The StarCraft 2 GOAT - An in-depth analysis - Page 9

On July 27 2024 19:29 lokol4890 wrote:


Alright, if we're just literally changing the meaning of words to fit our points, there is not much I can say. Dictionary defs pretty clearly equate dominate with being the best. No idea how you guys feel this confident reshaping the meaning of that word. This feels so weird. And funnily enough premobeats himself doubled down on dominating as synonym for best, e.g., noting serral's winrates against koreans since 2018. So this is the position you guys are in: you're at the same time arguing dominating somehow doesn't mean being the best in order to defend criticism of premobeats' analysis while premobeats's own words undercut your position. Wild

On July 30 2024 02:49 Charoisaur wrote:

Yeah sure, with enough mental gymnastics you can pretend like anything is a Serral nerf.

Here's my attempt doing the same as you, just in Maru's favor.

1. "For the efficiency score I simply divided the Tournament score by the sum of years a player was placed 4th or above."
This doesn't include Serral's weaker years (2013-2017) but it does include Maru's weaker years (2012-2017) so it's a huge nerf for Maru.
2. I compared the players aligulac rating and tournament winrate in their peak years, however did not award a bonus for their longevity which means I did not give them a bonus for achieving a high aligulac rating or finishing high in tournaments in more years than another players. This is a huge nerf to Maru as he has been a championship contender for 5 more years than Serral
3. despite the 2013-2018 era being way more competitive and the number of championship contenders and potential dark horses in tournaments being multitudes higher than in 2024, I gave the era only a 50% buff which means winning 2 tournaments in 2015 is the same as winning 3 tournaments in 2024. I gave such a low multiplier because the relative skill of players that are still active didn't decrease, seeing as Innovation continued to have the same winrates against fellow players like Zest, soO, Classic etc. This conclusion doesn't make sense because those benchmark players are affected by the same circumstances as he is (older age, responsibilities, approaching military). Therefore, this low multiplier is another nerf for Maru.
4. Another (only) 50% buff for pre-2018 in category 7
5. Another (only) 50% buff for pre-2018 in category 9
6. Another (only) 50% buff for pre-2018 in category 11
7. I didn't consider Proleague because that was difficult to do despite it being the most important tournaments for koreans back then with them being prohibited from playing in many overseas tournaments to focus on Proleague. As Maru was a phenomenal Proleague player this is again a huge nerf for Maru.
8. I didn't adjust for racial balance because that's difficult to do. However, this is clearly a nerf to Maru as Zerg was consistently the strongest race for most of Serral's career so including racial balance would have elavated Maru
9. The idea to only look at vs korean winrates makes sense. Technically Clem and Reynor should be included as they are on the same level as top koreans, but because that's difficult to do in aligulac I didn't do it. However, that's another nerf for Maru as he has better winrates vs both than Serral.
10. I put WESG in tier 4 because it didn't have enough top players in the ro8/ro16. Yes, the korean qualifiers included all the best players in the world but I don't feel it's the same as playing them in the ro8/ro16. Therefore, this is another nerf to Maru.
11. Average place achieved: I singled out the prime years because Serral's 2013-2017 would have decreased his score very much. Looking at all years would have been favorable for Maru, therefore it's another nerf for Maru.


This is how someone would have written the section who was as Maru biased as you are Serral biased. I mean, I don't even think your statistical approach was bad at all, I appreciate it, and understand that measuring racial balance or Proleague would've been really difficult. But you don't have to pretend like you're somehow nerfing Serral in every category when you could find just as many arguments for Maru nerfs.

On July 30 2024 05:00 Poopi wrote:
Yeah the text is funnily absurd and imo weakens the overall work with the constant « trying » to prove that the conclusion is 100% logic and factual.

On July 30 2024 04:59 PremoBeats wrote:


1. You have to decide: Either Maru's earlier years were weak and get abolished completely, or they stay and all of them get counted and you are able to praise his long career. You can't have the best of both worlds. If we discount these, we also have to discount them in the tournament score. Then Serral will be on top again, even if we include team results.
2. Why would I give a bonus for longevity in this metric, when longevity is obviously taken care of in HoF? Finishing high in tournaments is measured by average rank.
And achieving a high Aligulac rating would benefit Serral much more. I don't really understand your point here.
3, 4, 5, 6. The win rates of them didn't drop versus the rest of Koreans either. Unless you assume that EVERYONE suddenly got worse, this argument doesn't make sense. You do realize that the 50% weren't the only multiplier that addressed tournaments from different eras, right? But again.. I am open to revise this when my analysis of era comparison is finished.
7. So any replies to the difficulties on how to factor in Proleague/Team results? Or how to actually do it? This still has not been addressed.
8. How would you account for that?
9. This isn't even true. Serral is better versus Reynor. But why not include MaxPax while we are at it? Cause Maru is much worse against him? You do realize how tiny the amount of matches is in the big scope? Even if one player is 5% better than another versus the same player.
Also one has to take into account practicability and effect.
You know what will happen if we include ever more non-Koreans and not only the ones where Maru supposedly fares better, right? I hope that settles this point.
10. You want me to put it in Category 3? The equivalent of Miz's list? Then Maru gains 0,45 points... I probably don't have to mention that Serral gains much more with a more befitting 1st to 2nd place ratio.
11. Erm. No. This would have affected Maru a lot more, as he played more tournaments in that time frame, as Serral still attended school (look at the screenshot under 6.1 to get an impression).

Many of your takes are either untrue, illogical or will nerf Maru in another way. They cannot remotely be compared to the Serral nerfs I mentioned as far as I am concerned. But I am more than happy to focus on Proleague and imbalance. The more we can agree on, the better the next list will be.
Or are there more things that you think I did that were "unfair" to Koreans?

On July 30 2024 12:37 sc2turtlepants wrote:
While I admire your willingness to continue running in the same circles, rwala straight up said that he prefers subjective metrics (ie he's redefined 'GOAT' as 'My favorite player'), Chariosaur's comments have run in all the same veins, and Poopi won't even engage with the content because then he'd have to state a position that could be refuted when he'd rather just chime in every few pages with a 2-line quip that allows him to maintain an air of smugness. They aren't here to discuss metrics, they're here to complain about them.

On July 30 2024 14:32 Starcloud wrote:


I think you summed up this merry go round perfectly.

I must admit I appreciate the fact that Charoisaur has actually changed his opinion in last year and contributes to the much more to these discussions since he can see both sides now more clearly. Thumbs up for that !

This following chapter goes to the OP::

Overall I think this discussion is starting to be at a stand still, and the same things are just being rephrased over and over again. Imo the "truth", or how do I say, the balance between two sides stands somewhere in the middle. Like Charoisaur said, the statistics part is very well done and some of the conclusions are correct. However it gets whole lot complicated when you start to adjust those statistics by one way or another. Then it becomes more of an opinion instead of pure numbers. And however "correct" that opinion might be, its always up for debate. And in the end, I think you shouldnt focus too much on trying to change what other people think, but be proud of the work you have done. Thank you for an interesting and very detailed version of the debate !

On July 31 2024 16:17 ejozl wrote:
A big problem is that of the ace match, this is clearly a harder match. How would you tackle the problem of a team like CJ with a clear best player, that of herO and a stacked team, where even a player like Flash, or Life might not even get fielded. There is also the thing of sniping a specific player. Rogue might be a player who was used as a sacrificial piece, or sniper versus the other teams best player, this would clearly hamper his rating.

On July 31 2024 20:31 Charoisaur wrote:

Yeah I think that's a pretty good way of looking at it.
According to this you'd think winning a tournament is worth the same as getting 2nd twice?

On July 31 2024 20:31 Charoisaur wrote:

Yeah I think that's a pretty good way of looking at it.
According to this you'd think winning a tournament is worth the same as getting 2nd twice?