The record
Two records, both public, nothing deleted. What we stake: the singles, in units. What the model says: every call it publishes, graded.
Last updated 28 Sep 2026, 17:53
The model's calls, graded
Information, not what we stake. Every call the model publishes on a league on the record, locked before kick-off. A hit rate is not a profit: a short price needs a high one. That is why the singles we stake are a stricter set.
Monthly breakdown
Wins and losses, month by month. Tap a bar for the count.
Breakdown by market
Every market we have graded, with its hit rate on the record.
Markets with fewer than 10 verified picks hidden. Too small a sample to draw conclusions.
Calibration by confidence
For each probability band: what the model predicted vs what actually happened. Close to zero = well-calibrated.
Green = within 5pp (calibrated). Gold = over-delivering (actual beat the prediction). Red = under-delivering. Small samples are noisy. Weigh bands with more picks more heavily.
Breakdown by league
The same record, split by competition.
The model's next calls Locked before kick-off
Already locked into the record above, timestamped before their games kick off. Whatever happens, they grade in public. Every call, priced
Full pick history
Every graded pick, newest first. Filter by market, league, result or date.
| Date | Match | League | Pick | Odds | Result |
|---|---|---|---|---|---|
| 26 Sep |
Oldham v Salford City
1-4
|
League Two |
Over 1.5 goals
80%+
|
1.17 |
Won
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 8/12 of the sides' recent games (home 5/6 · away 3/6).
Landed, exactly as the model read it. It finished 1-4, 5 goals, clear of the 1.5 line.
|
| 26 Sep |
Cambridge United v AFC Wimbledon
1-1
|
League One |
Double chance: home or draw
80%+
|
1.11 |
Won
The model made Cambridge United or the draw a strong 80%+ read before a ball was kicked. The receipts: home unbeaten in 5/6 · away lost 2/6.
Landed, exactly as the model read it. It finished 1-1.
|
| 26 Sep |
Plymouth v Burton Albion
0-1
|
League One |
Over 1.5 goals
80%+
|
1.16 |
Lost
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 12/12 of the sides' recent games (home 6/6 · away 6/6).
Missed. One of the ~1-in-5 that a 80%+ call still loses, and we log every one. It finished 0-1, only one goal, short of the 1.5 line.
|
| 26 Sep |
Cambridge United v AFC Wimbledon
1-1
|
League One |
Over 1.5 goals
75%
|
1.30 |
Won
The model made Over 1.5 goals a strong 75% read before a ball was kicked. The receipts: over 1.5 agreed in 9/12 of the sides' recent games (home 6/6 · away 3/6).
Landed. The model's lean came good. It finished 1-1, 2 goals, clear of the 1.5 line.
|
| 26 Sep |
Wycombe v Reading
2-2
|
League One |
Over 1.5 goals
80%+
|
1.16 |
Won
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 11/12 of the sides' recent games (home 5/6 · away 6/6).
Landed, exactly as the model read it. It finished 2-2, 4 goals, clear of the 1.5 line.
|
| 26 Sep |
Rotherham v Crewe
1-1
|
League Two |
Double chance: home or draw
75%
|
1.28 |
Won
The model made Rotherham or the draw a strong 75% read before a ball was kicked. The receipts: home unbeaten in 5/6 · away lost 0/6.
Landed. The model's lean came good. It finished 1-1.
|
| 26 Sep |
Rotherham v Crewe
1-1
|
League Two |
Over 1.5 goals
80%+
|
1.16 |
Won
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 9/12 of the sides' recent games (home 4/6 · away 5/6).
Landed, exactly as the model read it. It finished 1-1, 2 goals, clear of the 1.5 line.
|
| 26 Sep |
Fleetwood Town v Rochdale
1-2
|
League Two |
Over 1.5 goals
77%
|
1.22 |
Won
The model made Over 1.5 goals a strong 77% read before a ball was kicked. The receipts: over 1.5 agreed in 9/12 of the sides' recent games (home 5/6 · away 4/6).
Landed. The model's lean came good. It finished 1-2, 3 goals, clear of the 1.5 line.
|
| 26 Sep |
Bristol Rovers v Exeter City
1-0
|
League Two |
Over 1.5 goals
80%+
|
1.17 |
Lost
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 8/12 of the sides' recent games (home 5/6 · away 3/6).
Missed. One of the ~1-in-5 that a 80%+ call still loses, and we log every one. It finished 1-0, only one goal, short of the 1.5 line.
|
| 26 Sep |
Fleetwood Town v Rochdale
1-2
|
League Two |
Double chance: home or draw
80%+
|
1.15 |
Lost
The model made Fleetwood Town or the draw a strong 80%+ read before a ball was kicked. The receipts: home unbeaten in 4/6 · away lost 3/6.
Missed. One of the ~1-in-6 that a 80%+ call still loses, and we log every one. It finished 1-2.
|
| 26 Sep |
Cheltenham v Chesterfield
2-1
|
League Two |
Over 1.5 goals
80%+
|
1.17 |
Won
The model made Over 1.5 goals a strong 80%+ read before a ball was kicked. The receipts: over 1.5 agreed in 12/12 of the sides' recent games (home 6/6 · away 6/6).
Landed, exactly as the model read it. It finished 2-1, 3 goals, clear of the 1.5 line.
|
| 21 Sep |
Valencia v Real Sociedad
2-3
|
La Liga |
Over 1.5 goals
76%
|
1.27 |
Won
The model made Over 1.5 goals a strong 76% read before a ball was kicked. The receipts: over 1.5 agreed in 7/12 of the sides' recent games (home 2/6 · away 5/6).
Landed. The model's lean came good. It finished 2-3, 5 goals, clear of the 1.5 line.
|
| 20 Sep |
Deportivo La Coruna v Real Betis
1-1
|
La Liga |
Over 1.5 goals
77%
|
1.21 |
Won
The model made Over 1.5 goals a strong 77% read before a ball was kicked. The receipts: over 1.5 agreed in 8/12 of the sides' recent games (home 5/6 · away 3/6).
Landed. The model's lean came good. It finished 1-1, 2 goals, clear of the 1.5 line.
|
| 20 Sep |
Juventus v Atalanta
2-0
|
Serie A |
Over 1.5 goals
77%
|
1.26 |
Won
The model made Over 1.5 goals a strong 77% read before a ball was kicked. The receipts: over 1.5 agreed in 8/12 of the sides' recent games (home 5/6 · away 3/6).
Landed. The model's lean came good. It finished 2-0, 2 goals, clear of the 1.5 line.
|
| 20 Sep |
Leeds v Crystal Palace
0-0
|
Premier League |
Over 1.5 goals
79%
|
1.18 |
Lost
The model made Over 1.5 goals a strong 79% read before a ball was kicked. The receipts: over 1.5 agreed in 11/12 of the sides' recent games (home 5/6 · away 6/6).
Missed. A lean, not a lock, and the game broke the other way. It finished 0-0, only 0 goals, short of the 1.5 line.
|
| 20 Sep |
Auxerre v Brest
2-1
|
Ligue 1 |
Over 1.5 goals
77%
|
1.21 |
Won
The model made Over 1.5 goals a strong 77% read before a ball was kicked. The receipts: over 1.5 agreed in 9/12 of the sides' recent games (home 4/6 · away 5/6).
Landed. The model's lean came good. It finished 2-1, 3 goals, clear of the 1.5 line.
|
| 20 Sep |
Fiorentina v Napoli
1-1
|
Serie A |
Over 1.5 goals
75%
|
1.25 |
Won
The model made Over 1.5 goals a strong 75% read before a ball was kicked. The receipts: over 1.5 agreed in 10/12 of the sides' recent games (home 6/6 · away 4/6).
Landed. The model's lean came good. It finished 1-1, 2 goals, clear of the 1.5 line.
|
| 19 Sep |
Celta Vigo v Racing Santander
5-0
|
La Liga |
Over 1.5 goals
79%
|
1.18 |
Won
The model made Over 1.5 goals a strong 79% read before a ball was kicked. The receipts: over 1.5 agreed in 9/12 of the sides' recent games (home 4/6 · away 5/6).
Landed. The model's lean came good. It finished 5-0, 5 goals, clear of the 1.5 line.
|
| 19 Sep |
Athletic Club v Alaves
0-0
|
La Liga |
Over 1.5 goals
76%
|
1.26 |
Lost
The model made Over 1.5 goals a strong 76% read before a ball was kicked. The receipts: over 1.5 agreed in 10/12 of the sides' recent games (home 6/6 · away 4/6).
Missed. A lean, not a lock, and the game broke the other way. It finished 0-0, only 0 goals, short of the 1.5 line.
|
| 19 Sep |
Paris FC v Strasbourg
2-1
|
Ligue 1 |
Over 1.5 goals
79%
|
1.17 |
Won
The model made Over 1.5 goals a strong 79% read before a ball was kicked. The receipts: over 1.5 agreed in 11/12 of the sides' recent games (home 5/6 · away 6/6).
Landed. The model's lean came good. It finished 2-1, 3 goals, clear of the 1.5 line.
|
Disclaimer: Odds shown are the best available at the time the prediction was recorded. Past performance does not guarantee future results. Please gamble responsibly. 18+ only. T&Cs apply.
A note from 20 August: nine MLS double-chance picks, three won and six lost, had been published under an automatic promotion. When promotions became a human decision they were moved out of this headline count. They stay graded in the full history above.
The singles we'd pick ourselves, with the research and the price behind each one
- The daily singles. Each one cleared three checks: the line landed in at least 9 of the two sides' last 12 games, a UK price of 1.35 or more, and our model's number clears what the price needs. Usually 0 to 5 a day, one unit each.
- The case for each, put together. The last 12 against what the price needs, who is missing, the head to head, and the best price at a UK bookmaker, in one place.
- The desk's card on every fixture it covers. Its verdict, every market graded, and the reason written down. And tomorrow's singles the night before.
Every pick published before kick-off, graded in public. The losing ones stay on the page.