| Forum Index > SC2 General |
|
BronzeKnee
United States5218 Posts
![[image loading]](http://s23.postimg.org/wf44ixxi3/Balance.png) Protoss is pretty much in free fall, especially in PvZ (haven't had a win rate that imbalanced for three years, and PvZ is more imbalanced than ever), and yet nerfs are planned for Pylon Overcharge and the Adept... interesting. http://aligulac.com/misc/balance/ | ||
|
Cascade
Australia5405 Posts
 Can we get some information about sample size and/or error bars or something please? These numbers are close to worthless without an idea of how large the uncertainty is. Maybe you can vertical make a line for lotv as well, as you have for hots? Interesting to see how PvT dropped a lot first month after hots, then immediately recovered for month 2, while TvZ had a 3-4 month advantage. | ||
|
BronzeKnee
United States5218 Posts
| ||
|
Cascade
Australia5405 Posts
On January 02 2016 17:08 BronzeKnee wrote: These aren't my stats. Click the link and you'll get all the information regarding sample size and such. The December stats are based on 2683 games. Yeah, I had a quick look, and wasn't very accessible. I'd like to have the information on the plot, to better understand what is random fluctuations, and what isn't. Hopefully the people making the plot will read this. Shouldn't be hard to add. Thanks though! 2683 seems like a lot. | ||
|
BronzeKnee
United States5218 Posts
I don't think there is random fluctuations with balance. There is map changes, patches, meta swings, ect that affect balance. With such a large sample size, it would be impossible to tell a random fluctuation from a meta swing. The second chart is something I pushed for a long time ago, and really tells the story for me. It predicts performance based on past results, and then compares the prediction to real performance. This helps control for players who go on big hot streaks and are very dominant in a single matchup from skewing the data. We continue to see Protoss underperform on that chart. | ||
|
Areanon
Afghanistan5 Posts
| ||
|
Topdoller
United Kingdom3860 Posts
Also the word "Balance" is misleading in the title of the OP. Shouldn't it say "Win Rate" or something or are you trying to promote your agenda here? Aligulac is nothing to do with balance, it never has and it never will, especially when the matchmaking system is designed to match player v player | ||
|
dNa
Germany591 Posts
On January 02 2016 18:04 Topdoller wrote: More stats which mean nothing, when you consider the maps currently in use. Also the word "Balance" is misleading in the title of the OP. Shouldn't it say "Win Rate" or something or are you trying to promote your agenda here? Aligulac is nothing to do with balance, it never has and it never will, especially when the matchmaking system is designed to match player v player those numbers are based on pro games, not ladder games. | ||
|
Topdoller
United Kingdom3860 Posts
On January 02 2016 18:11 dNa wrote: those numbers are based on pro games, not ladder games. So its nothing to do with Balance.I watched the majority of GSL qualifiers, where a lot of the top players struggled with the new meta and made massive mistakes loosing them the games. The map pool was one of the biggest deciders for the results "Note that this yields information about metagame balance near the top of the skill ladder, and is not to be confused with (although likely correlated to) actual game balance throughout the whole player population." | ||
|
207aicila
1237 Posts
On January 02 2016 17:56 Areanon wrote: When Aligulac begins to count maps in balance? The main problem of today's balance is imbalanced map pool This has been addressed before. The answer is probably never. While I don't disagree with your assessment, bear in mind Aligulac is not intended as a tool of checking and monitoring game balance, although you may use it as such. On July 11 2013 06:39 Grovbolle wrote: While I won't discuss the statistical/predictional value on the matter of maps or no maps, I can discuss some of the more simple issues: 1: Getting the info on maps played can be tough unless you are talking about the big leagues, let's not get into the problems with different versions of a map and some with forced spawns etc. 2: Backtracking the existing DB would take months since we can't do it in bulks like we did with the event NSM. 3: Number of capable volunteers = 5-10 4: The entire entry system/parsing system would need an overhaul. 5: As far as I recall, we store the rows of matches as sets, not games, meaning that they would needed to be split with an extra attribute to keep track of which games belongs to which set. I really wish we could have maps, but if you check aligulac.com/db you see that outside kiekaboe, conti, BB, MoP and shellshock, we really doesn't have nearly enough manpower to consistently keep it all updated, nor do we have the prestige of being affilliated with TL to attract them. | ||
|
Cascade
Australia5405 Posts
On January 02 2016 18:23 Topdoller wrote: So its nothing to do with Balance.I watched the majority of GSL qualifiers, where a lot of the top players struggled with the new meta and made massive mistakes loosing them the games. The map pool was one of the biggest deciders for the results "Note that this yields information about metagame balance near the top of the skill ladder, and is not to be confused with (although likely correlated to) actual game balance throughout the whole player population." The stats mean nothing? Really? This is the best information we have on balance... It's showing exactly what it says: tosses tend to lose against zergs lately. That is very important information imo, I don't understand how you can argue otherwise. People put all kinds of meaning into the word "balance", so maybe would've been better to not use that word in the OP, but it seems like your issue is deeper than that? Anyway, almost all definitions of the word "balance" will be correlated with the win-rates in the OP. They are not interchangeable (partially because people don't agree on what "balance" actually means), but "nothing to do with balance" just isn't accurate. Is it just a phase of the meta, and will straighten out in a month? Should we change map pool? Do we need to change the units? These stats can't tell, and I don't think anyone can tell. But it does provide as much information as we can get to speculate further, and I think that what we see here is reason to start thinking about changes to either the map pool or units. | ||
|
Beelzebub1
1004 Posts
On January 02 2016 18:23 Topdoller wrote: So its nothing to do with Balance.I watched the majority of GSL qualifiers, where a lot of the top players struggled with the new meta and made massive mistakes loosing them the games. The map pool was one of the biggest deciders for the results "Note that this yields information about metagame balance near the top of the skill ladder, and is not to be confused with (although likely correlated to) actual game balance throughout the whole player population." I agree with this, the map pool is absolute shit which always ends up screwing one race over for the season. I do believe that Protoss is in a bit of a learning curve right now at the top level vs. Zerg but nothing really looks particularly imbalanced, I saw plenty of Protoss having total micro fails and losing Disruptors for no good reason in important engagements or not even bothering to utilize the Warp Prism at all. I think a few tweaks here and there might be okay but we should aim for better map pool and another month or two of pro level games before the bigger buffs/nerfs start to get tossed around. | ||
Inflicted
Australia18228 Posts
On January 02 2016 17:11 Cascade wrote: Yeah, I had a quick look, and wasn't very accessible. I'd like to have the information on the plot, to better understand what is random fluctuations, and what isn't. Hopefully the people making the plot will read this. Shouldn't be hard to add. Thanks though! 2683 seems like a lot. It's there if you hover over the dots 987 TvZ 635 PvT 1061 PvZ | ||
|
zerge
Germany162 Posts
| ||
|
Big J
Austria16289 Posts
| ||
|
keglu
Poland485 Posts
On January 02 2016 18:04 Topdoller wrote: More stats which mean nothing, when you consider the maps currently in use. Also the word "Balance" is misleading in the title of the OP. Shouldn't it say "Win Rate" or something or are you trying to promote your agenda here? Aligulac is nothing to do with balance, it never has and it never will, especially when the matchmaking system is designed to match player v player Strong post First attacks poster that he has agenda when "Balance report" is name of this graph. Then discredits results because of matchmaking system when report is based on tournaments games. | ||
|
Cascade
Australia5405 Posts
On January 02 2016 19:06 Inflicted wrote: It's there if you hover over the dots 987 TvZ 635 PvT 1061 PvZ Oh, I missed that! How embarrassing, especially as I am coding interactive plots right now... :o Thanks! | ||
|
keglu
Poland485 Posts
On January 02 2016 19:06 zerge wrote: Please stop using winrates as an equivalent for balance. Balance means equal possibility to win given equal skill level. We don't have tools to measure skill level. | ||
|
Cascade
Australia5405 Posts
 anyway, thanks for keeping it up aligulac ppl! | ||
|
NKexquisite
United States911 Posts
TvZ - pretty good PvZ - uhhhhh | ||
| ||
StarCraft 2 StarCraft: Brood War League of Legends Super Smash Bros Other Games |
|
OSC
Korean StarCraft League
OSC
IPSL
Dewalt vs Bonyth
OSC
OSC
uThermal 2v2 Circuit
Replay Cast
Patches Events
|
|
|
EPT
