Pre-match probabilities
Every fixture in the leagues we cover, as soon as it enters the three-week window ahead.
- Home win, draw, away win
- Goal totals, both sides scoring
- Most likely correct scores
Available
Media & newsroomsThe number that makes the story,
The number that makes the story,
before kick-off.
Who will win the title, which match will be decided by the finest of margins, which side the table flatters: we model twenty-four competitions and deliver dated, verified probabilities and publication-ready graphics to your newsroom.
Real figures, recomputed matchday by matchdayPublic calibrationCredit “Rondo Research”
Who will be champions?
Ligue 1 2026/2027 · title probability, 50,000 simulated seasons
Research
Paris Saint Germain6th · 8 pts70%
Monaco1st · 13 pts14%
Olympique Lyonnais2nd · 11 pts5.0%
LOSC Lille4th · 10 pts4.9%
Rennes5th · 10 pts2.6%
Paris3rd · 11 pts1.2%The other 12 clubs combined: 2.4%
24Competitions modelled
14Leagues projected
50,000Seasons simulated per league
2,094Calibration matches
What you publish
Six ways to feed a football section.
Everyone has the table. What your readers are missing is what happens next, and how likely it is. Everything comes from the same model, published and verified.
Season projections
Who will win the title, who qualifies for Europe, who goes down: 50,000 seasons simulated over the remaining fixtures.
- Title, European qualification and relegation probabilities
- Projected points, position by position
- Leagues with a standard format
Available
Power rankings
Each team’s true strength, adjusted for the quality of its opponents — and the gap to the league table.
- Measured attack and defence
- Who is overperforming, who is underperforming
- The angle the table does not give you
Available
Champions League simulation
Every club’s path from the league phase to the final, simulated on the competition’s actual format.
- Top eight, play-offs, elimination
- Round of 16, quarter-finals, semi-finals, title
- Updated matchday by matchday
Available
Data stories
Data-led angles written by our team, ready for your journalists to run as they are or develop further.
- The figure, its context, its limits
- For a matchday, a derby, a run-in
- Sourced and dated
On request
Expert commentary
A modeller on hand to explain a figure — in an interview, on air or while you prepare a piece.
- What the model says, and what it does not
- The methodology in plain language
- Quotable as “Rondo Research”
On request
Live from the model
What you could have published this morning.
These graphics are not mock-ups: they are generated from the latest published model run and change with it. Under each one, the headline it lets you write.
The title race
Premier League 2026/2027 · title probability, 50,000 simulated seasons
Research
Manchester City1st · 15 pts58%
Arsenal2nd · 12 pts31%
Brighton & Hove Albion3rd · 10 pts6.7%
Liverpool6th · 9 pts1.6%
Brentford4th · 9 pts1.0%The other 15 clubs combined: 1.6%
Possible headline: “Manchester City given a 58% chance of the title after 5 matchdays”.
Too close to call
Ligue 1 · Matchday 6 · Saturday 10 October, 20:45
Research
Lorient12th in the table
Paris3rd in the table35%27%38%
Lorient winDrawParis win
1.3–1.4Expected goals
51%Three or more goals
55%Both teams score
Most likely scores1-113%0-18.7%1-28.5%
Possible headline: “Lorient–Paris: even the favourites, Paris, are no better than 38%”.
The table lies
Ligue 1 · table position against strength measured by the model
Research
Stronger than their position
Olympique Marseille17th in the table→4th on strength▲ 13
Lens15th in the table→3rd on strength▲ 12Paris Saint Germain6th in the table→1st on strength▲ 5Higher than their level
Angers SCO7th in the table→17th on strength▼ 10
Le Mans9th in the table→16th on strength▼ 7Paris3rd in the table→9th on strength▼ 6Possible headline: “Olympique Marseille: 17th in the table, 4th on strength”.
Formats
Delivered the way your newsroom works.
A graphic for print, a file for the subs’ desk, a feed for the website. The status of each format is stated: we do not sell what does not exist yet.
Graphics in your house style
The graphics on this page, adapted to your brand: print, web, social media, on-air graphics.
On request
Weekly data pack
Ahead of every matchday: probabilities for every fixture, updated projections and three written angles. A PDF and its source files.
On request
Data feeds
The same calculations in JSON: projected table, strengths, fixtures, calibration. For your competition pages and your apps.
On request
Embeddable widgets
A block to drop into an article that updates itself after every model run. Tell us which formats you need.
On request
How it works
From the latest matchday to your page.
A Dixon-Coles-type goals model, re-estimated after every matchday played. Nothing is adjusted by hand: what comes out of the model is what you publish.
01
The matches are played
The model re-estimates every team’s attack and defence from results across the whole competition.
02
The model simulates
Each fixture gets its own score grid; the rest of the season is replayed 50,000 times.
03
The model checks itself
Calibration — forecast against outcome — is recomputed and published as it stands.
04
You receive
Files, graphics or feeds, dated and credited “Source: Rondo Research”.
Why cite us
A figure you can stand behind.
- The methodology is public. Team strengths, score matrix, simulations: every step is described, so your readers can judge it for themselves.
- Calibration is published. When we say 30%, it should happen three times in ten — and the chart alongside shows that it does, across 2,094 matches.
- Every figure is dated. A graphic always carries the date of the model run it comes from: no out-of-date figure picked up by mistake.
- The scope is stated. Domestic cups are not modelled, and some leagues have no projection: we say so rather than publish a wrong figure.
Each point is a probability band; its size is the number of cases. On the diagonal, the model keeps its word. Here, across 2,094 matches (1X2 result), the mean gap between forecast and observed is 2.3%.
By type of media
One model, five ways to tell it.
Newspapers and magazinesDailies, weeklies, monthlies
The Saturday preview, the mid-season projection, the “who goes down” feature. A full-column graphic and a figure for the headline.
Example: each contender’s title probability, opening the sports section.
Television and radioPreviews, studio shows, regular slots
A figure the presenter can say in one sentence, and on-screen graphics that show it. Our modellers can explain it in the studio.
Example: the probability of each outcome, summed up in one line before going live to the ground.
Websites and appsCompetition pages, live coverage, match pages
Probabilities on every match page and a projected table on every league page, powered by a data feed.
Example: a “what the model says” module under every fixture.
NewslettersMorning briefings, themed newsletters
A graphic of the week and three lines of analysis, ready to paste in. The data pack is built for that rhythm.
Example: “the table is lying” every Monday, after the matchday.
Podcasts and videoShows, short formats
Angles to fuel a discussion, and a guest who can explain why the model thinks what it thinks.
Example: the most underrated team in the league, with the numbers to back it up.
A matchday, a derby or a run-in to cover?
Tell us the competition, the angle and the deadline. We will come back to you with the figures and a sample graphic, sourced and ready to publish.