03.7

Interpreting results

Combining stats + voting into a defensible team selection — and surfacing the disagreements.

Trials → [your trial] → Report

The report combines two sources of truth:

Stats side

  • Total goals, gains, losses per player
  • Per-position breakdown (which positions they played, how they performed)
  • Court time

Voting side

  • Average rating per player across all selectors
  • Per-position rating (a player might rate 4.5 at GS but only 3.2 at GA)
  • Spread of ratings (high standard deviation = selectors disagree)
  • Selector comments

Picking the team

The most useful column is usually rating × games-played-in-preferred-position. A player rated 4.5 across two GS games is a stronger signal than 4.8 across half a game.

Export to CSV for offline review with co-selectors who weren't present.

💡 Tip

The disagreement column. If three selectors rate a player 5/5 and one rates them 2/5, that's not noise — that's a conversation. Surface high-variance players to the selection panel for discussion before locking the team.

Need a hand?

Talk to the GameStats team.

We'll walk you through it 1:1 — no sales pitch, just answers. Whether you're trialling 80 players this weekend or migrating a season's worth of stats, we've probably seen it before.