Puzzle Strike Upgrade Pack, Part 3
Sunday, October 2, 2011 at 11:27AM I covered the non-gameplay components and the new chips in the Puzzle Strike Upgrade Pack here and here, and now it's time for the character chips. There's a complete set of 30 character chips in the Upgrade Pack, even though not all of them have changed. I thought this would be helpful for those who want to play mirror matches with the chips in the base set, such as Setsuki vs. Setsuki.
Let's start by talking about why any of the character chips are different at all. Shouldn't we not really mess with what's already been done? I think there's a value in letting things be for a while. Game balance doesn't have to be perfect, and can't be perfect, and there's a cost to changing balance stuff around. We have to consider the logistical difficulty of getting new chips into your hands, the possible confusion that might surround such an update, that any changes in gameplay in this upgrade pack take weeks or months to test and would cause a delay in working on the full expansion, and so on. It would certainly have been a whole lot easier not to change any game balance of the character chips, so with so many reasons stacked against making changes, why are there changes?
The Call to Action
It's because when a problem in a game is big enough, it demands attention and fixing. Often in board and card games, the approaches to this problem are a) do nothing and b) move on to the expansion and just forget about the base set. Neither of these strike me as what to do if you truly care about a game though. Because I have a serious commitment to Puzzle Strike being a tournament-viable game, if there is any problem in the base set that threatens that, I think I have no choice but to address it. And right now, this is the best method we have.
The main problem, which took months of tournaments to really fully discover, is that Valerie is too good. For those not familiar with the challenges we face in balancing any asymmetric game, the problem *isn't* that Valerie is the best character. That is no crime, and there is always some best character. It's only a problem if a character is too far beyond the "top tier," meaning too much power-difference between that character and the rest. As time went on, Valerie proved harder and harder to beat, dominating tournaments to the point that many players were losing interest in even competing. It's somewhat unusual that it took so long to discover the problem, and it lies in a false choice with her Burst of Speed chip:
For a while, there was an interesting tension between using the chip early (small benefit that snowballs over time) or late (big turn that might just win outright). In the end, tournament players found that playing it at the first opportunity--and usually using that extra turn to buy a Combine--was too good of a play to pass up. It's boring and really powerful. Interestingly, you might not even care about this in your games, because it was mostly a problem amongst the expert players and only after they played quite a bit. That said, if the game is degenerating at the expert level, some fix is required, so a new Valerie was the #1 goal here. Many versions of her were tried, but here's the final change. Burst of Speed is replaced with Three Colors:
Note that Three Colors is a move of hers from Yomi, and it's a series of three paintbrush swipes, each a different color. In a fighting game, it would function similarly to Fei Long's Rekkaken punches. In Puzzle Strike though, it allows her to chain together three actions...if you have a diverse enough deck to take advantage of it! During playtesting of this chip, the most encouraging thing about it was that the very best players called it a nerf (and of course it is, she was too powerful before) but some average players saw it as a buff! That's because they were not playing Valerie in a boring way before, they were trying lots of different strategies with different kinds of puzzle chips, and this new chip enables even more diverse strategies. So boringness is nerfed and fun is buffed, so to speak.
While addressing Valerie was the top priority, there were two other goals as well: address the "mono-purple" concerns and fix an unfortunate "flavor" problem with Lum.
Mono-Purple
Mono-purple refers to a strategy where you buy mostly purple chips (combine, crash, double crash) and money, and ignore the puzzle chips. Doing this is a kind of baseline strategy that should be beaten by more interesting strategies that use lots of different puzzle chips. The experts were long able to beat mono-purple with several different approaches, but then came the rise of mono-purple Valerie and her too-good Burst of Speed. This problem was compounded by Lum and DeGrey. Those two weren't too good, but they were very strong, and they happened to be most powerful when played in boring mono-purple ways. We got really unlucky that these types of characters ended up so strong, while some more interesting characters ended up weaker.
Puzzle Strike is one of the most complicated game systems I've ever worked with, as it's so deeply interconnected. Now that we know more about the system, I think we can better state the problem. It's that each character gives you a bit of a push in one strategic direction or another, and that if a character gets even a very small push toward buying purples, the game system magnifies this effect and it snowballs in a bad way. The good news is that we can avoid this problem by making sure characters push you in more interesting ways.
The Powerful and The Weak
I mentioned before that we happened to get really unlucky with which characters came out top tier or bottom tier in power. Let's start with the character we got most lucky with though: Setsuki. Her Double-Take chip opens up a world of strategies with all sorts of puzzle chips. There is a lot she can do, and just buying purple chips only is very suboptimal play with her. She's powerful and interesting and exactly what a Puzzle Strike character should be. There are no changes to her gameplay in the Upgrade Pack.
Geiger, like Setsuki, is pushed by his design to play in an interesting way. He's an "engine" builder, meaning he can get a whole bunch of draws and actions going, and he can use a deck with just a few purples that he can dig for. It's pretty cool but there's a major problem: Geiger turned out too weak to be relevant in tournament play. In the testing before the game's release, small changes in Geiger made him either too good or too weak, so he was hard to get right. He ended up too weak, but in the Upgrade Pack he has that same gameplay theme, but a high enough power level that there's another fun character who is a real choice in competitive play.
In Geiger, we got unlucky because such an interesting character was too weak. With Valerie, DeGrey, and Lum, we got unlucky for the opposite reason: because such boring characters are too strong. They weren't intended to be boring, of course, but each of those three characters happened to end up top tier and at their best when they focused mostly on purple chips. Valerie we already covered, so let's look at DeGrey.
In the base game, DeGrey's Troublesome Rhetoric chip was basically never used by experts. We could have replaced it with a wound chip and they wouldn't have even noticed. To me, that is a major design problem, even if it's not a balance problem. Further, his best strategy was to use No More Lies to trash Troublesome Rhetoric, then play a very small deck of just purple chips. Yes it was fun in that it was different from other characters, but it was also like playing a non-character who was just mostly degenerate. This is not the kind of character you want ending up being one of the very best in the game.
It actually took longer to figure out how to address this than it did to fix anything else, but we finally did it. Troublesome Rhetoric is now a chip you'll actually use, and No More Lies still offers you the best trashing ability in the game, it just doesn't do it to the extreme degenerate levels it did before. DeGrey feels more like an actual character now, rather than the lack of one (because he used to just trash his character chips).
Then there's Lum. We got the same kind of unlucky with Lum, meaning he is very powerful and also very boring. He's often at his best just buying purples, while some of the more interesting characters just happened to turn out too weak to compete with even footing against him. Lum has even another problem on top of that though: he doesn't feel like the gambling panda he's supposed to be. (He does in Flash Duel and Yomi though!) Some joke that he is "Lum, Careful Accountant" in Puzzle Strike. Originally, the idea was that a gambling panda would have more money than the other characters, but this turns out not to result in very interesting gameplay. Because his flavor was so far off the mark, I wanted to revamp him to make more sense, along with being more strategically interesting. Poker Winnings is out and Jackpot is in. Also, Panda's Bargain is more gambley and less "you just get tons of money-y." The new Lum, Gambling Panda:
Midori and Jaina were in the same boat as Geiger, basically. They have interesting varied strategies, but not enough power to be used in tournaments. Now they play similarly to before, but with high enough power to be real threats.
That just leaves Argagarg, Grave, and Rook. These characters were in mostly fine shape to begin with, so not much is different. Rook just has a wording change, but the same gameplay. Argagarg is just slightly weaker, partly to help the 4p game and partly because he was kind of dangerously strong in the new metagame. He's still a beast though. The main problem with Grave was that his ability to trash his crash gem on turn 1 and trade it for a 7 cost chip meant that we basically couldn't make any 7 cost chips or it would make Grave way too powerful. He can't do that trick anymore, and he got a little boost in another area to make up for it. And just to prove how fun a 7 cost chip can be, check out Custom Combo from my Part 2 post about the Upgrade Pack.
Conclusion
I know some players will still be upset that there are any changes at all, but as I said at the beginning, I think it was absolutely necessary to adjust game balance in a way that keep Puzzle Strike a tournament-viable game. The best case would be if we had addressed these issues before the original release. We certainly tried and tested as well as we could, and even then it took months of tournaments after release to fully grasp Valerie's power. So given the situation we're in, I think that offering players a rebalanced set of character chips is a far superior option to doing nothing, as would more commonly be done here. I hope you agree.
So that's the Puzzle Strike Upgrade Pack. So many players asked for playmats, that the four playmats alone are reason enough for lots of players to want it. The screens with the cool 8-bit art are a nice bonus. Some players were dying for a few new chips, and the three carefully chosen new chips should serve you well. Finally, the adjusted character chips power up the weakest characters while taking the purple out of the strongest ones. (You can see all those chips here, by the way.) All in all, I hope you find it an upgraded experience!
Reader Comments (30)
new Troublesome Rhetoric had soooo much discussion and back and forth, it could be an article unto itself.
Ditto the comment that Burst of Speed seemed fine to me until someone pointed it out, and then it seemed obviously way too good. Hindsight, 20/20, etc.
I immediately realized that BoS should be a first turn play every time. Your "general power" grows exponentially over time so it makes sense to use it early, especially since it gets itself out of the way. I just never realized how much further above the competition that put Valerie.
Alright, several comments on what I see here:
1. Sirlin said that Degrey was one of the better-than-we'd-like characters prior to the rebalance, and this is flat out false. Degrey was a low tier character. That being said, there were definitely some changes in order. No more lies and troublesome rhetoric were changed because, as Sirlin said, it defeats the point of asymmetric starts if the character is going to eliminate his unique and defining moves most of the time. Pilebunker was changed because of a group of playtesters who whined very loudly and repeatedly about it. There wasn't actually anything wrong with the chip from a game theory or strategic point of view; largely their complaint was that they only knew one strategy, and the old pilebunker happened to tear that strategy to shreds. It's not a fantastic reason, but it's the truth. However, anyone who would say that this pilebunker is nerfed over the old one is a fool; while it isn't as disruptive it grants quite a bit of additional draw control, and is at least as good as the old one. This new pilebunker definitely takes a lot more skill to get your worth with though, which is a good thing. So in the end I really do like that change.
2. I'm disappointed with some arguments I'm seeing in the comments for playing burst of speed the first time you draw it. While it's true that in >95% of situations it's the right thing to do, every single argument here has just swept under the rug and ignored loads of factors. For starters, there's the acceleration your opponent gets when you burst, and nobody seems to have considered the implications of playing burst with forks either, which can frequently lead to mid-game auto-kills (and yes, turn 1 burst is that stupidly good that it's better than likely mid-game auto-kills). The bottom line is that using burst later is not actually that minus EV, but it takes a heck of a lot more skill in order to achieve only a slightly reduced performance from turn 1 burst.
3. What Sirlin said about it being discovered post-release that turn 1 burst was so extraordinarily good is not 100% true. It was known before the game came out that was almost always the right play. What Steelcoil said is the truth; it simply took a while for a player to come along who put their mind to abusing it maximally.
4. Directed at x1372: I agree with you that Jaina's big buff in this rebalance is the removal of her prior 2 near-unwinnable matchups (Val and Lum) and that by comparison the pig on playing with fire is a pretty small buff. Just a disclaimer on your video evidence of her viability though: I only used Jaina in 7 of the 25 games I played that tournament, and I only used her against people whom I thought she would be especially effective. I play 7 different characters in tournament, so if I pick Jaina against someone it means I think that they're least likely to be able to handle her. Even still I think Jaina's a mid-to-high-tier character though? It's hard to tell, because the new version still seems so remarkably balanced even 2 months after it's character chips were finalized.
5. I haven't seen anything in the comments about the change to hex of murkwood, but Sirlin said to ask me or Chumpy if you were curious. The primary reason was, as he said, that Arg was too good in the new metagame. Consider for a second who were Arg's bad matchups previously? The answers were Grave, Degrey, Valerie, and Lum. Who got the biggest changes in playstyle in the rebalance? Also Grave, Degrey, Valerie, and Lum. In fact, not one of those 4 characters constituted a bad matchup for Arg with their new version. Most of them were ever favored to lose to Arg now. Nerfing hex fixed the problem by making Jaina, Midori, and Rook into bad matchups for him. A couple of other reasons for the hex nerf are basically more precise mathematical ways of stating that the old hex resulted in disparagingly random outcomes, and the new one fixes that. There are still more reasons after that, but they're pretty nuanced and it's not worth listing it all here.
I'd like to point out that against a strongly defensive pool, or a character like Rook, It might be better to save BoS for late game when you can take two turns immediately and crash for 4 twice, or some such amazingly offensive or defensive play.
Burst of Speed not only gives you an insanely fast rush, it can also speed up your econ too. So if a monopurp rush doesn't look favorable, you can just as easily use Burst to out econ them. Rook is definitely not one of those characters where that is necessary. Val already has Chromatic Orb to deal with Stone Wall if she ever needs it, but what really makes the matchup bad is that Val generally out classes Rook. Val can shell out 4gems even faster than Rook can with more consistency, and since Rook has no reliable way to defend against this, he just crumbles under the pressure.
But yes, Burst of Speed can be used to greater effect as a closer with a lot of set up and/or luck. Heavy deck control is a must and you need a proper bank to do it, but seeing as how T1 Burst is a lot more consistent, it leaves something to be desired from late Bursts. It's a shame, I had a lot of fun chaining Combo Practices with late Burst, even though I already had plenty of fun beating people up with monopurp. Ahhh the good old days.
Sirlin, sorry if I missed this, but does the upgrade work w/ the original Collector's Edition? (the one with the wooden chips). Or would I need to buy the new version of the base set?
Doug, I have talked about that before and I'm sorry to say that the upgrade works only with the second edition of puzzle strike (the cardboard chip version). I originally planned for everything to work with the collector's edition, but the manufacturing problems made that impossible. So the only way I could really continue was to switch all manufacturing over to this current method. I wish I had better news on that, I'm sorry.
I think that in 3,4 player games, the number of blue shields effects that can affect an attacking player should be capped. We had a game where surgical strike was consistently rewarded with 3 'really annoying' reactions. This is not reasonable at all.
Many puzzle chips are good once or twice but then become a wound once crashing becomes more important. For example, most of the risk in risky move I've found comes from simply having it, not playing it. I like the idea of having to choose what to play, but later in the game, the need to crash means there isn't really a choice. I would like to see a chip that has an ongoing effect that allows/forces you discard a chip each turn for some benefit, even if it's a small one.
Using reds in multiplayer can be risky if everyone has a blue, but most reds are already really good in multiplayer because of the ability to affect every player. If you're having that much trouble fighting off blues, you can just decide not to play your reds. Dealing with reactions when playing reds and crashing is part of the game; you have to make good decisions on whether playing it is worth the risk of getting reacted, rather than just playing it mindlessly.
And yes, most of the decisions come from the early-mid game. Having chips like Risky Move in your deck early help you indirectly do better later rather than directly. Maybe using it will give you a DCG 3 turns later, and then you'll do better than the opponent who does not have a DCG. There are still decisions in the late game. Maybe you bought Stolen Purples a turn earlier, and now you have to decide whether to play Stolen Purples or a Crash when you're at 9 height, when both of which can potentially win you the match. Or maybe you aren't deciding on whether or not to crash, maybe you're deciding on how you want to crash. Do you want to play it safe and crash 2gems and 3gems? or do you want to save up to a 4gem to potentially end the match? The amount of relevant decisions you can make aren't as great as in the beginning of the game, but decisions still exist during the endgame.
Dan: I was the one who asked about the hex of murkwood change on the first page, so thanks for the reply. Missed it until now.
So you gave Argagarg bad matchups on purpose? (or at least that was the main goal) That seems kind of weird to me. I always think of bad matchups as a side effect of having asymmetric characters, not an intentional balance tool. Especially in a game where blind-picks are advocated, forcing bad matchups seems like a bad idea