StarCraft II Would Benefit From A More Aggressive Tuning Philosophy

Thinking on Protoss performance after Jönköping

ESL Masters 2023 Summer came and went last week, capping off the latest professional event of StarCraft II. Sadly, it was a tough tournament for Protoss: while our friends from Aiur accounted for 50% of entrants, only 3 managed to make it to the top 16.

The Art of Tuning

I don’t think you can draw any conclusions from a single tournament. But I also think people are sometimes overly cautious in interpreting and forming judgments about the current game state. And personally, I think tuning is just as much an art and a philosophy as it is a hard-nosed, data-driven, objectively correct methodology.

For instance, back in the mid-2000s, Blizzard was more aggressive about patching and updating Warcraft III compared to the other RTS makers on the market. They were famous for regular developer updates and balance patches and all sorts of random changes like turning off demotions on the ladder (and then undoing that pretty fast, if I recall correctly). Notably, these changes weren’t directly tied to new content or expansion packs, in contrast with other studios.

Nowadays, this is a common model; War3’s iteration speed might even be considered slow by modern standards. But back then it was revolutionary. While everyone else on the market told their fans to wait for the next game or the next expansion pack, Blizzard was patching and updating and chatting with their fans all year round.

At the time, many of us were confused by this. This can’t be worth the investment! They’re not selling anything new! It’s a cost sink! And we were probably right, at least in material terms. But we failed to appreciate that this was a philosophical difference rather than a failure on the part of some Blizzard accountant to use Excel correctly. Blizzard updated their games because, well, they just really liked their games, and because they took a long view on the finances rather than seeing everything in one or two-year increments between expansion packs and new titles.

Schools of Thought

I find it funny that I remember Warcraft III as an aggressively tuned game, not only compared to other RTS titles, but also compared to other esports, too. Recall that the Counter-Strike of that era, for example, is literally remembered by its patch version. I find it funny because close to twenty years later, the modern approach to RTS tuning feels too conservative.

I think there was a window of time in StarCraft II, starting from the launch of Legacy of the Void in 2015 to the design patch patches of 2016 and 2017, when the title adopted a more aggressive philosophy. And I think that philosophy served the game well, contributing to years of growth paired with excellent top-tier esports competition.

Soon after, though, Blizzard shifted in the direction of a more static, Brood War-like approach. And that struck me at the time as a reasonable idea, though not one that I personally agreed with (I’ve been a proponent of regular StarCraft II design patches for a long time). Unfortunately, it didn’t quite pan out; 2019 was a fairly bad year of competition for StarCraft II, in which Zerg was so plainly overtuned that Rogue declared it publicly. And I’ve written before that I think one reason the game got into this state is because Blizzard took a more incremental and reactive approach to balancing and designing the game than they had in previous years.

I think it’s worth reflecting on that as we consider the direction of StarCraft II going forward. The game is far from dead. There’s still plenty of runway to make it better. I think it’s worth taking a chance and trying to iterate on some of the title’s weaknesses instead of defaulting not only to the old War III model, but to an even more conservative the-game-is-balanced-and-anything-we-do-might-make-it-worse mindset that I see as prevalent within community discussions today.

Which brings me to the subject of this article, which is that Protoss has not been doing great.

For Aiur!

I think there’s been pretty widespread acknowledgement that Protoss did not have a great 2022/2023 season. The general sentiment - perhaps most memorably articulated by Nathanias - was that Protoss was not going to do well at the World Championship in Katowice. And indeed, they did not.

I decided to run the numbers on this, going through the last three years of professional competition. I looked at the round of 16 race participation in Premier tournaments and marked each race as under- or over-performing, at 4 or below and 7 or above players participating, respectively. While this is a course-grained measurement, I figure in a game with healthy representation, under and over-performing should balance out.

Of the 15 Premier Events held in the 2022/2023 season, Protoss won 3 and was a finalist in 5. Terran won 4 while appearing in 10 finals; Zerg did the best, winning 8 and participating in 12. As for the rounds of 16:

GSL Code S is unique because it had a round of 20 followed by a round of 10. I decided to take the round of 10, because the racial distribution in the first stage mostly just speaks to whoever is left playing professional StarCraft II in Korea. I used 2 or 5 players as my respective limits for under- and over-performing:

I don’t expect racial representation to be exactly perfect, and this methodology is imperfect, too. But it seems clear to me that Protoss had a tough year - underperforming in 6 out of 15 events, but only overperforming in 1.

Anyway, it’s just one year, so I looked back at the previous two seasons, too.

In the 2021/2022 season, out of 22 events, Protoss won 6 and appeared in 9 finals; Terran won 6 and appeared in 10 finals; and Zerg again did the best, winning 10 and appearing in 18 finals. As for the rounds of 16:

(Note that for the rounds of 16, I excluded the EPT Circuit for this and the previous year; it features so many Protoss players that it’s difficult for them not to be overrepresented in the round of 16, and there are so many events that it skews the data. One example: More Protoss players entered the DH Masters 2021 Winter: Europe tournament than Zerg and Terran combined. But guess what two races were in the finals.)

Protoss again had the toughest year of all thee races, underperforming 7 times while overperforming 2 times.

Finally, in the 2020/2021 season, out of 20 events, Protoss won 4 and appeared in 11 finals; Terran won 5 and appeared in 7 finals; and Zerg once again did the best, winning 11 and appearing in 14 finals. As for the rounds of 16:

This looks pretty reasonable to me, at least as far as Protoss is concerned; Protoss underperformed 3 times while overperforming 2 times.

It’s Not Looking Great

I think one objection folks will raise to this methodology is that as far as it’s concerned, Zerg may have been underrepresented in 2020. So shouldn’t we buff Zerg, too? And that’s what I mean by tuning being more of an art and a philosophy than an objectively correct methodology. Zerg was consistently winning tournaments and appearing in finals; there was no concern about lack of Zerg representation in high-level StarCraft. We shouldn’t abandon common sense because there’s a single outlier year in the data.

As as Protoss goes, I think many people will look at this data and conclude that the tough years they had beginning in 2021 - and continuing on to this day - has more to do with player retirements than anything else, given that things were probably OK back in 2020. And I think others will quibble with the choice to exclude the EPT Circuit. Hey, at least Protoss does well in mid- to high-GM, right?

But another way of framing that is that multiple years of underperformance at the game’s highest-level global tournaments for one of the game’s three races is a perfectly normal and acceptable thing.

Which, well… maybe not?

I mean, setting aside the academic arguments that this is an asymmetric game that is difficult to appreciate if not all three races show up to the party; I hold the more practical perspective that there are not enough races in this game for one of them to consistently perform poorly for an extended period of time. This is not F1, there are not ten teams out on the grid - there are three.

And one of them is kind of sucking!

I think it’s worth doing something about that. I think racial representation in top-tier tournaments is a worthwhile goal, to be considered alongside other goals like theoretical racial balance (which is itself an incomplete goal, given how many constraints we impose on map makers to keep the game balanced). And I think we should strive to correct for the lack of Protoss representation in high-level tournaments. And the great thing is, we actually have a mechanism to make something happen - the Zerg cabal the professional player’s council.

I get that some folks blame the professional player’s council for worsening Protoss’s situation with the last patch. But I think the data shows this is a longstanding, pre-existing problem. And I think we’ve learned from the history of design and balance in StarCraft II that a conservative, passive approach to the game isn’t inherently good, either. One of the best periods in the game - the few years after Legacy of the Void’s launch - contained lots of large design changes and balance patches. So I think it’s worth being proactive and doing something, especially since we still have another year of StarCraft esports.

I’d love for the professional players council to work on this, or at least consider and make a public statement on whether racial representation is a problem and what their take is. Maybe it means map pool adjustments, or balance changes, or design tweaks, or something else entirely. I’m open-minded!

I think this is a challenging problem, given the many competing values at stake - not only game balance and racial representation, but also quality of games, impact to lower-level players, etc. But I think we should lean on the mechanisms we have and, if we think they’re flawed, suggest how we can fix them instead of giving up entirely. My take is that the professional players council is capable of handling this, with some tweaks to their approach compared to last time. So I hope they take on this work!

Until next time,
brownbear

If you’d like, you can follow me on Twitter and Facebook and check out my YouTube and Twitch channels.

P.S. I went back and forth between “aggressive” and “well-rounded” in the title. I support both, basically. But I thought aggressive captured the spirit of this particular article better.