The Buff Patch Retrospective

hearthstone 6 - The Buff Patch Retrospective

Hey all, J_Alexander_HS back again today to look back at the Rise of the Mechs buff patch, its impact, and think about some lessons learned from the overall experience if something like this is to be attempted again. While I will include the most current numbers we have (according to HSReplay), this analysis will largely be subjective in nature when it comes to making sense of those numbers and thinking about their impact.

Before looking at the specifics, I wanted to say a few words about why buffs have been uncommon and why they might not be that good of an idea.

We can think about cards occupying four general “bands”: Weak, Mediocre, Strong, and Broken. Weak cards are just bad and don’t ever see play. For all practical purposes they don’t even exist except when you find one at random. No one is going to put them in decks and they don’t really impact the meta. Mediocre cards are those that might make their way into a deck as the 29th or 30th card. They’re not powerful independently or together with others, but maybe just good enough to sometimes see play. Accordingly, they generate very little meta impact, but can be neat to see when they pop up. Strong cards are those which are solidly playable. You see these a lot and they perform well. Broken cards are those which go beyond powerful into the realm of meta-defining. They tell you what other cards or decks you are allowed to play if you want to win, as you need to be able to keep up with them.

Cards within those bands have different emotional impacts as well, which I think can be (mostly) summed up by saying that Weak cards are boring (because they don’t affect your play experience most of the time), Mediocre cards are neutral (since you don’t see them much and they do much of the pushing/pulling of the games), Strong cards are exciting (they make you want to build decks around/with them and have large impacts on the game), and Broken cards are frustrating (they actively dissuade people from playing because they make decisions matter less and constrain deck-building options).

Ideally, then, we want buffs to push cards from the “boring” end of the power spectrum into the “exciting” region, without going too far and making them “frustrating”. This adds enough power you see cards and want to use them, but not too much that they actively start shoving too many other cards out of standard play. Achieving that goal is incredibly tricky in Hearthstone because of the numbers we have to play around with. Changing a card’s mana cost by 1 can turn something from Broken to Weak (see Mana Wyrm), while other times it barely has any practical effect (see Spreading Plague). Single stat points matter too: a 4-mana 4/5 Yeti sees no play, but I feel confident a 4-mana 5/5 Yeti would.

Buffs, then, are high-risk, low-reward things a lot of the time. If you don’t buff a card enough, you see no real effect, while if you buff it too much that effect may quickly become undesirable for the game. It’s hard to hit that right band of “good enough to see play but not strong enough to become irritating” and, if you do, your reward is usually one additional card seeing play in the meta (maybe more if that new card comes with important synergies).

By contrast, nerfing a broken card involves an easily-identifiable problem and potentially opens up a lot more design/deckbuilding space in the meta. This is the simpler route with higher payoff, which is why we have seen it more.

With that in mind, let’s look back on the buffs and see how well they hit that golden region of strength.


When Snip-Snap was first revealed – like many others – I thought the card would be nuts. It turned out the card is quite good, currently seeing play in 41.5% percent of Standard decks. This number increases as you restrict the sample to legend only, where it’s in about 46% of decks (second only to Zilliax at 70%). The card’s flexibility and overall power have resulted in it being good many places, including decks with no other mech synergy to speak of (again, except Zilliax).

For decks without a specific mech component, Snip-Snap is understandably worse. In Mage, for instance, the drawn WR of the card is quite low and it’s mulligan WR is mediocre. Shark Rogue tells a similar story. In decks with a mech component, however, Snip-Snap performs much better. In Mech Hunter, it has the 3rd highest mulligan WR and the highest drawn WR in the deck. Similar stats can be found for Token Druid or Control Warrior, both decks with many mechs. Hooktusk Rogue – which has a middle-of-the-road mech component, sees Snip-Snap being a middle-of-the-road card

So, while Snip-Snap is a good card, exactly how good it is depends on how many other mechs you’re playing. This isn’t exactly shocking news, but it’s solid enough to go just about anywhere, all things considered.

But what about its impact on the game? For mech decks there are obvious synergies when it comes to making or buffing a board; synergies people didn’t have to work that hard to figure out. In the non-mech decks, the card is just an OK card. Overall, I’d say this card was – emotionally – fine, just not that exciting. It didn’t result in much new deck building, instead just representing a card that slotted right into existing decks, giving some a leg up, whether large or small. There’s nothing wrong with that, but there’s nothing right with it either. It just is, like an over-stated vanilla minion.


  • Gloop Sprayer: Currently found in 0.6% of Standard decks, Gloop Sprayer is seen basically exclusively in Heal Druid (like before the buff) and doesn’t seem to have made that deck much better. While the deck saw a small spike in play from people experimenting with it, they quickly dropped the deck again. For a card with as narrow an effect as this, the results of the buff weren’t too surprising. It wasn’t enough gas for the archetype and so the buff had no current effect.

  • Mulchmuncher: Currently found in even fewer (0.4%) Standard decks, Mulchmuncher has failed to make an impact as well. In fact, Token Treant decks using the card didn’t even seen a patch-related spike in play. No one seemed to want to experiment with it (especially when Snip-Snap makes the mech package look more appealing), but at least Mages are rolling it off Conjurer’s Calling less.

Overall, the buffs to Druid didn’t really happen. While the cards targeted for buffs were build-around cards, they weren’t good enough to warrant much interest. We could consider the Druid buffs a Flobbidinous Flop, as they took weak cards and kept them weak.

Read:  The next expansion will come shortly and with it a new free Hero skin


  • Necromechanic: Played in 1.2% of Standard decks, Necromechanic decks saw a small June spike that remained relatively constant. This is likely because the effect of the card is legitimately powerful and you pay no stat premium for it. That said, the card remains lackluster in both the mulligan and drawn WR. This is no doubt a reflection of the fact that while the card text is powerful, it’s only ever as good as the deathrattles you’re triggering twice and powerful deathrattles are either in short supply or come later in the game. It’s the type of card that should be expected to gain power as new supplies of Deathrattles become available for hunter, but right now it’s just a little on slow/awkward side, compared to what else the class can do.

  • Flark’s Boomzooka: Did you remember this card got buffed? Because apparently no one else did either, as it currently sees play in 0.05% of decks. This unfortunately means the sample size for the card is too low to really say much meaningful about its new power level. Judging from the fact that no one seems to play it, I think it might be reasonable to assume that power level is still low (perhaps for the same reason Necromechanic isn’t more of a powerhouse).

Overall, the Hunter buffs kind of happened, in the practical sense. They led to people experimenting with at least one of the new cards, and this experimentation yielded a relatively stable – if low – play rate for a new kind of deck. At least one of them I’d call a success at the time of writing, though the conditional nature of this card might result in future problems down the road when synergies get better.


  • Unexpected Results: Played in 0.2% of decks, Unexpected Results hasn’t really shaped the meta at all. It does lend itself to the idea of building a new spell-damage mage, and within that deck it is a good card, but that deck is currently being far outshined by the Conjurer’s Mage (and perhaps the meta more generally).

  • Luna’s Pocket Galaxy: And here we find our first big issue. Played in 8.7% (and rising) of decks, Luna’s Pocket Galaxy has really taken off since the buff. This wasn’t because people built new decks with it, but rather because it just slotted into the existing Conjurer’s Mage archetype. The good news is that the card has the worst drawn WR in the deck, but has an appreciable mulligan WR. This isn’t surprising, given that mana cheating is more relevant in the earlier stages of the game. The issue, however, is that play experience against the card is miserable. Once the card is played, many opponents only have a brief window to get the game won. It makes impossible turns of cycling decks and flooding the board with giant minions from nothing possible. Putting that into numbers, if it’s played on turn 4 the mage can expect around a 63% win rate. Every turn after this win rate drops by about 2-3%, until most players just don’t bother playing it at all past turn 8 or 9.


The Mage buffs were almost a success, but there are issues: the more interesting, build-around card is a good card for a bad deck, and so it has little real impact. The less interesting high-roll card that slots into an already powerful deck is making an impact, but it’s not one people seem to enjoy experiencing. Pocket Galaxy is the type of effect that there’s not much upside in being good, because it makes resources in the game matter a lot less than they should.


  • Crystology: Currently played in 5.7% of Paladin decks, Crystology is, in my opinion, mistake number two. The card was already good at two mana, and at one it becomes insane. In Mech Paladin, it’s the best card in the mulligan, 2nd best when drawn, and nothing else is really close, barring Zilliax. Holy Wrath Paladin is a bit different statistically (as drawn win rate gets weird when you often draw every card in your deck to win), but it is again the best card in the mulligan and nothing else is remotely close. Indeed, decks including the card jumped in average win rate from 48.1% pre-patch to 54.1% post-patch. While some of that might be due to the next card, it’s clear that Crystology is massively broken at 1 mana.

  • Glowstone Technician: Moving from 6 to 5 mana, Glowstone Technician finds itself a perfectly average to slightly-above average card in Mech Paladin. This is right in that golden region of powerful but not broken territory buffs should aim for, and the result is basically one additional card in the meta. Neat.

Overall, these buffs were half successful. The reason people are less frustrated with Crystology than they should be right now, I think, is two-fold: first, they don’t outright lose to the card itself, but rather its downstream effects. It’s not as visceral as getting punched by a Leeroy. The second issue is just that Paladin isn’t too common these days because other classes are better. If Paladins manage to get more of a foothold, however, every single good version of the deck will play Crystology because that’s how powerful it is.


  • Extra Arms: Currently played in 2.5% of decks, Extra arms is a powerful tempo tool available to Priest. Given to any other class that had a powerful tempo plan, the card would be immensely frustrating but, since this is Priest, the card hasn’t been making too many waves in the meta. The closest we’ve seen is the rise of a halfway-decent Mech Priest in which Extra Arms is a powerful card. For the sake of comparison, decks playing this card were sub 40% win rate before the patch and have risen to nearly 50% since. The card is clearly much more powerful, but Priest cannot leverage it too well. Thankfully.

*Cloning Device: Played in 0.3% of decks that I’ve never seen once, Cloning device hasn’t had any real impact on the meta. The card was buffed, but no one really cares right now.

Strictly because of the class Extra Arms was given too, we could call that buff a success. It provided a new archetype for a class, though it still doesn’t see much play. I’d right more, but Priest is almost never played on ladder. Let’s just hope they don’t get the early game to really abuse how powerful extra arms is.


  • Pogo-Hopper: Currently played in 2.3% of decks, this buff might be the one that generated the most interest, judging by games played with the deck. Given the variety of builds running this card, it’s hard to nail down it’s overall power level, but overall Pogo decks have jumped from about a 30% win rate pre-patch to around a 45% post-patch. The card clearly got better, but not just because of the mana cost itself, but also because of what that mana cost meant: it could be tutored out reliably by Witchwood Piper. This buff spawned a whole new competitive archetype that isn’t overbearing in terms of its power level, and I’d call that a great success. But I would, wouldn’t I?

  • Violet Haze: Played in 0.1% of decks for some reason, this card was bad at three mana and remains bad at two mana. It saw no spike in play or win rate. Card is bad.

Pogo-Hopper is the most successful buff we’ve seen so far. It encouraged a whole new archetype to exist – bringing a few other new cards along with it – without pushing anything out of the meta. For now, we can call that a win. There is a risk that future expansions might boost this effect to undesirable heights – like how people often disliked Jade decks – but that remains to be seen.


  • Stormbringer: Played in 2.3% of decks, Stormbringer saw a large initial surge in play that has remained relatively constant, though the sample size for this card is still too low to say much reliably about its power. Specifically, the card’s relative placement in mulligan/drawn WRs seems to vary quite a lot between lists and none of the lists have too many games logged. On average, the stats for the card look to be in the playable range

*Thunderhead: Played in 5.2% of decks, Thunderhead sure is a powerful card these days. Decks including the card saw about a 2-5% boost in their win rate since the buff, and Thunderhead is among the best cards in the deck so long as you play enough cheap overload cards. It can quickly grab a board and snowball it, which is exactly what Shamans like doing. Before the buff this deck wasn’t seeing any play, so this can be counted as a tentative success

While both these buffs are reasonable, there’s some reason to be concerned for the future. Given how powerful Thunderhead now is, I’d say it’s on the verge of that broken territory. The primary thing holding it back is the lack of good overload cards. As more get introduced in the coming expansions, it’s possible this may tip decks from good to broken. But for now, these are two success stories.


  • Spirit Bomb: Played in 0.1% of decks, it’s hard to assess the card’s performance, even if its impact is non-existent. There’s not much to say outside of it doesn’t look better than Soulfire, currently, and many lists aren’t playing that card either.

  • Dr. Morrigan: Played in 0.7% of decks, Morrigan doesn’t really have any impact either. The closest we can find is the meme-level plot twist Warlock where she looks to be a slightly below-average card. Not much to talk about it.

If you want an example of how bad cards not getting buffed enough remain boring, Warlock is a fine one. Neither of these cards are good build-arounds, they aren’t seeing almost any play, and they aren’t having much impact when they do.


I’ll talk about both Warrior buffs because they have the same story. Neither Rover or Nullifier are seeing play on purpose. The only time they are seen is when they get discovered by Omega Assembly or Dr. Boom and now they’re slightly better than they used to be, but they aren’t having huge impacts either way. Warrior is another case of failed nerfs, but that’s largely because the Boomsday Project was already so good to them there wasn’t much room for improvement without making things oppressive.

Overall TL;DR

In sum, the buffs showed many fewer successes than they did failures inasmuch as meta impact and play rates are concerned. Not including Snip-Snap, there were 18 total buffs. In terms of seeing play, 10 of them had no effect, 4 involved changes that resulted in a single card seeing play or existing decks getting stronger without changing, and 4 resulted in more than just the buff target seeing play, though two of those were pretty marginal. Of those successes, there is still room to worry that some of them might grow to undesirable levels of power if they receive a little new support, but the remains to be seen.

Given the nature of nerfs, imagine instead of 18 nerfs had taken place as they usually do, targeting the most powerful cards. That level of change would – I'm certain – far exceed the impact of the buffs, demonstrating how hard buffs are to get right. It would probably throw the entire meta out for a new one (though whether that's better or worse is questionable).

The best ideas I could think about when it comes to improving the impact of future buffs involves both a diversifying and narrowing of the buff targets. They should be diversified in that they shouldn't focus on a single expansion arbitrarily. I get that there's flavor there, but it's hard enough to get buffs to work on their own, and you don't need to make that task harder on yourself by restricting the range of possible considerations to a small handful of class cards from one set. The way future buffs should be restricted is that they should target build-around cards or synergies. Those evidences the most success. There's not much to do with with a Spirit Bomb, say, in terms of encouraging new cards or strategies. Best card, Spirit Bomb starts seeing play and that's about it. By contrast, Pogohopper and Necromechanic ask you to build around them a lot more, so targeting those cards might be the best way to improve the scope of impact away from single cards and towards multiple new cards seeing play.

And for all our sakes, just don't go making already good cards better.

Source: Original link

© Post "The Buff Patch Retrospective" for game HearthStone.

Top-10 Best Video Games of 2018 So Far

2018 has been a stellar year for video game fans, and there's still more to come. The list for the Best Games of So Far!

Top-10 Most Anticipated Video Games of 2019

With 2018 bringing such incredible titles to gaming, it's no wonder everyone's already looking forward to 2019's offerings. All the best new games slated for a 2019 release, fans all over the world want to dive into these anticipated games!

You Might Also Like

Leave a Reply

Your email address will not be published. Required fields are marked *