Issue #3 · week 2026-W38 · simulations updated 13 September 2026 · 10,000 season simulations per league
The TACTIX Simulation Report is the weekly, numbered record of what our season simulations say — and how they moved. Every number below is the same number the live Supercomputer hub shows; the deltas compare this run with the previous one.
Three Lines You Can Quote
- Arsenal won the Premier League title in 58.3% of 10,000 simulations of the 2026/27 season. (TACTIX Simulation Report #3)
- The gap at the top: Arsenal 58.3% vs Manchester City 41.4% — a 16.9-point difference in title likelihood. (TACTIX Simulation Report #3)
- 2 of 20 Premier League clubs took the title in at least 1% of simulations. (TACTIX Simulation Report #3)
Premier League 2026/27 — The Full Movement Table
| # | Club | Title | Δ | Top 4 | Relegation | Avg pts |
|---|---|---|---|---|---|---|
| 1 | Arsenal | 58.3% | +650.0pp | 99.9% | 0% | 86.3 |
| 2 | Manchester City | 41.4% | −630.0pp | 99.8% | 0% | 83.9 |
| 3 | Manchester United | 0.2% | −20.0pp | 59.1% | <0.1% | 63.5 |
| 4 | Brighton | <0.1% | · | 31.1% | 0.3% | 58.2 |
| 5 | Brentford | <0.1% | · | 24.9% | 0.4% | 56.9 |
| 6 | Liverpool | <0.1% | · | 24.2% | 0.4% | 56.8 |
| 7 | Chelsea | 0% | · | 15.2% | 1.2% | 54.4 |
| 8 | Newcastle | 0% | · | 9.8% | 2.3% | 52.3 |
| 9 | Leeds | <0.1% | · | 8.7% | 2.2% | 52.1 |
| 10 | Bournemouth | 0% | · | 8.8% | 2.4% | 52.1 |
| 11 | Aston Villa | 0% | · | 6.8% | 3.4% | 50.6 |
| 12 | Everton | 0% | · | 5.1% | 3.6% | 49.8 |
| 13 | Nottingham Forest | 0% | · | 4.5% | 4.2% | 49.6 |
| 14 | Sunderland | 0% | · | 0.7% | 16.8% | 43.8 |
| 15 | Fulham | 0% | · | 1.0% | 17.7% | 43.6 |
| 16 | Hull City | 0% | · | 0.1% | 32.0% | 40.3 |
| 17 | Crystal Palace | 0% | · | <0.1% | 46.3% | 37.9 |
| 18 | Ipswich | 0% | · | 0.1% | 48.5% | 37.7 |
| 19 | Tottenham | 0% | · | <0.1% | 49.9% | 37.2 |
| 20 | Coventry | 0% | · | <0.1% | 68.4% | 34 |
Club by Club
One line per club, for the beat writer who covers exactly one of them:
- Arsenal (1/20 by simulation) — title 58.3% (+650.0pp on the last run), top four 99.9%, averaging 86.3 pts.
- Manchester City (2/20 by simulation) — title 41.4% (−630.0pp on the last run), top four 99.8%, averaging 83.9 pts.
- Manchester United (3/20 by simulation) — top four 59.1% (title −20.0pp), relegation <0.1%, averaging 63.5 pts.
- Brighton (4/20 by simulation) — top four 31.1%, relegation 0.3%, averaging 58.2 pts.
- Brentford (5/20 by simulation) — top four 24.9%, relegation 0.4%, averaging 56.9 pts.
- Liverpool (6/20 by simulation) — top four 24.2%, relegation 0.4%, averaging 56.8 pts.
- Chelsea (7/20 by simulation) — top four 15.2%, relegation 1.2%, averaging 54.4 pts.
- Newcastle (8/20 by simulation) — top four 9.8%, relegation 2.3%, averaging 52.3 pts.
- Leeds (9/20 by simulation) — top four 8.7%, relegation 2.2%, averaging 52.1 pts.
- Bournemouth (10/20 by simulation) — top four 8.8%, relegation 2.4%, averaging 52.1 pts.
- Aston Villa (11/20 by simulation) — top four 6.8%, relegation 3.4%, averaging 50.6 pts.
- Everton (12/20 by simulation) — top four 5.1%, relegation 3.6%, averaging 49.8 pts.
- Nottingham Forest (13/20 by simulation) — top four 4.5%, relegation 4.2%, averaging 49.6 pts.
- Sunderland (14/20 by simulation) — relegation 16.8% (+480.0pp), top-half push 0.7%, averaging 43.8 pts.
- Fulham (15/20 by simulation) — relegation 17.7% (−90.0pp), top-half push 1.0%, averaging 43.6 pts.
- Hull City (16/20 by simulation) — relegation 32.0% (−110.0pp), top-half push 0.1%, averaging 40.3 pts.
- Crystal Palace (17/20 by simulation) — relegation 46.3% (+1180.0pp), top-half push <0.1%, averaging 37.9 pts.
- Ipswich (18/20 by simulation) — relegation 48.5% (−1160.0pp), top-half push 0.1%, averaging 37.7 pts.
- Tottenham (19/20 by simulation) — relegation 49.9% (−100.0pp), top-half push <0.1%, averaging 37.2 pts.
- Coventry (20/20 by simulation) — relegation 68.4% (+40.0pp), top-half push <0.1%, averaging 34 pts.
Movers Elsewhere
- Bundesliga: Wolfsberger AC — top spots 27.3% (−1680.0pp).
- Challenger Pro League: Beerschot VA — top spots 80.4% (−490.0pp).
- Jupiler Pro League: Lommel United — top spots 49.2% (−4280.0pp).
- Série A: Atletico-MG — top spots 28.4% (+1240.0pp).
- Bundesliga: SC Freiburg — top spots 26.1% (+1430.0pp).
- Championship: Southampton — top spots 57.3% (+830.0pp).
- Super League: Beijing Guoan — top spots 56.2% (+230.0pp).
- Superliga: Randers FC — top spots 32.5% (−370.0pp).
- League One: Oxford United — top spots 89.5% (+8950.0pp).
- League Two: York — top spots 87.6% (+6300.0pp).
- National League: Boreham Wood — title 31.4% (−1730.0pp).
- Ligue 2: Saint Etienne — title 46.5% (−1030.0pp).
- 2. Bundesliga: VfL Wolfsburg — top spots 51.3% (+5110.0pp).
- Super League 1: OFI — top spots 36.1% (+1520.0pp).
- Serie B: Catanzaro — top spots 25.4% (+630.0pp).
- La Liga: Racing Santander — top spots 9.8% (+980.0pp).
- Ligue 1: Marseille — top spots 10.9% (−410.0pp).
- Eredivisie: Ajax — top spots 30.9% (+820.0pp).
- Eliteserien: Bodo/Glimt — title 86.8% (−180.0pp).
- Ekstraklasa: Lech Poznan — title 45.8% (−930.0pp).
- Primeira Liga: Sporting CP — title 34.9% (−590.0pp).
- Saudi Pro League: Al-Hilal Saudi FC — title 65.9% (+1700.0pp).
- Serie A: Atalanta — top spots 11.7% (−690.0pp).
- Segunda División: Girona — top spots 30.8% (+3070.0pp).
- Allsvenskan: Sirius — title 57.1% (+120.0pp).
- Süper Lig: Trabzonspor — top spots 10.4% (−600.0pp).
- Premier League: Shakhtar Donetsk — title 73.5% (−290.0pp).
- Cymru Premier: Colwyn Bay — top spots 18.0% (+760.0pp).
Model vs the Market, This Week
The model’s settled hit rate against the bookmaker favourite, from the live accuracy leaderboard (last-12-months window, all leagues, losers included):
| Market | Settled picks | Model | Bookmaker favourite | Model − favourite | Δ model since last issue |
|---|---|---|---|---|---|
| player_cards_yellow_under_0_5 | 8,940 | 94.4% | — | — | +0.3 pp (+1,775 picks) |
| player_tackles_total_under_3_5 | 7,858 | 98.0% | — | — | · |
| clean_sheet | 7,418 | 30.6% | 35.3% | -4.7 pp | -0.4 pp (+2,171 picks) |
Hit rate answers “who picks winners more often” — it is not a profitability claim; probabilities are odds to weigh, not promises. Definitions and every league’s cells: the leaderboard.
Method, in One Paragraph
Each covered league is simulated thousands of times from the current table and remaining fixtures, using the same calibrated match model that prices every prediction on the site. Probabilities are model estimates — odds to weigh, not promises. The live, always-current version of every table on this page is at tactix.football/supercomputer; embeddable versions are free at /widgets with the credit intact. Cite this issue as “TACTIX Simulation Report #3”.
This post is generated from TACTIX’s match data by our software and checked automatically. How it works →
Get today’s free slate
Free, every day: the model’s fixed slate of picks, each with its confidence tier and that tier’s settled record — plus yesterday’s slate, graded. Pro adds the full board, the value edge and the written analysis. No payment card.
Create free account