FootballBetting

Tools & Analytics · Updated

The Scenario Machine

We play the rest of the season out 10,000 times, in all four English divisions. Title, promotion, play-off and relegation chances for every club, with the projected points to go with them, and the same model prices every match along the way.

Arsenal: title
Man City: title
Crystal Palace: relegation

The model's clearest three calls in the Premier League, out of 10,000 simulated seasons. Every club's full set of chances is in the table below: these are simply the ones it is most confident about.

Season Simulation · Live Scenario Machine

10,000 simulated seasons · 2 games played, 36 to go per club

# Club Pts Proj pts TitleTop 4Rel
1 Arsenal 6 73 51% 95%
2 Man City 6 71 39% 91%
3 Man Utd 3 58 3% 36% 2%
4 Chelsea 6 56 2% 28% 3%
5 Brentford 4 54 0.9% 17% 6%
6 Liverpool 2 54 0.7% 18% 6%
7 Coventry 0 53 0.8% 17% 7%
8 Brighton 3 53 0.7% 17% 6%
9 Ipswich 3 53 0.7% 16% 7%
10 Bournemouth 1 51 0.3% 10% 12%
11 Hull 6 51 0.3% 10% 12%
12 Everton 4 50 0.3% 9% 13%
13 Aston Villa 0 50 0.2% 8% 15%
14 Newcastle 4 50 0.2% 9% 12%
15 Leeds 4 49 0.1% 6% 19%
16 Sunderland 3 48 0.2% 5% 22%
17 Nott'm Forest 1 47 0.1% 4% 27%
18 Fulham 0 45 2% 32%
19 Crystal Palace 0 42 0.9% 51%
20 Spurs 0 42 1% 49%

Percentages are how often a club finished in that zone across the simulated seasons. Proj is its average final points total. Shaded rows mark promotion or European places, the play-offs and the relegation zone.

Why Not "What Does My Club Need" in August?

Because in August the answer is "anything". There are 380 Premier League matches still to play, and with three outcomes each the number of possible seasons runs past 180 digits. Every club can still reach almost any total. A tool that says your side needs "roughly 85 points, by any route" has told you nothing at all.

So for most of the season this page answers the question that can be answered: how often does each club actually end up in each zone? We simulate the remaining fixtures 10,000 times, match by match, and count the finishes. As real results come in the table firms up and the percentages move with it.

Then, once roughly eight games remain, the permutations have collapsed far enough for exact answers, and the page switches to the run-in solver. That is when "four points from the last two games and the title is certain" becomes a fact rather than a guess: and when a dead rubber can be told apart from a decider.

All Four Divisions, Not Just the Premier League

The Premier League has two races: the title and the top four. The EFL has four in every table: automatic promotion, the play-offs, mid-table safety and relegation, which is why the Championship, League One and League Two projections are often the better read.

The zones follow the real rules rather than a template. The Championship and League One send two clubs up automatically with four more into the play-offs; League Two sends three up and takes four into the play-offs. Relegation is three clubs from the Premier League and the Championship, four from League One and two from League Two.

Every division refreshes on the same rebuild, so these projections never disagree with the tables on our league pages.

What the Simulation Cannot See

It reads goals, and only goals. It has no idea who has been bought, who is injured, which club banked parachute payments or who changed manager in June. A club relegated from the Premier League after a poor season stays rated on that poor season until it starts scoring again: bookmakers will rate it higher, and sometimes they will be right.

It also treats last season as evidence rather than gospel. Every rating is pulled part of the way back toward its division average, and a club that changed division is pulled harder still, because a record built against different opponents says less about the new company it keeps. As real results arrive they take over the rating: by the spring the current season is doing most of the work, so a club that has been transformed since August is judged on August onwards, not on last May. Without that step a dominant lower-division champion would project as a title favourite one tier up, which the history of English football does not support.

So read it as a starting price, not a forecast. The number worth acting on is the gap between our percentage and the bookmaker's: and closing that gap is a judgement you make with the team news the model has never seen.

Frequently Asked Questions

What does the scenario machine actually calculate?

It plays the rest of the season out 10,000 times in each of the four English divisions, then counts how often every club finishes as champion, in the automatic promotion places, in the play-offs or in the relegation zone.

Each match along the way is priced by the same model behind our football predictions.

Where do the ratings come from?

With no games played there is no table to read, so every club starts rated on the season it just played: including clubs that changed division, who keep their real record adjusted by one division step. Once the season is under way the current campaign takes over: it carries about half the weight after a dozen games and most of it by the spring, which is also how the promotion adjustment retires itself.

The method is set out on our how we rate page.

Why is my club rated lower than the bookmakers rate it?

Because the model reads results and nothing else. It knows no transfer fees, no parachute payments and no squad valuations: a relegated club that scored badly stays rated badly until it scores again.

That gap is sometimes where value hides. Compare prices at the best betting sites.

When does the "what does my club need" solver appear?

In the run-in. Once about eight games remain the permutations collapse far enough for exact statements, four points from two games, or a rival needing to win twice.

Until then the projection above is the useful view, alongside this weekend's betting tips.

Do the percentages add up?

Yes, and it is the check worth making. There is one title, so that column totals 100%; four play-off places total 400%, three relegation places 300%.

We publish the workings rather than a verdict, the Prognosis Engine does the same for a single match.