Lille–PAOK

3–4

HT 0–3

Kick-off: Thursday, 23 October 2025 at 19:00
Status: Match Finished

What the model said, and what happened

Incorrect — we gave this result 18%

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

18'PAOKS. Meite (assist: A. Živković)
23'PAOKA. Živković (assist: G. Michailidis)
42'PAOKG. Konstantelias (assist: F. Chalov)
57'LilleB. Andre (assist: C. Verdonk)
68'LilleH. Igamane (assist: T. Meunier)
74'PAOKA. Živković

Match statistics

Match statistics
Lille		PAOK
22	Shots	8
11	Shots on target	7
11	Shots inside the box	6
3	Goalkeeper saves	8
13	Corners	2
3	Offsides	0
65%	Possession	35%
551	Passes	309
486	Accurate passes

What we expected, and what happened

Expected goals before kick-off: Lille 2.01 · PAOK 0.97. On the pitch: 3-4.
Shots Lille 22 · PAOK 8; on target 11 and 7.
Lille kept the ball more (65%).

Referee

Nikola Dabanovic, Montenegro→

3.00

yellows per match

0.00

reds per match

matches of data

2 matches is a small sample — these averages are noisy and should not be read as a tendency.

Model prediction

Why this prediction?

The model puts the most likely outcome at 61.8%: “Lille to win”. Second comes “a draw” at 20.4%.
Behind that distribution are the expected goals: 2.01 for Lille, 0.97 for PAOK.
The highest double chance value is 82.2% (1X — Lille win or draw).
These values were computed from 1,730 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.

Lille to win61.8%

Draw20.4%

PAOK to win17.8%

When the model has said 60–65% in the past, 1,274 of 2,042 predictions were correct (62.4%). Accuracy record →

(assist: K. Despodov)

78'LilleH. Igamane (assist: Chancel Mbemba)

90+9'PAOKTomasz Kędziora · red card

7	Fouls	15

1	Yellow cards	5

0	Red cards	1

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.

When the model has said 50–55% in the past, 3,625 of 7,026 predictions were correct (51.6%). Accuracy record →

Double chance

1X — Lille win or draw82.2%

12 — any result other than a draw79.6%

X2 — draw or PAOK win38.2%

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 (20.4%).

Goals market

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

The model gives a probability below the baseline (54.4%). 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 Lille score2.01

Attack: Lille score above the league average (18% more).

Opponent's defence: PAOK concede 31% below the league average — a defence that seriously brakes even a strong attack.

Result: These two effects give an expectation of 2.01 goals (including home advantage).

How many goals PAOK score0.97

Attack: PAOK score above the league average (48% more).

Opponent's defence: Lille concede 19% fewer than the league average, which pulls the expectation down.

Result: These two effects give an expectation of 0.97 goals (including the away penalty).

Most likely scorelines

1–010.6%2–010.3%2–110.0%

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 1730 matches played in this league before kick-off; the team with fewer has 111 matches. No data from after the match was used. The model's track record is published unfiltered.

Week 3 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.