Bridges 2026 Scoreboard
495 team Bridges games · 3 Jan – 30 Sep 2026
| # | Player | Tier | |||
|---|---|---|---|---|---|
| 1 |  Straight |
S | 66.7% | 57.9% | +2.58 |
| 2 |  Dawson |
A | 59.5% | 50.5% | +1.08 |
| 3 |  zedify |
A | 56.9% | 51.7% | +0.54 |
| 4 |  explicy |
A | 56.9% | 54.1% | +0.26 |
| 5 |  ikyy |
A | 54.8% | 47.0% | +0.38 |
| 6 |  mlclm |
B | 52.5% | 57.9% | −0.31 |
| 7 |  Border |
B | 47.9% | 49.1% | −0.95 |
| 8 |  NIECHOPPA |
B | 47.7% | 48.0% | −0.68 |
| 9 |  creio |
B | 47.2% | 50.5% | −1.12 |
| 10 |  chrzz |
C | 35.5% | 40.2% | −2.63 |
| 11 |  Mohtus |
C | 34.1% | 44.0% | −2.49 |
| 12 |  Pufferfishh |
D | 27.8% | 39.2% | −4.01 |
Adjusted win %: how often a team would win if this player replaced an average player from the group, after accounting for who they played with and against. Ranked by this to start with: click Adjusted win %, Raw win % or Carry to sort by that column, and click again to flip the order.
Raw win %: actual wins ÷ games played. Carry: own skill rating minus teammates' average rating (+ carries, − gets carried).
Tiers (adjusted): S 62%+ · A 53–62% · B 42–53% · C 30–42% · D under 30%.
| # | Player | |||||
|---|---|---|---|---|---|---|
| 1 | Dawson |
18.1 | 1.47 | 3.88raw 3.60 | 3.08raw 3.51 | 0.42 |
| 2 | zedify |
17.7 | 0.99 | 3.79raw 3.63 | 3.23raw 3.46 | 0.29 |
| 3 | Straight |
16.8 | 1.43 | 4.09raw 3.77 | 2.88raw 3.26 | 0.44 |
| 4 | Border |
16.7 | 1.72 | 3.59raw 3.60 | 3.61raw 3.63 | 0.47 |
| 5 | creio |
15.7 | 1.41 | 3.38raw 3.51 | 3.47raw 3.48 | 0.40 |
| 6 | ikyy |
15.1 | 1.35 | 3.72raw 3.46 | 3.26raw 3.61 | 0.37 |
| 7 | explicy |
13.3 | 1.53 | 3.88raw 3.69 | 3.31raw 3.43 | 0.45 |
| 8 | mlclm |
12.2 | 1.57 | 3.55raw 3.72 | 3.53raw 3.38 | 0.46 |
| 9 | Mohtus |
11.9 | 1.32 | 3.01raw 3.38 | 3.94raw 3.58 | 0.37 |
| 10 | NIECHOPPA |
11.9 | 1.20 | 3.38raw 3.40 | 3.75raw 3.80 | 0.32 |
| 11 | Pufferfishh |
10.9 | 1.09 | 2.77raw 3.08 | 4.21raw 3.88 | 0.28 |
| 12 | chrzz |
10.1 | 1.00 | 3.09raw 3.25 | 3.85raw 3.77 | 0.27 |
Ranked by kills per game to start with: click any column to sort by it, and click again to flip the order (team conceded sorts fewest first).
Kills / game: other players killed per game (knocking someone into their own goal counts). Goals / game: goals scored per game. Goal ratio: K/D for goals – their goals per game ÷ their team's goals conceded per game.
Team goals / conceded per game: goals their team scored and let in per game. The big number is adjusted for who they played with and against (an otherwise even 3v3 with average players from the group; an average team scores and concedes 3.51); the small number is the raw average.
Adjusted win rate over the last year
One panel per player, week by week. 50% is the group's average player. Based on 505 team Bridges games over the last 12 months (Nov 2025 – Sep 2026).
Show as a table (end of each month)
Skill is allowed to drift slowly from week to week, fitted to all the games at once; how fast it can drift was picked by testing which setting best predicted games it hadn't seen. Values between a player's games (for example the April and July breaks) are the model's best estimate, not new results. Lines start at each player's first game and stop at their last.
The better value in each row is highlighted green (for team conceded, lower is better). The numbers are the same as on the Scoreboard and Stats tabs; team goals and conceded are the adjusted ones. Head to head counts the games where they were on opposite teams.
Team maker
Tick who's playing and pick a team size. Every way to split them into two teams is listed from fairest to least fair, with each team's chance to win. If there are more players than places, it shows the fairest teams for each choice of who sits out.
The number next to each player is their skill: 0 is an average player from the group. A team's strength is its players' average skill, and the win chance comes from the gap between the two averages (the same model as the adjusted win %). A gap of 0.4 is about 60/40.
Auto uses as many players as it can (7 ticked makes 3v3 with one sitting out). Teams are always the same size; an uneven game like 4v3 isn't suggested because the model can't tell how much the extra player is worth.
Skill comes from this year's team games across 2v2, 3v3 and 4v4, so players who are much better in one format than another may play a little above or below it. Your ticks and team size are remembered on this device.