Fix MMR range for Toss and Zerg

General Discussion
671

Yes, they can; because you’re not looking at it the same way as - at least I am, I assume for Sentry as well.

Let’s use an extreme example from a different game: In League of Legends, there are so many characters and so few highest-rank players that multiple characters have single digit games played in a season at that rank, and as a result, not infrequently end up with 100% win rates.

Does that mean absolutely anything at all, balance-wise? By itself, not a thing. Maybe none of those under-a-thousand players like the character, or the character is too similar to some other character whose utility is more useful in higher ranks. Maybe the character is simply niche, so only gets picked when it would do very well.

However, League, unlike StarCraft 2, has gone on record to say their balance approach rarely, if ever, should boil to your skills are lacking - SC2’s older development teams have explicitly said that ie. 6-8 pools are not a problem because its success rate in Gold-and-up is horrible, therefore anyone in Bronze or Silver simply needs to learn to play a better game.

And this isn’t true either, wow! It had been listed multiple times. Like… ???

3 Likes
672

Are we reading the same thread?

Because there have been multiple instances of information and long form explanation offered by Sentry and I don’t understand how you could miss it if you’re using your eyeballs.

This isn’t what that fallacy means.

Have you considered the possibility that there is more than one person who thinks the basic premise of this thread and of the argument that Bourne is making are simply that stupid?

Because that’s reality. That’s how that works. There are, in fact, multiple people behind the screen.

If you would like, I’m happy to perform a fresh tackling of any evidence-backed claim you would like to make, and use evidence to demonstrate why it doesn’t make sense; as long as you actually read it and then learn from it.

673

@OP:

IMO, watching the pro scene, Zerg and Terran have a similar skill cap. Protoss has a lower skill cap. Thus Protoss dominates the ladder, because lower skilled players can beat higher skilled players. This is balanced by the fact that Protoss is a bit weaker at the highest pro level.

674

First i dont need to be able to predict what they think. Secondly i dont need to do it for tens of thousands of people.

I have Stated multiple Times that its Just a statistical development. Meaning Not ALL people do xyz because of abc. Its enough If SOME people do it and thus revealing Statistical imbalances.

Its actually crazy that you deny Data science as a whole. You want to study people and people behaviour to predict what they gonna buy/do/think. This is Not a new or Abstract concept. Im sorry for you If you missed that, but that isnt my Problem.

And i have presented you the Proof of concept for Marketing or Design. So i dont know why you would assume i have No reason to believe what i say. Because gues what? 37% of the Players Play terran and only 27% Play zerg. Is this Just “random”? I have listed all the sus “coincidenes” multiple Times by now.

Its actually the Other way around: YOU have No reason to believe what you believe. And you still Need Proof or any Argument that you can ignore the Player preferences (37 % terran and 27% Picking zerg) and still proceed with your Claim that lower AVG mmr means Harder.

675

The evidence for which being (paraphrasing): the statistics cannot be treated the same because the faction selection is not an immutable attribute.

This applies to all levels of skill. If you are willing to argue that several tens of thousands of players are losing more frequently because the bad players play Terran, you have to be willing to say that (on the other side of the skill spectrum) 15 or 20 players are unable to win Premier Tournaments because “the top level players just don’t pick Protoss.”

676

This isn’t really a controversial take, honestly. Artosis just made a video about this not long ago.

youtube. com/watch?v=fVogp9bicB4

1 Like
677

This is only one part of the reasoning, but yes, and it is also the centerpiece of that logic.

Yes, but no, it does not equally apply to all levels of skill. We actually can observe this from the ladder data itself: Roughly speaking, the racial proportions of players at a rank becomes more equal the higher up you go on the ladder.

The more skillful or knowledgeable a player is, the more willing they are to experiment and try new things, and part of this is that they will have a stronger foundation, but part of this is they will be more able to identify when it’s “Ah, I suck” versus “Ah, X is garbage” without getting frustrated with themselves or the game.

You can observe this with not just SC2, but basically any real time game. I’ve watched it happen with players of every type of action game, platformer or combat centric. There’s this weird trough in the middle where people are comfortable to succeed but not enough to experiment, but people failing will try new things all the time and people once comfortable will try new things to see if they like them more than what they have been doing.

Yes, this is a reasonable dichotomy to bring up.

However, please note that no response to this - in any way - invalidates anything that’s been said.

Even if Sentry disbelieves this particular quote (which I doubt), it does not mean anything. Your entire argument here is building to an ad hominem; that Sentry “cannot be rational” because they hold two beliefs that are diametrically opposed and seem to be contradictory.

While I do agree with your argument - Protoss success in tournaments is markedly low, and it also may be because the samples are biased, and it is actually statistically more likely for the samples to be biased because we’re looking at such a small population - as a result of the former, I don’t see why it has any relevance besides being a word substitution on the argument. This is in contrast to the last time - near the head of the thread - when such a word substitution based argument happened.

We’re not talking about “top level protoss doesn’t win tournaments”; we’re talking about “Terran average MMR is lower”. We’re talking about “Terran is the weakest race”, “Terran has the highest skill floor”, “Terran requires the most skill to succeed”.

678

How are they losing more ? I dare you to say: lower AVG mmr.

“Because Players who are Casual/new/unfamiliar with the Game Pick terran more often”

I fixed it for you. You arent Bad because you picked terran and Picking terran doesnt make you Bad. Also a Casual Player can Turn into a competetive one, a new Player can Turn into an experienced one. A Player that is unfamiliar with the Game can Turn into being familiar with the Game. The Starting Point is important.

First: its Not the Point of the topic.
Second: No, you cannot outright compare those 2. There are many very plausible reasons why new Players will Pick terran but there is no plausible reason for pro Players to Not Pick protoss.

That being Said ofc it can be because of randomness. But the Thing is: i can directly Change small Things in sc2 and have a massive Impact on the Tournament Race distribution. But you have to Change Something massively in Order to have terran Not being the starter Race.

I believe protoss performance in Tournament is because of many reasons. Ofc one reason is because i dont think there are that many Players for protoss on the Same Level as serral reynor and Maru but that alone isnt it ofc. And also its pretty important to Note what pros are saying. I mean i think we all know there are many many reasons for protoss Being sh!t in Tournies and you actually agree but you Just try to Argue for the Sake of it and try to make the 2 Statements Look the Same. But obviously they are Not.

1 Like
679

40% at GM is hardly “equal”

That’s the only way to get lower MMR, Simple Jack.

680

But Plat to Masters IS pretty much equal. And end of Last Season GM was Equal between t and p

So you mean Like on average every terran needs to win much more only because of the Low League distribution ? You realize thats fallacy of composition? :slight_smile:

Ok, See: you have 2 very different behavoirs: in lower League terran is overrepresented and in Higher League terran is equally represented. To Claim A should influence B is pretty much illogical. High Level terrans dont struggle at all, so why should Low Level terrans struggle because of difficulty? Doesnt make Sense. Again: there are dozens of reasons why the avg mmr distribution is how it is. ONE possible influence is difficulty. Yeah. Thats its. Its a POSSIBILITY.

681

Cool, so if we delete over half the ladder, things look pretty even.

Good for us pick a single datum and draw conclusions based on that. Brilliant science work.

682

Can i end this thread ?

TERRAN BIO IS THE HARDEST STYLE TO PLAY IN ALL OF SC2.

Mech is relatively easy in terms of apm and multitasking and can be played by people with mediocre mechanics.

So to summarize terran is the hardest race if you choose bio.

683

Its even all around. As Establishment by Splitting Them into upper and lower mmr. Only that in Low Level Play terran is more popular.

On average you will get exactly the Same mmr If you are an unskilled or a skilled Player.

You mean Like you do it with gm right now or how you use Activity stat?

GM is Not constantly 40% protoss

1 Like
684

Fine, 27-29% of the population bounces regularly between 35 and 45% of the GM for the last 5 years clustering more towards the middle. You happy? It’s called an “average.”

685

IDKY, but Artosis actually said it’s the opposite. I guess because you have to defend without Marines? Anyway, I can see how It’s easier in TVZ, but not in TVT (and it’s all but impossible in TVP).

686

No. Prove it please :slight_smile:

Still doesnt Change the fact that Generally speaking the Game is more even in Terms of Race distribution the Higher the Player Skill grows. Maybe more or less law of average ? You Come from an imbalanced start to a balanced Outcome.

How would you describe this ? I mean is this even possible If 1 Race is sooooo muuuuuuuuuuuuuch haaaaaaarder than the Other races ?

687

Yeah, so… Except for GM? So if there’s 7 leagues and if I just remove 4 of them, BAM! Things look even! And I’m definitely not cherry picking.

688

Except 6 of Them are looking Equal. Its Just that the Race representation evens Out. But the avg mmr gain is Equal for a skilled and Low skilled Player. The Starting Point is Just uneven which is very logical (for everyone that is Not Bourne, ofc).

GM as they only outlier where its fluctuates heavily and t and p Change positions

689

The masters league bug is purposefully maintained so that the number of master league terrans appears higher.

1 Like
690

How. How are they even if their race representations aren’t?

The only fluctuation Protoss has in GM is between being massively overrepresented (27% of population => 35% of GM) and overwhelmingly overrepresented (27% =>45% of GM).