UEFA Champions League · League Stage - 2

Bayer Leverkusen–PSV Eindhoven

1–1

HT 0–0

Kick-off: Wednesday, 1 October 2025 at 19:00
Status: Match Finished

What the model said, and what happened

Ended level — we gave the draw 21%

This verdict goes into the public record unchanged; we do not filter out the ones we get wrong. The model's track record

Goals and red cards

65'Bayer LeverkusenChristian Kofane
72'PSV EindhovenI. Saibari (assist: G. Til)

Match statistics

Match statistics
Bayer Leverkusen		PSV Eindhoven
14	Shots	6
4	Shots on target	1
12	Shots inside the box	4
0	Goalkeeper saves	3
7	Corners	4
2	Offsides	2
46%	Possession	54%
499	Passes	573
445	Accurate passes

What we expected, and what happened

Expected goals before kick-off: Bayer Leverkusen 2.07 · PSV Eindhoven 1.38. On the pitch: 1-1.
Shots Bayer Leverkusen 14 · PSV Eindhoven 6; on target 4 and 1.

Referee

G. Nyberg→

3.48

yellows per match

0.12

reds per match

matches of data

Model prediction

Why this prediction?

The model puts the most likely outcome at 53.7%: “Bayer Leverkusen to win”. Second comes “PSV Eindhoven to win” at 25.7%.
Behind that distribution are the expected goals: 2.07 for Bayer Leverkusen, 1.38 for PSV Eindhoven.
The highest double chance value is 79.4% (12 — any result other than a draw).
These values were computed from 1,944 past matches.

This is a prediction, not a guarantee; a probability is not a certainty.

These sentences are built by a fixed template; every number in them comes straight from the model's computation, none invented while the text was written.

Bayer Leverkusen to win53.7%

Draw20.6%

PSV Eindhoven to win25.7%

When the model has said 50–55% in the past, 1,798 of 3,471 predictions were correct (51.8%). Accuracy record →

8	Fouls	11

0	Yellow cards	2

0	Red cards	0

The measurements our data provider sent for this match. A row that was not measured shows “—”; writing 0 into an empty field would claim we measured it and found none.

Both teams to score65.2%

When the model has said 65–70% in the past, 764 of 1,176 predictions were correct (65.0%). Accuracy record →

Double chance

1X — Bayer Leverkusen win or draw74.3%

12 — any result other than a draw79.4%

X2 — draw or PSV Eindhoven win46.3%

These three values add up to 200%, not 100% — and that is not a mistake. The options are not mutually exclusive; every outcome is counted in two of them. For the same reason 1X plus X2 exceeds 100% by exactly the probability of a draw (20.6%).

Goals market

The 3+ goals prediction is not published for this match.

The model gives a probability below the baseline (55.3%). In backtesting the model's edge in the "fewer goals" direction could not be confirmed; we don't publish an unconfirmed prediction.

What these probabilities rest on

Expected goals comes out of a team's attack weighed against its opponent's defence. A strong attack is damped when it meets a strong defence — which is why comparing the ratings on their own is misleading.

How many goals Bayer Leverkusen score2.07

Attack: Bayer Leverkusen score close to the league average.

Opponent's defence: PSV Eindhoven concede 21% fewer than the league average, which pulls the expectation down.

Result: These two effects give an expectation of 2.07 goals (including home advantage).

How many goals PSV Eindhoven score1.38

Attack: PSV Eindhoven score 73% more than the league's average team — one of the sharpest attacks in the league.

Opponent's defence: Bayer Leverkusen concede 26% fewer than the league average, which pulls the expectation down.

Result: A strong attack meets a strong defence. The two largely cancel and the expectation drops to 1.38 goals — markedly lower than the attack alone would promise.

Most likely scorelines

2–19.4%1–18.8%1–06.8%

A single scoreline is always less likely than an outcome: “home win” is the sum of 1-0, 2-0, 2-1 and dozens more. So the most likely single score can be a draw while the most likely outcome is a home win; the two do not contradict each other.

Ratings were computed from 1944 matches played in this league before kick-off; the team with fewer has 107 matches. No data from after the match was used. The model's track record is published unfiltered.

Week 2 of the season. This prediction was computed from previous seasons. Summer transfers, budget changes and squad rebuilds aren't in the ratings yet — the model can only see that a team has been rebuilt once new results come in.

We measured this: the bias is only pronounced in the opening week. In this week the model's overall accuracy does not separate from the rest of the season — that is, we found no measurable flaw. The one tendency across all opening weeks is that home wins are called slightly too often (said 43.0%, got 41.9%; n=583).

This sample is small and not statistically established, which is why we do not correct the model for it. The correction we tried (pulling probabilities toward the league average) didn't win on holdout, so it wasn't added. Read opening-week predictions with a little more caution.