PvT isn’t 55%+, and you should stop calling people troll if they do not think like you, not only the average data and tournament data is showing protoss is not broken, you can see the overall tournament results and upper brackets. All you have to say it’s ‘‘muh 40% of P at ladder’’, and that’s all, cry if you don’t like P dominating at GM, but do not try to create the dumb narrative that P is dominating everything because if P was so good we should be seeing more P at the upper bracket tournaments and them even winning, more protoss getting results at every tournament, we should see rising noname players and some known players rising into the top.
http://aligulac.com/misc/balance/ , during all the time aligulac is tracking the winrates, just in two months P had over 55% and T had over 55% 7 times, and the more important thing, RIGHT NOW, it’s almost around 50%,it’s basically perfect, you can’t improve that.
Yes.
Yes.
No. Premier tournaments are still played by individuals. Being a world champion isn’t exacly balance indicative. Btw interesting things (from statistical point of view) are happening in EU tournaments. I have written more about it in chapter 2.2:
Follow the blue line, that’s PvT, over 45% the matchup is on the 5% range, under 45% means T is having over 55%wr,each dot is one month, count the dots and you will see more times P being under the 45% than T.
It’s true that sometimes p showed advantage in the numbers and they were left, but because those numbers were on the 5% range.Also is worth mentioning how P% work, if you see Z matchups, they usually follow a more steady trend, specially zvt, but P matchups involves a lot of spikes, some month is at 43% and the next one can be at 50%, some T were claiming insane WR and P being extremely broken and that was just a spike, then as I predicted during those times, the WR went down without needing any patch and we are at 50% now, and they basically were asking for nerf everything protoss has.
" just in two months P had over 55% and T had over 55% 7 times, and the more important thing, RIGHT NOW , it’s almost around 50%,it’s basically perfect, you can’t improve that."
“each dot is one month”
For a while I thought that I haven’t pushed aligulac’s possibilities to limits, but apparently I used it correctly.
You have an error somewhere, there can’t be 8 data points in a span of 2 months, which has only 2 datapoints. Hence my questions. Well, we all keep making mistakes.
Using data across 10 years may nicely show how Blizzard improved at making balanced game and prove that it is in great state compared to the past, however should not be taken into account while considering current balance.
But I believe we both agree on that, just like with the rest.
Oh, that’s it, you misunderstood the phrase, it was just two months in 10 years of data (the total of aligulac timeline), not 7 points of data in just two months. Then the current data can be seen too, its around 2,5 periods of data per month, current period (well, last since the current one started today) is
Start
Oct. 8, 2020
End
Oct. 21, 2020
Games
4665
PvT
616–602 (50.57%)
PvZ
548–597 (47.86%)
TvZ
387–413 (48.38%)
Mirrors
672 PvP, 378 TvT, 371 ZvZ
data from a healthy amount of games and showing a near perfect balance,everything in the 2% range.
Why do you think games from different periods should receive different weight?
Again, that is obviously false. If you sum all the wins and losses in the past 2 years, Protoss has a positive win-rate vs terran. Furthermore, Protoss is dominating GML more than any other race has ever dominated GML, meaning their current imbalance is larger than any imbalance that has happened before. Additionally, they have won the most tournaments in 2020.
It is not a matter of opinion. Protoss is overpowered as a matter of fact.
Why do you place increased weight on some games but not on others? By your logic it’s fine if I select 1 game where a Protoss wins, put 100% of the weight on that, and declare Protoss OP.
Also patently false. By dividing into discrete periods, games are, by definition, not weighted the same. The math is very simple. If you have 1 game in a “period” and a 1000 games in another “period” then 1 game has 1000x the weight of games in the other period. It’s simply a form of cherry picking.
Every game should have the same amount of weight, and that requires a time series linear regression.
This is by far one of the best thing that has come out of the Blizzard post regarding going into maintenance mode and only making balance changes when warranted, is when he talks about how it’s all up to you the players to win or lose.
This forum is gonna be even worst when it comes to balance now that they’ve taking a step back from the game. Cant wait to see people crying about balance but still play the game religiously. And I’ma sit back, eat some poprlcorn and watch the ensuing salty tears come flowing out like a river.
Why do you think games from different periods should receive different weight?
We live in reality, not in the past. Past 1-36 months may be reasonable to take into consideration, but how far do you want to look into the past? Early LotV? HotS, WoL? I am sorry to disappoint you, but such data will not support your point of view. Patches changes, balance changes at some point it’s no use looking behind. If any race was weak at any point in the history, it is not an argument to give it 60-70% win-ratio now.
I do have a very special place in my heart for MaNa vs ForGG from Katowice 2015, but I do not believe old matches are relevant to current balance (with an exception for what not to do).
Again, that is obviously false.
That’s my genuine opinion. I did not create it on empty words, but supported it by evidence and data. If you show that there is something more than “Protoss OP” behind your words, perhaps you will convince me. If it’s obviously false, feel free to enlighten me. Protoss has a positive win-rate vs terran.
When you take probability into account, one of races must have positive win-rate.
Win-ratio is usually under 52% in favour for Protoss. So every 50 games as a Terran, you will win only 24, and Protoss shall win 26 out of pure balance. Would you notice that after playing 50 PvTs? I would not. That seem to be fair, personally I accept 5% difference, and even if win-ratio was 40% for Protoss in both match-ups, I would not stop playing the race. I see no problem in current state. Protoss is dominating GML more than any other race has ever dominated GML, meaning their current imbalance is larger than any imbalance that has happened before.
That’s funny one. Where to start with?
Not Protoss, but top players. Not ever, but in Jule 2020, (funny how situation was back in 2011), not dominate, but win a few more games.
Meanings of dominate:
Being the largest (true), most important (false, all races), most noticable (false - Zerg), the best (possible, arguable, not by large margin, though), having control (false). (6)
If you want to use a word to describe Protoss appearance in GM league, the best world is “largest” group of players, not dominating, there are no reason to use such a strong word. (also it has emotional meaning, what means the word itself can, and often is, used for manipulation emotions).
Protoss has largest amount of players in GM, around 50% EU, that’s a fact.
Using this pure number as game balance indicative is a mistake.
Detailed explanation in chapter 2.1:
Additionally, they have won the most tournaments in 2020.
Tournament wins are not indicative.
Most means more than other races, so above 34%? Did you actually count that?
Detailed explanation in chapter 2.2 of link above.
It is not a matter of opinion. Protoss is overpowered as a matter of fact.
Unless proven by a solid research, it’s a subjective opinion. Pretty often not supported by anything. From info I gathered, nobody attempted recently to properly prove the thesis. There is also no point in, unless you are a member of Activision balance team. They get paid for such research and you not. It is time consuming, and balance evolves quickly, so such stuff becomes obsolete very fast.
Also Aligulac is not exacly correct, however it is 2nd best balance-related source I have found so far.
Also patently false. By dividing into discrete periods, games are, by definition, not weighted the same. The math is very simple. If you have 1 game in a “period” and a 1000 games in another “period” then 1 game has 1000x the weight of games in the other period. It’s simply a form of cherry picking.
True, but that is not the case, what indicates you have not checked the source properly. And that is okay. We can be lazy, just be more careful in the future.
Every game should have the same amount of weight
Absolute nonse unless you put reasonable time limit. Put away WoL and HotS of the equation. Also games before November 2018 should not be taken fully into account, for big changes added to the game during that period. Old data is good for avoiding mistakes and comparision, but is not relevant to current balance.
“Today is cold, global warming is false.”
“I am just using the most recent data.”
Incorrectly, of course. If you analyse the data properly and show your thought process, your opinions will be taken seriously. Cherry picking is not the only error you can do while analysing data.
At this point I would suggest everyone (including me) to learn more about truth, false, opinions and process of making them. It is something most education systems around the globe do not include in teaching procees and many people are lacking such skills.
Too lazy to reference properly, so I will leave it just here like that:
6. htt ps://dictionary.cambridge.org/dictionary/english/dominating