luckadjusted
Premier League Matchday 5 · 2026-27

Matchday 22 · Premier League 2015-16 · January 17, 2016

Stoke 0–0 Arsenal: who deserved to win?

Stoke and Arsenal drew 0–0. Replayed with the way both teams played, Stoke win 4 of 10, Arsenal 3, and 3 end level. Stoke made the better chances: 0.9 expected goals to 0.5.

Full time

GoalsxG
Stoke City 0 0.92
Arsenal 0 0.49

January 17, 2016 · bet365 Stadium · Referee: Craig Pawson

Luck index −12 Stoke about even for the draw

Each team’s own shots from this match, resampled into new matches 4,000 times, each a goal at its expected-goals chance: with the way both teams actually played, how often does this performance win? How the replay works

Ten replays. Stoke win 4, 3 end level, Arsenal win 3.

  1. 1 STK 1 ARS 0
  2. 2 STK 1 ARS 0
  3. 3 STK 2 ARS 0
  4. 4 STK 2 ARS 0
  5. 5 STK 0 ARS 0
  6. 6 STK 0 ARS 0
  7. 7 STK 0 ARS 0
  8. 8 STK 1 ARS 3
  9. 9 STK 0 ARS 1
  10. 10 STK 0 ARS 1

The narrowest view of luck

The same match, with only the coin flips re-rolled: every shot kept where and how it was taken, its outcome redrawn at its chance (75% of finishing counts as luck), own goals at the league’s rate.

There is no betting line for this match, and no forecast from before kickoff: matches of past seasons are replayed, not forecast.

How 4,000 replays of the 90 minutes ended

Stoke win 45%, 35% end level, Arsenal win 21%. The line is what actually happened: Draw.

Show as a table
After 90 minutes in the replaysShare of replays
Stoke by 3+4.7%
Stoke by 212.1%
Stoke by 128.0%
Draw34.6%
Arsenal by 115.7%
Arsenal by 24.1%
Arsenal by 3+0.8%

Where the luck was

FinishingShotsGoalsExpected goalsAbove expected
Stoke1400.92−0.92
Arsenal800.49−0.49

Own goals aren’t shots, so they’re counted on their own. A side above expected finished better than its chances, or met a goalkeeper on a bad day; 75% of the gap counts as luck (why that share).

When it happened

Stoke’s chance of winning

Numbers mark the biggest luck moments, listed below.

Our in-game win probability: the score and the minutes left, with goals for the rest of the match at the league’s rates. A draw counts half, as in the replays. Shaded above the middle while Stoke was ahead in it, below while Arsenal was.
#WhenWhat happenedWin probability for
1 20' Olivier Giroud didn’t score from a shotFinishing · 19% chance +5% STK
2 46' Olivier Giroud didn’t score from a shotFinishing · 6% chance +2% STK
3 55' José Luis Sanmartín Mato didn’t score from a shotFinishing · 25% chance +8% ARS
4 56' Bojan Krkíc Pérez didn’t score from a shotFinishing · 15% chance +5% ARS
5 64' José Luis Sanmartín Mato didn’t score from a shotFinishing · 6% chance +2% ARS
6 84' Joel Nathaniel Campbell Samuels didn’t score from a shotFinishing · 4% chance +2% STK
7 89' Jonathan Walters didn’t score from a shotFinishing · 7% chance +3% ARS
8 89' José Luis Sanmartín Mato didn’t score from a shotFinishing · 7% chance +3% ARS

Adjust the assumptions

Think finishing is more skill than we do? Change how much of it counts as luck and the coin-flip replay of the 90 minutes reruns in your browser.

Only coin-flips re-rolled

Stoke 45% · draw 35% · Arsenal 20%

Final margin
even
Luck in goals
0.3 for Arsenal
Luck-adjusted margin
Stoke by 0.3
Replay with same day's form
Stoke 45% · draw 35% · Arsenal 21% unchanged: why
How much of each is luck?

0% treats it as skill, 100% as pure luck. The tick is our fit.

What counts

About these numbers

Expected goals per shot from StatsBomb’s open data; the replays, the in-game win probability and the luck shares are our own models, fitted and tested on every shot of four full league seasons and four tournaments, these matches included (methodology and validation).

All matchday 22 matches, 2015-16 · Stoke · Arsenal

More from matchday 22, 2015-16

Every match of matchday 22, 2015-16