Sign in Register

← All posts

Recap Friday 23 October 2026

Football Results & Model Review Friday 23 October 2026: Mönchengladbach and Nice stun favourites; Middlesbrough rout Bolton

Friday 23 October gave us 47 completed fixtures with a pre-match prediction attached, and the honest summary is that it was a day of two halves for our model. It was a modest day on winner calls, but a very good one on goals. Of the 47 matches, our winner call was correct in 22. That is a long way from a triumph, and we will not dress it up as one.

The goal markets told a happier story. Our both-teams-to-score calls landed in 30 of 47, and our over 2.5 goals calls landed in 27 of 47. Several of the games we have picked out below were exactly the kind of open, goal-laden contests those markets are built to anticipate, even when the model had the wrong team winning them. Season-to-date, our winner accuracy across all leagues stands at 47%, so the day sits right around our usual standard.

Two were comfortable hits, and four were misses, some of them instructive. We will go through each one candidly, then round up the rest of the day's results from around Europe and beyond.

Bundesliga (Germany)

VfB Stuttgart 2-3 Borussia Mönchengladbach

Final score
2-3
Winner call
Model called VfB Stuttgart win (67%) · Result: Borussia Mönchengladbach won · MISS
Predicted score
2-0
Both teams to score
Model called Yes (56%) · Actual: Yes · HIT
Over 2.5 goals
Model called Yes (64%) · Actual: Yes · HIT

Stuttgart 2-3 Mönchengladbach was the sort of result that makes a forecaster wince. Our model made VfB Stuttgart clear favourites: a 67% win chance, with 18% for the draw and just 14% for Borussia Mönchengladbach. Expected goals were 2.31 to 0.97, and the most likely scoreline was 2-0, at 10%. The visitors winning was, by our numbers, a fairly unlikely outcome, and our winner call was a MISS.

The script began as expected. Ermedin Demirović put Stuttgart in front in the 5th minute, with Dženan Pejčinović providing the assist, and for a while the model's picture looked sound. Then Mönchengladbach found their rhythm. Joe Scally levelled in the 38th minute, set up by Zento Uno, and the visitors went into the second half with the match all square.

After the break, the favourites came undone. Shuto Machino put Mönchengladbach ahead in the 51st minute, assisted by Wael Mohya, and Mohya then scored himself in the 66th, with Scally supplying the pass. That made it 1-3, a long way from the 2-0 our model had pencilled in as its most likely outcome.

Demirović pulled one back in the 83rd minute, assisted by Jamie Leweling, but it was not enough to rescue a point. Our other markets fared better: we called both teams to score at 56% and over 2.5 goals at 64%, and both were HITS. So the model read the match as a lively one, just not the winner. The winner call at 67% was a MISS, and a reminder that a strong favourite still loses regularly.

Full match analysis, xG and correct-score grid β†’

Serie A (Italy)

Torino 3-2 Monza

Final score
3-2
Winner call
Model called Torino win (57%) · Result: Torino won · HIT
Predicted score
2-1
Both teams to score
Model called Yes (59%) · Actual: Yes · HIT
Over 2.5 goals
Model called Yes (61%) · Actual: Yes · HIT

Torino 3-2 Monza was a win for the side our model fancied, and a rare clean sweep on the day. We had Torino at 57% to win, with the draw and a Monza win both at 21%. Expected goals were 2.02 to 1.15, and the most likely scoreline was 2-1, at 10%. The final score of 3-2 was one goal different from that modal call, and our winner call was a HIT.

The events that shaped the game began in the 39th minute, when Kian Fitz-Jim scored for Torino, with Giovanni Simeone credited with the assist. Monza responded early in the second half: Gustavo Varela scored in the 47th minute to keep the visitors in the contest, and the match stayed open from there.

Torino then pushed on. Rafik Belghali scored in the 61st minute, assisted by Gvidas Gineitis, and Nikola Vlašić added another in the 72nd. Monza were still in it, though, and the game went deep into stoppage time. In the 90+4th minute there was an own goal from Nathan Patterson of Monza, the last event of a frantic finish.

On the goal markets we were also right. Our model gave 59% for both teams to score and called yes, which was a HIT. It put over 2.5 goals at 61% and called yes, also a HIT. A match in which the favourite wins and the goals flow is exactly what the Poisson machinery hopes for, so this was a satisfying one to file under correct.

Full match analysis, xG and correct-score grid β†’

Ligue 1 (France)

Stade Brestois 29 1-2 Nice

Final score
1-2
Winner call
Model called Stade Brestois 29 win (62%) · Result: Nice won · MISS
Predicted score
2-0
Both teams to score
Model called Yes (55%) · Actual: Yes · HIT
Over 2.5 goals
Model called Yes (59%) · Actual: Yes · HIT

Stade Brestois 29 1-2 Nice was another favourite undone. Expected goals were 2.06 to 0.99, and the most likely scoreline was 2-0, at 10%. Nice won it, so the winner call was a MISS, and the margin for error on a 62% call is exactly this sort of afternoon.

Brest did lead. Ludovic Ajorque scored in the 15th minute, and for a long stretch the match looked as if it was following our model's script. Nice had other ideas. Sepe Elye Wahi equalised in the 61st minute, with Mohamed Amoura providing the assist, and the game turned on that goal.

The decisive moment came in the 83rd minute. Sofiane Diop put Nice ahead, set up by Wahi, to make it 1-2. It was the last event listed, and it meant there would be no late equaliser from the spot.

Our goal markets again held up. We put both teams to score at 55% and called yes, a HIT, and over 2.5 goals at 59%, also called yes and also a HIT. So the model understood that this would be a contest with goals at both ends, just not who would come out on top. The winner call at 62% was the miss, and we own it.

Full match analysis, xG and correct-score grid β†’

Premier League (England)

Ipswich 1-1 Nottingham Forest

Final score
1-1
Winner call
Model called Nottingham Forest win (39%) · Result: draw · MISS
Predicted score
1-1
Both teams to score
Model called Yes (62%) · Actual: Yes · HIT
Over 2.5 goals
Model called Yes (60%) · Actual: No · MISS

Ipswich 1-1 Nottingham Forest was the closest call of our six, and the one where our model was least sure. It gave Ipswich a 36% chance, the draw 25% and Nottingham Forest 39%. Expected goals were 1.51 to 1.58, and the most likely scoreline was 1-1, at 11%. The scoreline we flagged as most likely was exactly what happened, which is a small consolation.

The winner call is another matter. We leaned towards Nottingham Forest at 39%, the draw arrived, and so that call goes down as a MISS. It was a thin lean, barely ahead of Ipswich, but a miss is a miss, and we are not going to hide behind the slimness of the margin.

As for the story, Neco Williams gave Nottingham Forest the lead in the 13th minute, assisted by Daniel Muñoz. Ipswich took until the 41st minute to respond, when Julio Enciso scored the equaliser. There were no further goals listed after that.

The goal markets split. We gave 62% for both teams to score and called yes, a HIT. We had over 2.5 goals at 60% and called yes, but the match did not deliver enough goals, so that was a MISS. A game that finished level with a modal 1-1 prediction is, in a sense, a model vindicated on shape but not on the call it actually made. We will take the partial credit.

Full match analysis, xG and correct-score grid β†’

Championship (England)

Bolton 1-5 Middlesbrough

Final score
1-5
Winner call
Model called Middlesbrough win (54%) · Result: Middlesbrough won · HIT
Predicted score
1-1
Both teams to score
Model called Yes (55%) · Actual: Yes · HIT
Over 2.5 goals
Model called Yes (55%) · Actual: Yes · HIT

Bolton 1-5 Middlesbrough was the most lopsided of our featured matches, and our model had the right side on top. It gave Middlesbrough a 54% chance of winning, with Bolton at 23% and the draw at 23%. Expected goals were 1.08 to 1.79, and the most likely scoreline was 1-1, at 11%. The winner call was a HIT, although the size of the victory was far beyond anything we had pencilled in.

Middlesbrough set the tone early. Aidan Morris scored in the 13th minute, with Amario Cozier-Duberry the creator, and Will Lankshear made it two in the 31st minute, set up by Tommy Conway. In the 39th minute there was an own goal from Lewis Brunt of Middlesbrough, a strange moment in an otherwise one-sided first half.

Bolton got something back just before the break. Thelonius Bair scored in the 45+2nd minute, assisted by Gaizka Larrazabal, which at least gave the home side a goal to take into the interval.

The second half belonged to Middlesbrough. Cozier-Duberry scored in the 73rd minute, assisted by Riley McGree, and Lankshear added his second in the 78th, with McGree again the provider. It finished 1-5. Our goal markets were both HITS: both teams to score at 55%, called yes, and over 2.5 goals at 55%, called yes. The model had the direction right, but the scale of the win was well beyond its central expectation.

Full match analysis, xG and correct-score grid β†’

La Liga (Spain)

Alaves 1-1 Malaga

Final score
1-1
Winner call
Model called Alaves win (63%) · Result: draw · MISS
Predicted score
1-0
Both teams to score
Model called No (48%) · Actual: Yes · MISS
Over 2.5 goals
Model called Yes (51%) · Actual: No · MISS

Alaves 1-1 Malaga was a thoroughly miserable evening for the model, with every market we called going the wrong way. We made Alaves strong favourites at 63%, with the draw at 24% and Malaga at only 13%. Expected goals were 1.92 to 0.79, and the most likely scoreline was 1-0, at 13%. The draw cost us the winner call, which was a MISS.

The goal markets were not kind either. Our both-teams-to-score probability was 48%, so we called no, but both sides scored, which makes it a MISS. We put over 2.5 goals at 51% and called yes, but the match fell short of that line, so that was a MISS too. It is a rare clean sweep of errors, and it is only fair to say so.

The match itself ran largely against our expectation. Malaga took the lead in the 44th minute through Adrian Nino, with Ramón Enríquez credited with the assist, which put Alaves behind at the break. That went against a model that had barely given the visitors a chance.

Alaves did eventually respond, though it took until the 89th minute. Aitor Mañas scored the equaliser to make it 1-1, and that was the last goal listed. Expected goals favoured Alaves by a wide margin, so a late equaliser rather than a win hints that the model's read of the balance of play was not entirely off. It simply did not turn into the result we had called.

Full match analysis, xG and correct-score grid β†’

Elsewhere

Beyond the six featured games, the day's other results were a mixed bag for our model, though several favourites did what we expected. In the Eredivisie, GO Ahead Eagles beat Telstar 3-2, and in the Primeira Liga, Vitória SC beat Maritimo 5-0, both matching our winner calls. In the Ukrainian Premier League, Dynamo Kyiv beat Kudrivka 3-1, Polessya beat Obolon'-Brovar 3-0, and Kolos Kovalivka won 2-0 at Veres Rivne, all as we had called. In Saudi Arabia, Al-Nassr won 3-1 at Al-Faisaly FC and Al-Ittihad FC beat Al Diriyah 2-0, again correct calls.

The biggest scoreline of the day came in Sweden's Superettan, where Oddevold beat GIF Sundsvall 7-0, a HIT for us at 69% on the winner. We also had hits in Austria, where SCR Altach beat Austria Vienna 2-1, and in Ligue 2, where Nancy beat Laval 1-0 and PAU beat Grenoble 2-1.

The misses came in clusters, and drawn games were a large part of the story. Motor Lublin drew 2-2 with Zaglebie Lubin, VfL Bochum drew 1-1 with VfL Wolfsburg, SV Darmstadt 98 and FC St. Pauli played out a 0-0, IF Elfsborg and Malmo FF also drew 0-0, and Dunkerque drew 1-1 with Metz. In each of those, our model had backed the home side or Wolfsburg and the draw spoiled it.

There were also straight upsets. Akhmat lost 1-0 to FC Orenburg, Vasteras SK FK lost 2-1 to Halmstad, Alanyaspor lost 2-1 to Kocaelispor, and Clermont Foot lost 2-0 to Guingamp. These were mostly games where our favourite held only a modest probability, which is a useful reminder that a call at 39% to 45% is a lean rather than a conviction. A further batch of results we have not listed here are all counted in the day's totals.

So what did Friday 23 October tell us? Our winner calls were correct in 22 of 47, which is a fair reflection of how hard it is to name winners at football's busiest hours, and the season-to-date figure of 47% shows the day was broadly in line with form. The goal markets were the bright spot, with 30 of 47 on both teams to score and 27 of 47 on over 2.5 goals. The model tends to read the shape of a match better than its winner, and several of today's upsets still came with plenty of goals. If you want to dig into the numbers behind any of these games, the match pages on Fixture360 carry the full probabilities, and our model accuracy record keeps a running, honest tally of hits and misses.

← Previous: Football Results & Model Review Wednesday 21 October 2026: Brugué's double for Levante, plus misses at PK-35 and Reading

This post was written with AI assistance from Fixture360's own model data β€” every figure, score and event comes from our database, and the stat cards are generated directly from it. Predictions are statistical estimates for entertainment and information only, not betting advice. Please gamble responsibly (18+).