Designated Balance Discussion Thread - Page 904

On February 13 2014 00:19 SC2Toastie wrote:
http://aligulac.com/players/29/results/?after=2014-01-23&before=2014-02-12&race=ptzr&country=all&bestof=all&offline=both&game=all&op=Filter

In these games, jjakji is 62–17 (78.48%) in games and 30–6 (83.33%) in matches.

On February 13 2014 00:22 Ghanburighan wrote:


That's easy:

List 103 includes 24 Jjakji v P games (17-7, 71%). There are 445 PvT games in the list, Jjakji contributed 5% of the games.

List 104 includes 12 Jjakji vP games, (8-4, 67%). There are 226 PvT games in the list, Jjakji contributed 5% again.

On February 13 2014 00:33 Pirfiktshon wrote:
Actualy ForGG made some Serious Dent in those Stats as well... between Jjakji and ForGG if their stats were taken out of that we would see some serious IMBA hahahahaha

On February 13 2014 00:35 TheDwf wrote:

But there are other Koreans, Zerg or Protoss, who similarly skew the ratios towards their respective races; that's the problem with those winrates.

On February 13 2014 00:40 Glorfindel! wrote:
Well if the stats had shown 50% and almost no TvTs atleast gone could argue its because fewer players play Terran or something. But the super low amount of TvT and still the imblanace in PvT is just.. Scary :/

On February 13 2014 00:35 Ghanburighan wrote:
Btw, TheDwf, I updated my post with Jjakji's actual contribution in terms of win percentages while you were typing.

On February 13 2014 00:38 SC2Toastie wrote:

The low count AND the fact it is still this imbalanced in PvT is indicative of a large balance problem.

Don't worry, you're preaching to a converted. And actually, I think those numbers fail to represent the extent of the problem, just like international winrates for TvZ were not as bad as the 35:65 in Code S at the end of WoL.

On February 13 2014 01:35 Faust852 wrote:
Belshir was a good map where blink wasn't totally broken.

On February 13 2014 01:39 Big J wrote:


Cloud Kingdom was a good map where blink wasn't totally broken.

On February 13 2014 01:11 Pirfiktshon wrote:


I don't know my Fellow Terrans some days it feels like we are climbing MT. Everest where the mountain just keeps getting Taller and Taller as we climb it.....

The Current system of Analyzing statistics seems lack luster.... Especially because SC2 Statistics and the way the system is built is almost always meant to give us the 50% winrate statistic in most situations....

I think analyzing performance of players vs the performance of the other taking into account what they have to go through in order to perform the same and accomplish the same results is the best way but having 3 different races and having them intentionally different mechanics makes that difficult.....

On February 13 2014 02:22 Snusmumriken wrote:


Cloud kingdom was the best map in sc2 history imo.

As much as it hurts me to say, I think there are bigger issues atm than tvp which is pretty broken. The swarmhost is just terribad. I really hope they change its DESIGN, because tweaking balance wont fix that unit.

On February 13 2014 02:22 SC2Toastie wrote:

The current system has a massive flaw in it; winrates go towards 50% at any time because imbalance just means less games -> the only way we can state there's an imbalance going on is by looking at a significantly lower count of games player by one race.

The problem with finding another way is that you always need to use some sort of number to define the skill of a player.

One way we could maybe do that is use the Aligulac's system of player performance; if a low Protoss beats a high terran, that influences statistics more than when a high terran beats a low protoss: call it an upset adjustment, I don't know.
Problem with that is the fact Aligulac responds somewhat slow to a player falling behind in skill.

All in all, the low count of games we currently see on top of a 10% imbalance in TvP indicate there's something very, very wrong with TvP, which I doubt will be solved easily with patches on the scale we currently see.

On February 13 2014 02:26 SC2Toastie wrote:

The problem in that is the fact you need about 2/3 remaxes just to beat one Void Ray Collosus Archon army. Zerg cannot go head to head with a Protoss army and needs 30 minute of chipping damage and making static defense and casters to be able to handle the Protoss army. Especially on the current maps, the third (and usually the forth too) are soo easy to take.

On February 13 2014 02:26 Snusmumriken wrote:


nah, the easiest way to spot imbalance is and has always been if one or two RACES dominate upper parts of the big tournaments. If protoss keeps on winning, but its not one or two protoss who does all the winning, then its a clear sign of imbalance (its ok if its just one or two since they may simply be much better than everyone else, see MVP 2011). This is what we see now with dear, sos, mc, san etc. winning. So its not one or two players who simply dominate regardless of race, its clearly the race that carriers whoever plays it to victory (exaggeration).

Winrates are just poor guides for so many reasons some of which youve mentioned, that theyre next to useless. Only in conjunction with other data are winrates interesting. or rather one should say, if winrates are imbalanced then that is almost definately a sign of imbalance, but if winrates arent imbalanced that doesnt exclude imbalance.