Curious what hero came out on top of the Vainglory Autumn Finals in North America? We’ve dug through all of the draft pick data and compiled every hero’s win-ban rate, their presence in matches, and more.
Every Hero and Their Participation
The first graph depicts how many games that hero won, was banned from, lost and was neither played in or banned in (absent). The second graph displays each hero’s win-ban rate vs lose rate from the matches they were in or were banned from.
Win-ban rate is useful for giving us an idea of how successful a hero was in a tournament and can also show us if there are balance issues with a hero. Win-ban rate by itself isn’t that insightful, you need the context of how often the hero was played. For example, Petal had a 100% win-ban rate, which makes her sound insanely overpowered – until you learn that she was only played once.
Adagio
Ardan
With Fortress being banned so often, Ardan found himself as the primary support pick and played in a total of 26 (of the 34) games.
Blackfeather
No Win-Ban Rate
Blackfeather wins the award for least popular hero in this tournament. He wasn’t played a single time. We actually had to double-check that he was enabled for this tournament.
Catherine
While a popular support pick in this tournament, Catherine frequently found herself losing to the avenging father, Ardan.
Celeste
Celeste was the obvious queen of these finals. She was present in 24 matches and banned from another 8. Despite being so popular, she still managed to maintain a positive win-ban rate.
Fortress
Our king of aggression, Fortress came in just ahead of the Starqueen and snagged himself the title of king. He has not only the highest win-ban rate (out of any hero played more than once), but is tied for the highest presence missing only two matches. Because he was banned so often, however, this opened up an opportunity for Ardan to steal some limelight.
Glaive
Joule
Koshka
Koshka pounces in just behind Celeste as the 3rd popular and successful hero in these matches.
Krul
Petal
While Petal technically has a 100% win rate, she was only played once and is omitted from the top choices.
Phinn
Many tried and just as many failed with Phinn. Phinn has the highest number of matches played out of any 0% win rate hero.
Ringo
Rona
Rona beat out Koshka in terms of bans, but left the fray with a slightly lower win-ban rate despite this.
SAW
The “Tier II Poop Strat” was attempted three times in this tournament and failed each time. Anyone who’s played against the strategy before, undoubtedly feels a sense of justice when they hear that.
Skaarf
Skye
Taka
Vox
Kings and Queens of the Vainglory Autumn Finals
This graph compares all of the heroes on two stats: win-ban rate—of the matches they were in or were banned from, what percent were wins or bans, not losses— and presence—the percent of matches they were in or were banned from, compared to the total number of matches (there were 34 total matches in the Vainglory Autumn Finals). Win-ban rate lets us see how successful a hero is, and presence lets us put that into some context by showing how much mind share a hero actually received. For example, by only looking at the win-ban rate, it would appear that Petal “won” this tournament; however, when factoring in presence, you know that her numbers were skewed due to only playing a few matches, meaning she likely wasn’t as popular as other choices because she wasn’t banned or picked often.
Celeste and Fortress had the highest presence in this tournament, and—ignoring Petal—Fortress had the highest win-ban rate, indicating that he was most successful hero in this tournament.
Now we take a slightly different perspective: the same win-ban rate compared with matches played—the percent of matches won or lost, not banned from or absent from. This allows us to ignore bans and see how much actual play a hero received.
When we look at this, we can see how Ardan ended up replacing Fortress in a lot of these matches because Fortress was so often banned or picked by the other team. We can also see how some heroes were almost never—or actually never—banned. Like Catherine, who was played 65% and present 68% of the time. This gives an indication that she wasn’t a threatening hero in the tournament, which is further backed up by her negative win-ban rate.
For similar content, check out our 1.11 Tier List and expect a similar analysis of this coming weekend’s Vainglory Autumn Finals in Europe. Also check out some other posts about the North American Vainglory Autumn Finals and VIPL:
9 Comments
ChazIsidore
Dec 16, 2015 9:27 amI’ve pulled similar stats and gathered some thoughts on thee EU live finals. I don’t have a good location to post (reddit maybe?).
The meta differences are amazing. Fortress had a 29% win rate in EU, Koshka 29%. Adagio as 80% in EU with high use rate.
Both metas agree on a couple things: Blackfeather and Taka are bad. Ardan and Rona are good.
ChazIsidore
Dec 13, 2015 7:12 pmThis is great. I was going to do the same thing for my own curiosity, but didn’t have a chance to watch all the matches.
I wonder if having pure win rate might be a better stat for success. It seems to me, looking at actual win/loss is more important than the possible win a ban implies. It is also simpler, in that 50% is the break point between good and bad.
Ardan, for example, was only OK this tournament, with a 46% win rate (12-14). This is probably due to the Fortress (77% win rate) situation, but still.
Rona was only slightly better that Glaive win rate wise, with 58% vs 55%. Koshka was better than both with 65%.
Win rates for Celeste (54%), Skaarf (53%), and Ringo (55%) were close to identical but the two crystal heroes were banned much more often.
Looking at win rates, it seems to me that Fortress was just dominant, and probably should have been banned more often. Any time the second team didn’t ban him, they were looking at a less than 25% win probabiliy. I would also theorize that some of Koshka’s strong numbers are due to the Fortress-Koshka combination, and if that was the case than banning Fortress also neutralizes Koshka. And then the heroes left are closely balanced.
LC
Dec 15, 2015 12:08 pmFully agree, win/ban rate as a unit of analysis is really murky.
Ban rate should be its own unit of analysis although I guess there are some observations that can be gleamed from them combined.
Jackaloupe
Dec 15, 2015 3:46 pmThat’s why win-ban rate isn’t used by itself in this. It’s put along side all of the information, including losses, wins, bans, and absences.
Win-ban is used to show a hero’s dominance.
Wins and presence are used to see how successful that hero actually was.
A1io
Dec 12, 2015 1:38 amWill there be a 1.12 Tier List? I found the 1.11 Tier List article very helpful.
MrTouchnGo
Dec 12, 2015 6:24 amYes, there will! Keep your eyes out for it.
Thommy8
Dec 11, 2015 10:18 amThis is fantastic! Thanks for putting this together.
Skaarf is in the graphs, but I think the little lizard creature is missing from the first part of the post.
Jackaloupe
Dec 11, 2015 3:54 pmYep, we’ve added him in now. Good catch!
Avrye the Small
Dec 11, 2015 10:15 amYay for fortress!!!