Hatchery vs warp gate

[Adventurer-16000863]

@TerranicII-1450

That might be a symptom of the same problem though. I don’t think Zerg’s overall population has ever been greater than Terran or Protoss

You’re saying masters/diamond being 40% zerg at the time doesn’t indicate the probability of zerg being selected for grandmaster. You’re saying Zerg should be 33% of masters/diamond and as such a 40% gm representation is 7% above expected. Except that’s exactly what I am saying almost identically. Protoss representation should be 33% of masters and yet they make up 45% of grandmaster. Thus it proves the point that modern Protoss dominance is the largest imbalance the game has ever had.

Assuming anything except 1.0 / NumberOfRaces makes assumptions about the probability of racial selection, and removing those assumptions is painful and complicated. But if you were to do it, you’d find that about 10% of people either pick the strongest race or avoid the weakest which is an obvious conclusion to come to if you look at historical trends. The moment zerg is nerfed, protoss representation in masters spikes and zergs goes down by an identical amount. The point is, the argument that zerg’s lower population should be predictive of GM is wrong because it ignores race switching. 1/3rd is the proper expectation for masters league, meaning a masters player is equally as likely to pick each race IF the game is balanced.

If zerg is equally likely to be picked in masters league & zerg’s population has 100 mmr higher average, then they should be the most represented in grandmaster. But the reality is that zerg is highly disadvantaged and this makes it less likely to be picked by masters players and the few who pick it are less likely to be advanced to grandmaster. And that’s how you end up with 2x as many protoss in gm as zerg. Probably 1/5th of the protoss in GM are zergs who just don’t want to play zerg on this patch, and who could blame them.

[TerranicII-1450]

@Adventurer-16000863

You’re saying masters/diamond being 40% zerg at the time doesn’t indicate the probability of zerg being selected for grandmaster.

I am saying that the same mechanisms that cause a race to be overrepresented in GM would also affect their representation in Masters and Diamond. Zerg’s overall population has always been smaller than the other two races, yet they were disproportionately represented in the upper leagues throughout the BL Infestor era. Based on the numbers that you provided, that suggests that the BL Infestor era was worse than the game’s current balance (which is not to defend any issues that currently exist).

@Adventurer-16000863

Protoss representation should be 33% of masters and yet they make up 45% of grandmaster. Thus it proves the point that modern Protoss dominance is the largest imbalance the game has ever had.

If league distribution is affected by balance, why would you even assume that Protoss should make up 33% of Masters? They don’t actually hold that much of the population. The population numbers pretty much go Terran > Protoss > Zerg, whether that is due to the familiarity of the races or personal preferences.

@Adventurer-16000863

Assuming anything except 1.0 / NumberOfRaces makes assumptions about the probability of racial selection, and removing those assumptions is painful and complicated. But if you were to do it, you’d find that about 10% of people either pick the strongest race or avoid the weakest which is an obvious conclusion to come to if you look at historical trends. The moment zerg is nerfed, protoss representation in masters spikes and zergs goes down by an identical amount. The point is, the argument that zerg’s lower population should be predictive of GM is wrong because it ignores race switching. 1/3rd is the proper expectation for masters league, meaning a masters player is equally as likely to pick each race IF the game is balanced.

I am wondering where you pulled that “10%” number from, as there doesn’t seem to be a basis for it. If 10% of the total population was switching based on balance, we would see wide swings in the overall population between patches; instead, the population seems to have remained Terran > Protoss > Zerg throughout SC2.

@Adventurer-16000863

If zerg is equally likely to be picked in masters league & zerg’s population has 100 mmr higher average, then they should be the most represented in grandmaster.

Zerg is the least likely race to be selected. You wouldn’t expect them to be the most represented race.

@Adventurer-16000863

But the reality is that zerg is highly disadvantaged and this makes it less likely to be picked by masters players and the few who pick it are less likely to be advanced to grandmaster. And that’s how you end up with 2x as many protoss in gm as zerg. Probably 1/5th of the protoss in GM are zergs who just don’t want to play zerg on this patch, and who could blame them.

Edit: There seems to be no justification for assuming such wide swings in the overall population.

[Kelthar-1277]

Terranic, you couldn’t possibly be suggesting that Batz is just pulling numbers out of his rear and making things up, could you? That has never happened before!

Man, I almost was able to type that with a straight face.

[Adventurer-16000863]

@TerranicII-1450

I am saying that the same mechanisms that cause a race to be overrepresented in GM would also affect their representation in Masters and Diamond

Yes, exactly.

@TerranicII-1450

If league distribution is affected by balance, why would you even assume that Protoss should make up 33% of Masters?

Because that’s how it would be under a balanced game scenario.

@TerranicII-1450

I am wondering where you pulled that “10%” number from, as there doesn’t seem to be a basis for it

Spikes in racial population immediately following a balance patch.
Tracking individual accounts to see if they race switch.
Mining replays from websites to see if a player switches races in the replays.
Checking player rank histories to see if they are ranked for multiple races.
Calculating population growth via demotion metrics & seeing if they are adequate to describe population growth (they are not, races grow via players switching races in addition to players being demoted).

@TerranicII-1450

Edit: There seems to be no justification for assuming such wide swings in the overall population

This one is easy to verify by simply glancing at a population chart for masters and diamond. If protoss in masters go up, where did they come from? They came from diamond. We’d expect diamond to lose protoss, but they don’t, they also gained protoss. What about platinum league? Surely that’s where all the protoss are coming from! Nope, platinum also gained protoss. What about gold? Nope, gold didn’t lose any protoss. Maybe they came from silver league? Darn it, silver league didn’t lose any protoss. What about bronze! Nope, bronze also didn’t lose protoss representation.

Hmm. Where are all these protoss coming from who are filling up the ladder!

Oh wait it’s race switching.

Players switch races away from the weakest race. It’s just a fact of reality. Players like to win and when you play zerg you get railed and that sucks and it’s not fun so over time players stop playing it. The proper assumption for Masters league is 33/33/33 for a balanced game scenario. This assumes Masters league players are equally likely to select each race (e.g. racial strength has no impact on masters league because there is no difference between racial strength in a balanced game scenario).

@Kelthar-1277

Terranic, you couldn’t possibly be suggesting that Batz is just pulling numbers out of his rear and making things up, could you? That has never happened before!

How does PhD K. consistently pick the lowest IQ response conceivable? It’s amazing. He does this every time. He’s easily the greatest troll to ever post on battle.net. I have no idea how he can do this. It’s obviously on purpose because someone this dumb wouldn’t be able to use a computer to post on the internet, and that’s what gives the trolling away. It’s incredible. Thinking on this level, my brain would have an aneurysm, but he can do it. Wow, what a talent.

[TerranicII-1450]

@Adventurer-16000863

Yes, exactly.

You still seem to be missing the point. 33% is not the target that you would expect for Zerg or Protoss based on their populations. Their populations have been rather consistently below that.

@Adventurer-16000863

Because that’s how it would be under a balanced game scenario.

No, it is not. The problems with that assumption are the existence of random, and selection bias.

Random reduces the total percentage of the population that the three races have to work with, lowering their average.

Selection bias results in an uneven distribution amongst the races regardless of balance, meaning that the ideal percentage in each league is not 33% of what remains when you take random out. In a perfectly balanced game, each race would still have a different target that is based on their population. Even your assumption is that only 10% of players are switching races based on balance.

It’s weird that you seem to be missing these points because they actually make your case stronger. If Protoss if 40-45% of GM when their overall population is or should normally be roughly 27-31%, then the balance is actually further off than if the population was expected to be at 33%.

[Adventurer-16000863]

@TerranicII-1450

It’s weird that you seem to be missing these points

It’s weird that I have a theory supported by a mountain of evidence and you’ve got “what ifs”. Yet the language you use to describe my argument is “assumptions” and “bias” aka a literal inversion of reality.

[TerranicII-1450]

@Adventurer-16000863

It’s weird that I have a theory supported by a mountain of evidence and you’ve got “what ifs”

There is an obvious bias in the way that players select races, independent of balance. This can be shown in the trend that Terran has a larger population than Protoss, and Protoss has a larger population than Zerg. This trend has held even in periods where balance was terrible for Terran or Protoss, meaning that even if balance has some effect on race distributions, it would not result in a 33% distribution for each race in a balanced game. This is a problem with your theory that you are either blind to or purposefully ignoring.

[Adventurer-16000863]

@TerranicII-1450

There is an obvious bias in the way that players select races, independent of balance

That’s called a “what if”, honey. If players select one race over another, independent of balance, you need a mechanism for why that is. But specifically you need that for masters league because masters league is what predicts GM league. GM is selected out of masters league after all. So why is it that people who have mastered the game would have a preference for one race over another without the race itself being the cause. Remember, the race can’t be the cause because you said it’s independent of race. So you’re supposing that masters level players incidentally, by chance, happen to favor one race over another, for a reason other than racial strength, when in reality they are in masters because they make decisions that prioritize win-rate. If masters level players prioritize win-rate it’s obvious they are going to favor the stronger race.

It is mathematically impossible to create a theory where one race is selected more than the other, independent of racial performance, for a league that selects players based on performance. Players who don’t pick race based on performance will automatically be kept out of the league by definition. You’re saying 2+2=7 again. You’ve created a theory that contradicts itself. If you are in masters league it’s because you prioritize win-rate. If you pick race based on some factor other than racial strength, you will be filtered out of masters league and thus not part of the set. So how exactly do you create a masters league that somehow doesn’t select players by performance? Because that’s what your theory is predicting and it’s absurd.

This is the same basic fallacy you used earlier in the thread. You’re using a variable that has no relavance to the causal relationship being described. You’re saying that you painted your car red because cars with bigger engines drive faster. Players favor Protoss for some reason that has nothing to do with the design of protoss. It’s seriously a completely absurd argument.

@TerranicII-1450

This trend has held even in periods where balance was terrible for Terran or Protoss, meaning that even if balance has some effect on race distributions, it would not result in a 33% distribution for each race in a balanced game. This is a problem with your theory that you are either blind to or purposefully ignoring.

Maybe in bronze league you can make the argument that people don’t select race based on strength, and I might agree with that because to know what race is strongest requires game mastery which is something bronze players don’t have. But, master’s league players? Never in a million years will that argument hold water because masters league forces players to make decisions based on win-rate, that’s how you get into masters.

[Asamu-1302]

@Adventurer-16000863

If players select one race over another, independent of balance, you need a mechanism for why that is. But specifically you need that for masters league because masters league is what predicts GM league.

You actually don’t need to know what the specific reason is.
There could be multiple factors that have nothing to do with balance that impact representation, like commonality of aesthetic preferences: you wouldn’t get equal results asking which race has the best/coolest looking units or design, with gameplay entirely disregarded, and that is something that will influence race selection even for players in masters/GM.
Some people like dudes with guns/vehicles, some like technologically advanced aliens with laser/plasma weapons, and some like bug/xenomorphic aliens. Those preferences aren’t going to be equal. You see this in all sorts of games, where certain factions, races, units, weapons or archetypes are more/less popular even among competitive players for reasons that have nothing to do with power level.

There are only a few noteworthy periods of overall representation shifting outside of a ~3% range for any races; early LoTV where Protoss representation dropped dramatically due to the horrific state of PvZ at the time and a couple of other periods of extreme imbalance, like early HoTS when hellbat drops were dominating the meta being some examples.

@Adventurer-16000863

So why is it that people who have mastered the game would have a preference for one race over another without the race itself being the cause

There is no reason to assume it should be equal at any specific level, even if balance were perfect, which it obviously never has been, and balance is clearly different depending on skill level based on per league win rate stats and representation differences.

Masters representation is influenced by balance in diamond, which is influenced by balance in plat, etc… And all of that is influenced by overall race representation, which, based on how little it fluctuates, is likely primarily based on some arbitrary factors that do NOT appear to result in representation being equal (according to what data is available).

@Adventurer-16000863

Maybe in bronze league you can make the argument that people don’t select race based on strength, and I might agree with that because to know what race is strongest requires game mastery which is something bronze players don’t have. But, master’s league players? Never in a million years will that argument hold water because masters league forces players to make decisions based on win-rate, that’s how you get into masters.

If what you’re saying were true, masters+ representation would shift far more than it does as balance shifts, and pros would race switch far more frequently than they do based on the state of balance, but they don’t - in fact, pros almost never switch races, even at times when their race is clearly disadvantaged due to a clearly poor balance state.

Protoss players at the top still stuck with Protoss when it had a ~37% win rate vs Zerg at the pro level in early LoTV, for example. Clearly balance was, at best, only a minor factor in race selection.

[Adventurer-16000863]

@Asamu-1302

You actually don’t need to know what the specific reason is.

You do if you want to prove it isn’t caused by race. Because “thousands of players coincidentally don’t like race X for some reason except race X sucks” isn’t a good argument. I’ll read the rest of your post later, I don’t have time right now.

[Asamu-1302]

@Adventurer-16000863

Let’s explain why that is. Serral wins → zerg is nerfed → nobody wants to play zerg anymore → serral wins some more → zerg is nerfed more → nobody can play zerg anymore.

This explains why Zerg was the lowest overall representation even when Zerg was the best performing race at practically all levels of play (as was the case for most of the first 9 years of SC2). /s

@Adventurer-16000863

You do if you want to prove it isn’t caused by race.

No, you just need to show that balance isn’t the only factor, or that other factors outweigh it, which should be obvious given how little overall representation actually shifts.

Frankly, this whole discussion about representation is way off-topic for the thread anyway, and you clearly have some gaps in your logic and either refuse to acknowledge or try to twist any point that you disagree with, so I’m done with it.

Stick to the topic from the OP (which has some very thorough explanations in the first few posts), or make a new thread if you want to discuss a different topic.

[Adventurer-16000863]

@Asamu-1302

No, you just need to show that balance isn’t the only factor

I did, you just can’t understand the argument. It’s ironclad. Masters league selects for players who prioritize win-rate. That’s the definition of masters league. If win-rate isn’t prioritized, the player isn’t a part of master’s league. For your theory to be true, you’d need players from masters league to pick en masse based on some factor other than win-rate when the definition of masters league prevents that from happening. You’ve created a concept that cannot exist in real life, and you refuse to define what factor is driving the trend. If players don’t pick based on race, what are they picking based on, and why does it coincide with race. You’ve solved 1 small problem and introduced 3 enormous problems.

The obvious answer is that masters league selects players who prioritize win-rate, and that includes players who prioritize selecting the strongest race.

So to reiterate:

1. What variable is driving the trend.
2. Why does this variable coincide with race if not caused by race.
3. Why does masters league select for this variable and not for win-rate.

You’d need a complex & convoluted theory to explain each of these points when a simple answer is standing right in front of you: players who select protoss are more likely to be in masters league because protoss increases win-rate which is what master’s league selects for.

[Eliwan-1354]

@Adventurer-16000863

If win-rate isn’t prioritized, the player isn’t a part of master’s league.

(stares at the huge quantity of non-Protoss players in both masters and GM)

… ?

If your suppositions are true, this shouldn’t be the case. Therefore, your suppositions are not completely true.

Therefore, balance isn’t the only factor.

[Adventurer-16000863]

@Eliwan-1354

(stares at the huge quantity of non-Protoss players in both masters and GM)

… ?

If your suppositions are true, this shouldn’t be the case. Therefore, your suppositions are not completely true

This argument works in the non binary cases as well aka for the average case. If a mystery variables causes zerg to be 5% less common in Masters, what is this variable (if not race), why does it coincide with race (if not caused by race), and why does masters league select for it.

The “we don’t know what causes it, even though it perfectly coincides with a certain explanation we hate, but we definitely know it isn’t that one thing that we hate because we say so” isn’t a good argument. Tell me what variable it is. Tell me why that variable coincides with Protoss and not the other races. Tell me why Masters league selects for it when Masters league is supposed to select for winrate.

[VichelChaos-2623]

To be honest, its true that I haven’t played recently (couple of years, have been super strapped for time), and most of what I watch are tournament finals where Serral does his stuff and makes defending everything while having a strong economy seem easy. I’ll hop onto ladder for a while when I get some time.

I might be talking out of my butt because of that, but I don’t remember protoss economy outpacing zerg to such extent. In my experience and a lot of what I saw, protoss typically were ahead on economy until 3 base injects kicked in and then things started to shift. Are there new tactics of the protoss using that period to get an advantage by leveraging the economic lead into keeping the zerg pinned down with blink stalkers?

[Eliwan-1354]

@Adventurer-16000863

why does masters league select for it.

It does not, simply.

@Adventurer-16000863

If a mystery variables causes zerg to be 5% less common in Masters, what is this variable (if not race), why does it coincide with race (if not caused by race),

Well, given the last; my assumption would just be player preference? The probability of a random Zerg from the entire pool on the ladder being exceptionally good at the videogame is not the same as if we do that with Protoss or Terran; and it isn’t necessarily anything wrong or right with those races that makes that happen.

If I didn’t already know, I would be surprised if the race with the highest population-relative number of exceptional members isn’t Terran; since that’s the one where it seems most likely that skills from other RTS would transfer over.

There is no other reasonable answer to your stated question, and I find it a certainty that you will complain about this response.

@VichelChaos-2623

but I don’t remember protoss economy outpacing zerg to such extent.

You remember correctly.

[Adventurer-16000863]

@Eliwan-1354

Well, given the last; my assumption would just be player preference?

Player preference is required to align with racial strength because players who don’t prioritize win-rate are kicked out of masters.

Let’s set up a scenario where masters is perfectly balanced 33/33/33 & the game is balanced. Let’s say zerg is hit with a nerf that gives it a win-rate of 1%. Anyone who picks zerg won’t be in masters league because to get into masters league requires a high win-rate and you can’t have a high win-rate by picking zerg if zerg has a 1% win-rate. So anyone who picks zerg is demoted and they learn that really fast so they switch race to protoss or terran.

In the real world, there is a spectrum of players with a spectrum of winrates in masters, but the average is 55% wins and 45% losses. The balance scenario isn’t that severe, it’s probably in the ball park of 58%. That means a player who is average in masters would go from a 55% win-rate to a 52% win-rate if he doesn’t select race for racial strength. The ones who select wrong will be demoted. Players preferences can’t drive this trend unless the preference is based on racial strength.

@Eliwan-1354

There is no other reasonable answer to your stated question

There is no way to answer those 3 questions because it creates a definition that can’t exist. If players in masters league do not select race based on winrate then they won’t be in masters league because masters league selects players for their winrate. It is impossible for preferences to drive this trend unless those preferences are that players prefer the stronger race.

“We don’t know why people prefer protoss, even though masters requires them to prefer protoss for the strength, because we say so” isn’t a good argument. List why players prefer protoss. Explain why it has nothing to do with balance. Explain why masters league selects for these players with no regard for their winrate.

There is no justification for why Master’s league prefers these players over those who prioritize win-rate. There is no explanation for what the player’s motivations are, except that you deny that racial strength is the motivation. There is no substance to your argument except “maybe there is a super secret hidden variable that magically explains everything without the need for racial strength, oh and by the way I refuse to explain what that variable is or how it works”.

[Adventurer-16000863]

@VichelChaos-2623

To be honest, its true that I haven’t played recently (couple of years, have been super strapped for time), and most of what I watch are tournament finals where Serral does his stuff and makes defending everything while having a strong economy seem easy. I’ll hop onto ladder for a while when I get some time.

I might be talking out of my butt because of that, but I don’t remember protoss economy outpacing zerg to such extent. In my experience and a lot of what I saw, protoss typically were ahead on economy until 3 base injects kicked in and then things started to shift. Are there new tactics of the protoss using that period to get an advantage by leveraging the economic lead into keeping the zerg pinned down with blink stalkers

Let’s pick a random pro game to see. Dark v Hero game 1 of GSL:

P v Z
0:54 - 17 vs 14
1:40 - 10 vs 18
3:06 - 29 vs 26
3:52 - 37 to 28
5:04 - 45 to 46 (zerg takes the lead for a brief moment)
5:55 - 58 to 54
7:57 - 72 to 70.
9:14 - 84 to 78.
10:30 - 71 to 77 (protoss takes economic damage due to bad positioning, zerg leads in eco from here on out).

So as you can see, Protoss have the stronger economy. Don’t take my word for it - use your own eyes:

https://www.youtube.com/watch?v=3n7SIRz9k34&t=9808s

Chronoboost is busted. Anyone with eyes can see it. Protoss eco is stronger than zerg’s, and their army is more powerful for the same cost, which means protoss is simply a better race.