[WCG] 2010 Korea Round of 16 Day 2 - Page 26

Post a Reply

Prev 1 24 25 26 All

Letmelose

Korea (South)3227 Posts

August 13 2010 15:37 GMT

#501

On August 14 2010 00:18 revy wrote:

Show nested quote +

On August 14 2010 00:08 Letmelose wrote:

On August 13 2010 23:14 revy wrote:

On August 13 2010 04:47 Letmelose wrote:

On August 13 2010 00:40 huameng wrote:

On August 12 2010 17:37 Letmelose wrote:

On August 12 2010 17:30 TwoToneTerran wrote:

On August 12 2010 17:29 Letmelose wrote:

On August 12 2010 17:16 TwoToneTerran wrote:

On August 12 2010 17:08 Letmelose wrote:
[quote]

Take Jaedong away, and the entire statistics will be changed for his race than any other player out there in the modern era. That was the point I was trying to make. Don't pretend otherwise, terrans have always been the dominating race in Starcraft ever since the 1.08 patch, and is least affected even if you take away the player of the highest score within the race. Jaedong just happens to be awesome enough to keep the zerg race even statistically.

Keep in mind that OSL has an Ro36 format instead of the Ro32 format of the MSL when you make the statistic.

holy crap you are a broken record. You just stated that Bisu had the same impact in the overall scheme of things with my little rating system. Why are you saying I'm pretending Jaedong isn't great? You keep shoving words down my throat in defense of Jaedong, I've never said anything bad about Jaedong. You are really mad at something and it shouldn't be me, so I suggest you leave me be.

The statistic will scale fine, by the way. I start with the group stages which are the tournament proper. OSL pre Ro16 is a big changing mess each season anyway, especially with the match fixing scandal and ACE recruitment and blaaaah.

Why can't you deal with your own stastics being thrown at your face? Zergs have not been dominating the post-Savior era, and if it appears to be so, it's the power of Jaedong. What part of the total number of points being relied upon one player more so than any other race don't you get?

Haha you're gonna fucking love the OSL stats

Haven't seen it, but I guess Jaedong's great success in it, coupled with the lack of success from protoss have made your argument that zergs have wiped the floor with everybody in recent years more convincing?

It's not just the individual leagues I'm talking about here. Check the Proleague stastics, check out the number of terran players overall, be it top ranked Kespa players, or simply in the Proleague rosters. Wipe that. Just take all the televised games in recent years and see which race has won the most. I haven't checked any of these, but I'm pretty sure most of it will have terran in front of the other two races.

Haha why do you even post? All of your posts just say "terran is imba but because of JAEDONG MY HERO zerg looks okay. I refuse to back this up in any way but STOP DENYING IT!" Do you think this is a normal way to try to prove a point?

I think sc2 whining has bled into bw -_-

Time to back my theory up.

1) Since the start of the 2007 season (not the year 2007), YGOSU records the following:

TvT 1883 matches
ZvZ 1777 matches
PvP 1320 matches

TvZ 2632 : 2311 in terran’s favour
ZvP 2325 : 1915 in zerg’s favour
PvT 2313 : 2094 in protoss’s favour

Terran is 4726 : 4405 overall against other races
Zerg is 4636 : 4547 overall against other races
Protoss is 4228 : 4419 overall against other races

Terran’s have won the most, and lost the least in non-mirror match-ups. Terran versus terran also happens to be the most played mirror in the game.

2) This is how the record is changed if we count Kespa officiated matches only:

TvT 1038 matches
ZvZ 782 matches
PvP 633 matches

TvZ 892 : 753 in terran’s favour
ZvP 694 : 617 in zerg’s favour
PvT 802 : 737 in protoss’s favour

Terran is 1629 : 1555 overall against other races
Zerg is 1447 : 1509 overall against other races
Protoss is 1419 : 1431 overall against other races

Terran is the only race that happens to have a positive overall record in non-mirror match-ups. Terran versus terran remains the most played mirror in the game.

3) This is how the record is changed if we count Proleague matches only:

TvT 742 matches
ZvZ 601 matches
PvP 490 matches

TvZ 486 : 399 in terran’s favour
ZvP 424 : 375 in zerg’s favour
PvT 462 : 429 in protoss’s favour

Terran is 915 : 828 overall against other races
Zerg is 823 : 861 overall against other races
Protoss is 837 : 853 overall against other races

Terran has again won the most, and lost the least in non-mirror match-ups. Terran versus terran remains the most played mirror.

4) This is how the record is changed if we count individual league matches only:

TvT 208 matches
ZvZ 136 matches
PvP 129 matches

TvZ 250 : 249 in zerg’s favour
ZvP 185 : 148 in zerg’s favour
PvT 244 : 214 in protoss’s favour

Terran is 463 : 494 overall against other races
Zerg is 435 : 397 overall against other races
Protoss is 392 : 399 overall against other races

Zerg is the only race that happens to have a positive overall record in non-mirror match-ups. Terran versus terran remains the most played mirror.

Zerg is the only race that happens to have a positive record in both their match-ups. I guess this is where the consensus that zergs have dominated the other races in recent years stems from.

Considering how much individual league statistics are influenced due to the nature of brackets, the relatively small amount of gaming pool and greater number of games played at higher stages (Bo3, Bo5), it’s not a far fetched thing to say that the might of a few players do change the overall appearance of things in my opinion. I’m just claiming the obvious here, nothing against the common consensus of any well reasoned follower of the progaming scene. How my opinion that terrans are still the most dominating race in the progaming scene, and that the results of individual leagues are swayed greatly by the most successful gamer of our era is somehow so far stretched that it borders on insane ramblings of a blind fanboy is beyond me.

Are you kidding me? Divide your own data here:

TvZ : 53.25% Terran Favor
ZvP : 54.83% Zerg Favor
PvT : 52.48% Protoss Favor

Combined: Favor in good + (1-Favor) in bad
T : 0.5325+(1-0.5248) = 1.0077
Z : 0.5483+(1-0.5325) = 1.0158
P : 0.5248+(1-0.5483) = 0.9765

A) TvZ isn't the most imba matchup by this data
B) We're talking 2.48-4.83% imbalance in matchups, is this so insurmountable as to complain this much?
C) Using this rough combined metric (seems plausible to me) things are god damn balanced, but if you want to split hairs Z has it better than T by some insignificant margin.

Do not take away from this: Z is imba
Do take from this: Nobody is imba, Toss needs a little love, STFU with Terran imba garbage

Edit: This is the kespa+nonkespa data. I'm at work so I cant run the other numbers now.

What part of terran players participating in more televised matches, having more victories and having less defeats despite the greater number of games in non-mirror match-ups don't you get? Why do you think I bothered to list the number of mirror match-ups? Do you think it's a coincidence that terran versus terran happens to be the most played mirror in the scene at every level imaginable? Is it a simple flip of the coin that terran players have represented their team far more than the other two races?

The greater the pool of games we take, the more dominant the terran presence becomes. It's not just about the winning ratio, although it's interesting that you decided to present it as an average of the winning ratio between the two non-mirror match-ups, when I thought a simple calculation of the actual number of games won and games lost in non-mirror match-ups would have been sufficient. There are plenty of other factors to consider, a few of which I have mentioned above.

I find myself repeating the same thing ad infinitum here, and I hope this will be the last time I mention this.

Terrans have pretty much always have been the race to be in the progaming scene. The difference is not much, but it is substantial enough for most people to have noticed. It didn't matter which statistic you took, be it the number of individual leagues won, number of televised matches won, the number of top ranked players, the overall winning ratio in televised matches etc. However, terran superiority over the other two races (however slight it is, it was there) has become less significant in recent years, especially in the individual league with the success of some select few individuals, with Jaedong, being by far the most noticable.

It does not change the fact, that terran players today, are still the most played, most trusted by the coaches, most successful in terms of the number of games won in the scene. To claim that zerg as a whole race, have dominated the scene in recent years seemed to me the arguments of someone who got a little too overwhelmed by the number of individual league trophies taken by zerg in recent years. Surely, it had a lot to do with the individual success of the most successful player of our era. I don't see what was so sacrilegious about my argument.

Perhaps there are more Terran players than others? Maybe that's why there are more mirror matchups. Infact that I remember there ARE more terrans than the other two races, that's why mirror matchups are pointless to look at. Why are there more terrans? Probably because Boxer, Oov, and Nada. Terran used to be terribly op or at least all the good players were terran for some reason, it isn't anymore.

I didn't save the data but % for kespa only support your claim, it was something like 52% for PvT and ZvP and 54% for TvZ. Again at work. The fact of the matter is a difference as small as 2-4% in a single matchup is so incredibly small.

The number of registered progamers are roughly equal if I remember right. What's your source for the number of progamers for each race? Care to back it up? I seem to recall YGOSU, although its list is probably not entirely accurate (no, I'm not going to individually count from the shitty official Kespa webpage), listing zerg progamers as being the most numerous. Although, the number that actually appears on television, however, is a entirely different kettle of fish.

That's not what interests me, however. Don't you think that it's a little odd that despite your percentage figures, it doesn't change the fact that terran players won more games, while losing less games than the other two races lost in non-mirror match-ups at progaming level? Your percentage figures don't take into consideration the overall number of games played, otherwise it would be impossible for terrans to have won more, while losing less.

Repeating myself for the umpteenth time here.

More terrans on televison. Playing more games. Winning more games. Losing less games.

Letmelose

Korea (South)3227 Posts

August 13 2010 15:52 GMT

#502

On August 14 2010 00:37 Wings wrote:

Show nested quote +

Thanks for pointing the mistake out, makes the whole thing a lot more even.

Wings

United States999 Posts

August 13 2010 16:04 GMT

#503

On August 14 2010 00:52 Letmelose wrote:

Show nested quote +

Thanks for pointing the mistake out, makes the whole thing a lot more even.

I've also raised a few other points as well in a few edits. Please check my earlier post (end of pg 25).

Letmelose

Korea (South)3227 Posts

August 13 2010 16:18 GMT

#504

On August 14 2010 00:37 Wings wrote:

Show nested quote +

Let me ask just one thing here. See that bolded section?

So you said:
TvZ 2632 : 2311 in terran’s favour
PvT 2313 : 2094 in protoss’s favour

If I add that up... and please tell me if I'm doing something wrong

Terran win/loss vs other races: (2632 + 2094 : 2311 + 2313) = 4726 : 4624

And yet you state... "Terran is 4726 : 4405 overall against other races"

Okay. Either something is really wrong with what I've done (please enlighten me), or YGOSU adding programs are screwed up, or... your math sucks. Where did 4405 come out of??

As it stands, 4726 : 4624 seems MUCH MORE reasonable to me.

I haven't read any of the rest of your argument but if I see something equally questionable I'll ask it here.

EDIT 1:

Why do you CONSTANTLY state that TvT is the most played MU? How does that affect anything? It merely shows that Terran is the singular most popular race, completely unsurprising since Boxer is the leading figure in SC. In fact, I think that answers MORE questions that degrade your argument: since Terran is more popular, it naturally affects more "skilled players" as per the rules of proportion, and when those "skilled players" become progamers, more of them end up being Terrans (proportionally speaking vs Zerg & Toss). As a result, it makes perfect sense that TvT always remains the most popular MU, and I don't understand how that's supposed to help your argument.

EDIT 2:

So you basically proved, with your point # 4, that Zergs have been dominating over Terrans (and Toss) in the individual leagues as per the timescale that you chose. Doesn't that single point disprove your entire argument? Let's assume Terran WAS imba (Proof by contradiction). Then how can it not show in a huge part of the proscene (individual leagues)?? And not only does it not show, Zergs were extremely dominant according to your stats. If you assumption was correct, then the imba trend should manifest itself in what we consider to be literally HALF (or greater, to some people) of Professional Starcraft, and Zerg shouldn't have been SO dominant.

EDIT 3:

I have a key concern with your "methodology". You say Terrans have been more "dominant/imba" simply due to a difference in win differential; Terrans have positive wins, while the others might have negative, etc.

However, this methodology is, to me, clearly faulty in some cases. Let's look at your Point # 1 (after we fix your ridiculously-well-placed math error). As stated in EDIT 1, Terrans make up a larger proportion of gamers (due to popularity, or whatever). Using laws of probability, their difference in wins/losses SHOULD be larger than the difference of other races; the PERCENT PROBABILITY should remain the same/constant. Let's see:

Win/loss difference for Terrans: 102
Win/loss difference for Zergs: 89

Let's take the number of TvT's & ZvZ's: (1883 & 1777) as an estimate test proportion number for the number of Terrans/Zergs. 102/1883 = 0.054. 89/1777 = 0.05. That means if we assume a test population of equal numbers of T & Z (say, 1000 each), that means the win differential for Terrans should be 54, and the one for Zergs should be 50.

That's like... nothing. Barely any difference. How is that supposed to help your point?

The only thing I get from seeing your stats is that... wow... Protoss is getting raped by Zergs...

1) Like I mentioned before, YGOSU lists more zerg progamers than terran. The overall number of progamers for each race is roughly the same. The number of progamers on the Proleague roster today for zerg and terran are almost exactly the same. This can only mean that more terran players on being sent to represent his team, despite the overall pool of players being roughly the same.
2) I already mentioned what I believe to be the reasons for the difference in statistics in the individual leagues, compared to the progaming scene overall. While the scene greatly relies upon the existence of the individual leagues for its prestige, I believe that the nature of the individual league (already mentioned), hardly makes the statistics derived from it the representative of the entire scene.
3) I was mistaken with many of my points due to the error, but my point still stands. Terrans, I believe still have the highest winning percentage overall. The difference was substantially less than what was proposed before, but it still helps my point that in the post-Savior era, terrans have been more dominating as a race than zerg.