Ember Spirit

Ember Spirit has the most favourable matchups of any hero in this dataset — twenty-three pairings clearing the significance test in its favour, against a single one going the other way. The nearest rival is Phoenix at nineteen. By that measure it is the most reliably advantaged hero here.

It never averages a win. The bracket curve runs 46.7, 46.7, 46.8, 48.7, 48.9, 48.9, 49.7, climbing three points across the range and stopping just short of fifty at the top.

Those two paragraphs are the whole page, and the gap between them is the most instructive thing on this site. Twenty-three measurable advantages, one measurable disadvantage, and a losing record. Whatever Ember Spirit is good at, it is not translating into games won.

The explanation is partly arithmetic. A reportable matchup only covers games where that specific opponent appeared, and twenty-three heroes out of 127 is still a minority of drafts. But Ember Spirit is not alone, and the company it keeps is the real finding: five heroes here carry ten or more favourable matchups while never crossing fifty at any bracket — this one, Shadow Demon, Treant Protector, Snapfire and Windranger. The pattern is common enough to be structural rather than a curiosity.

The matchup list is also unusually well evidenced, which rules out the easy dismissal. Hoodwink rests on 368 games, Tiny on 387, Tusk on 358, Windranger on 301. These are not thin results that will evaporate next patch; the margins on the larger entries are four to five points, tight enough to trust the direction and roughly the order.

There is one more angle worth the space. Across all 127 pages, Ember Spirit is the hero that appears most often on somebody else’s unfavourable list — it shows up on twenty-three of them, ahead of Phoenix on nineteen. Read from the other side of the table, it is the single most frequently dreaded opponent in the game, which makes the losing record harder to dismiss as a measurement artefact and more interesting as a fact about the hero.

The practical reading is that Ember Spirit is a draft-dependent hero whose dependence runs deeper than usual. Against Void Spirit at 63.5%, Terrorblade at 63.1% or Tidehunter at 61.3% the numbers are emphatic. Across everything else, they are not, and there is a great deal of everything else.