?
arrow_backBack to match
Winline Insight · BO3
result vs prediction
YNTloss 1:2
series probability
25%75%
probability at match start · final 1:2
No Sorrywin 2:1
The prediction was locked in using data as of the match start: hero winrates, form and head-to-head were computed only from games before it began. This is how the model's honest accuracy is verified.
Teams' go-to heroes · 6 months
Who plays what and how often they win. "Signature" — from 6 games with a 60%+ winrate: such a pick statistically raises the win chance.

YNT

Mars
Marsoften banned
Dominic · 10 games
40%
Grimstroke
Grimstroke
jesv · 9 games
33%
Pugna
Pugnaoften banned
Malr1ne · 9 games
44%
Sand King
Sand Kingoften banned
Dominic · 9 games
56%
Tiny
Tiny
DarkLord^ · 9 games
33%
Death Prophet
Death Prophetoften banned
dSa · 8 games
25%
Bane
Bane
dSa · 7 games
57%
Chaos Knight
Chaos Knightoften banned
DarkLord^ · 7 games
29%

No Sorry

Bane
Banesignature
BLACKARXANGEL · 40 games
60%
Skywrath Mage
Skywrath Mage
y0nd · 34 games
53%
Mars
Marsoften banned
LastHero- · 29 games
45%
Hoodwink
Hoodwink
y0nd · 28 games
46%
Tiny
Tiny
Worick · 28 games
57%
Storm Spirit
Storm Spiritoften banned
Worick · 26 games
58%
Ember Spirit
Ember Spiritoften banned
Worick · 24 games
54%
Keeper of the Light
Keeper of the Lightoften banned
BLACKARXANGEL · 18 games
50%
Patch meta · 3 months
The most picked heroes of the pro scene and their winrate. Picking a meta hero adds points to the team in the model.
Mars
Mars
479 picks · 660 bans
48%
Tiny
Tiny
473 picks · 448 bans
49%
Bane
Bane
454 picks · 277 bans
57%
Dragon Knight
Dragon Knight
414 picks · 495 bans
52%
Sand King
Sand King
414 picks · 566 bans
54%
Storm Spirit
Storm Spirit
401 picks · 653 bans
56%
Pangolier
Pangolier
368 picks · 574 bans
53%
Grimstroke
Grimstroke
349 picks · 297 bans
46%
Prediction components
Form (series winrate)31% vs 61%
Hero pool strength56% No Sorry
Head-to-head1 – 4
Model total75% No Sorry
Prediction by game vs result
Probabilities were locked in using statistics as of the match start: the draft was assessed by hero comfort, counter-picks and late-game potential. We compare against the actual result.
Game 1prediction correct
YNT 25%No Sorry 75%
The model gave 75% on No Sorry — the game was taken by No Sorry.
Game 2prediction missed
YNT 25%No Sorry 75%
The model gave 75% on No Sorry — the game was taken by YNT.
Game 3prediction correct
YNT 25%No Sorry 75%
The model gave 75% on No Sorry — the game was taken by No Sorry.
Model check result
Model favoriteNo Sorry · 75%
Series resultNo Sorry 1:2
Series predictioncorrect
Accuracy by game2 of 3