|
On September 09 2010 16:50 rycho wrote:
yeah could someone explain why every race seems to have like 55% win? its probably basic math but i'm kind of missing how this is possible
It's been explained a few times already, but it's due to:
1) Diamond players also getting opponents from lower leagues. Naturally Diamond players have a higher chance to win such a game, so their win percentage rises above 50. If it wasn't clear - what's meant by "Diamond games" here are the match histories of Diamond players; there's no way of knowing what kind of opponents they got (though naturally the vast majority of those were fellow Diamond players).
2) The stats only include the last 0-25 games of each player (all the history that is publicly available) so even if all games were Diamond vs Diamond the stats wouldn't need to add up to 50% if people were more likely to stop playing after a win streak than a losing streak (leaving more wins than losses on average to the match history of their last 25 matches). EDIT: So exactly what the dude above said, he might have been a bit more understandable 
|
On September 07 2010 00:18 Neo.NEt wrote:
Either the results of this aren't "statistically significant" (56%>53%) or I've been making some poor assumptions. Apparently Terran > Zerg on 4 player maps? This is news to me.
Should not really be, is generally harder to get your overlords into place to scout good on a 4p map. Also that means it is harder to scout where the terran is in the beginning and you're more susceptible to some kind of early harassment push or such. Also more importantly considering it, LT and Kulas are the two most favored maps for T so they add up to the percentage quite a bit for sure.
|
eh isnt 50% working as intended with the matchmaking system? i dont understand this topic :X
|
All the statistics are too close to 50% to conclude something from them.
Everyone knows the imba maps anyway.
|
Random is underpowered on Lost Temple, needs a buff. It's ok though because they hold the highest win rate on any map, 56.59% on Blistering Sands.
I think this statistic alone disputes the significance of these numbers. Since they are all so close together, the only real factor in these players wins were probably if the map fit their playstyle, or just their individual skill in the game played. I think it'd be more fitting to do maybe the top 100-200 players, and see what the statistics look like for those?
|
On September 03 2010 20:43 Alexstrasas wrote:Show nested quote +On September 03 2010 20:36 Tufas wrote:
If I read this correctly, most wins go to random players. I love that. Statistic agrees with me, blistering sand is the worst map for terran (same percentage as kulas ravine). Another interesting fact : No one likes to play desert oasis. i actualy find it very very weird that ppl bash desert oasis while no one mentions scrap station. Scrap station is a piece of shit map 10x worse than desert oasis, anyone not playing zerg should turn it off by default
Just me that refuse to cross out maps even if I hate them just because it's ignoring a part of the game which is taunting you "You can't master this map as that race haha oh maw gawd". Screw that, I'll go in there and lose all day every day until I know how to use the map for my own gain.
|
On September 03 2010 22:18 Silu wrote:
If you look at the numbers being as relative and not absolute, neither of these matter that much. The idea is to gauge what the maps are like when compared to each other for one race. You're correct that 56% for Zerg doesn't quite mean the same as 56% for Terran (the latter being a bigger sign of "imba" in this case, since they still manage a big win percentage even with relatively more mirror matches included in it). However the numbers are very comparable among a single race.
I'm quoting this because I think a lot of people are misinterpreting these stats and using them to judge whether the maps are balanced between the races. The fact is that these stats have no implications on balance arguments and are only interesting because they can tell us things like "Zerg is better on Scrap Station than on Delta Quadrant" which is purely about which map a given race prefers, but not whether either map is imbalanced in favor of a particular race.
The reason, as has been stated in previous threads with similar statistics, is that Blizzard's matchmaking system skews the results towards 50/50 for each player. Therefore, when you're comparing win rates of the different races in diamond league, you'll always see the overall percentages hover around 50%, no matter how imbalanced the map pool is. Imbalance will weed out a race and cause less players of that race to make it to diamond, however, it won't change the win percentage of the players in diamond very much at all. That's why these statistics are only comparable within a single race, to let you know which maps to thumb down if you want to increase your win percentage.
|
On September 09 2010 23:30 Najda wrote:
Random is underpowered on Lost Temple, needs a buff. It's ok though because they hold the highest win rate on any map, 56.59% on Blistering Sands.
I think this statistic alone disputes the significance of these numbers. Since they are all so close together, the only real factor in these players wins were probably if the map fit their playstyle, or just their individual skill in the game played. I think it'd be more fitting to do maybe the top 100-200 players, and see what the statistics look like for those?
Well. Obviously the Random stats are by far the least significant, because of the far smaller sample size. For the same reason the stats from the top100-200 would tell absolutely nothing because of the minuscule sample size.
|
Lots of data mining, none of it useful for analysis. So you're responsible for the stats to take forever to load? Way to annoy.
|
The fact that you can thumb down maps destroys these stats completely.
|
This is awesome, . Looking at the number of games played you can also get a pretty good look at what maps players are removing. The Kulas Ravine numbers for terran are pretty surprising. I wonder what they would look like if you could refine it to only TvZ though.
|
On September 10 2010 01:06 Dagobert wrote:
Lots of data mining, none of it useful for analysis. So you're responsible for the stats to take forever to load? Way to annoy.
Mind to elaborate what on earth you are talking about?
|
|
|
|
|
|
EPT
