game continuation.

        In short; I’m going to have to think on this some more but I am starting to think my analysis was wrong in this example. Maybe it is better to accept the bad postion and play into the threats rather hold onto cards hoping for better options later.

  14. Tiandes says:

    Wow, that’s a lot of stuff to process for a newbie like me.

    The thing is, so far, I’ve been little to no tim in those situation. Usually in arena I rarely got multiple 2 drops or 3 drops on a turn 2 or turn 3 recpetiveley. It’s mor ea matter of playing on curve.

    How does it affect your decision?

    Concerning the Scarlett. I admit, in your example, playing her is not the optiml play. But usually, on turn 3 (or turn 2 with coin) it’s a given for me. Either I get great board advantgae, or I pretty much dictate my opponent next turn if he’s a Mage, Rogue or Druid, where he’ll clearly go hero power to at least get rid of the shield and attack it with a minion.

    • Smashthings says:

      As I stated at the beginning of the article, this guide wasn’t aimed at ‘newbies’. All the same, I do hope you got something from reading it.

      Regarding your Question:

      If you only have a single 2-drop to play on turn 2 then usually the only decision you need to make is whether playing it is better than hero power (It is worth noting in two of the examples playing nothing was better than dropping the minion on curve!). Playing the minion is usually better, and so therefore little thought was required. BUT, if you pay close attention you will probably be surprised how often (assuming you are drafting enough early game) having a choice between 2-drops occurs.

      The simplest bit of advice I can give to you is this: If you only have one 2-drop in hand, play it. If you have a choice of 2-drops, think carefully about what one is best.

      And also, Scarlett is usually a good turn 3 play (even against mage). If you played her in Example 3 instead of my suggestion (which was Blademaster) then you haven’t committed a terrible misplay, but you did miss a tactical shot.

      The point of my guide is trying to teach ‘optimal play’; players who consistently get 5 wins in Arena are players that are really good at finding the ‘good moves’ in a position (such as playing Scarlett). Players who average 7 wins don’t play good moves, rather, they play the most optimal moves. And thats what makes them such strong players.

      • cheesypuff says:

        First of all thanks for that very educational article. I’m one of those players who average 5 wins. I really learned a lot from this article; it made me think tactical in other ways than i usually do. Paradoxically did i manage to make the right plays in all of the example without having the exactly same reasoning as you. Especially the case about the SC and IBM dilemma am i still wondering why arcane missilies + 2drop/HP wasnt mentioned. That was actually my main reason for favouring IBM over SC!?

        • cheesypuff says:

          Oh… I actually think that my draft is a bigger barrier for reaching infinite, so if you eventually are going to write another article soon an article on the drafting topic would be great.

  15. Daren says:

    Hello there, on example number 3, I wanted to point out, that if scarlet crusader seems stronger play (frostbolt being way too common) he can answer with and then ping again + frostbolt on turn 4 wasting your efforts. So the potential win of blademaster trumps scarlet stronger tempo generation. But only thanks to the farseer. Mad bomber being a frostbolt bait could be a very strong play if we didn’t have the farseer in hand. (and we get to save coin for later)

    • Not only is Frost Bolt common, but there is also Flame Cannon which is a very good draft pick. That being said, I would still play IBM as it specifically requires cards from hand to deal with and will also use up the turn giving you a good window for Crusader on turn 3 against which the opponent either needs to play a sizable minion to avoid your shield or waste almost all of his mana for the turn removing the shield and then half the mana on turn four to ping it again. Basically, force the opponent to use actual cards or face an overwhelming opposition, something that playing Crusader on turn 3 instead of IBM doesn’t really force.