Football, modelled — and explained
How the numbers behind a match actually work: xG, scoreline distributions, calibration, and reading a fixture the way the model does. No tips — the thinking.
How Momus uses the bookmaker's odds
Bookmakers spend a fortune pricing every match — so we don't ignore them. Here's exactly how Modal folds the odds into the model: de-vigged, used as an anchor when data is thin, and treated as the thing to beat.
Everything Momus Modal does, in one place
A quant model reads every match, grades itself in the open, and hands you the whole picture — not a tip. Here's the full tour of what you actually get.
The value board: where the model disagrees with the market
The betting market is sharp, so the interesting matches are the ones where a model and the market disagree. That gap is the whole game — here's how to read it.
Reading the receipts: a public, graded track record
An analysis you can't check is just a tip. Here's how we grade every prediction against the real result — in the open — and why calibration matters more than a headline accuracy number.
Model odds vs market odds: what 'value' really means
The edge in football isn't picking winners — it's spotting when the model and the market disagree. Here's how comparing model probability to market odds reveals value, explained plainly.
How to read football match probabilities (without a stats degree)
1X2 probabilities, over/under, BTTS — what the percentages on a match analysis actually mean, and how to read them like an analyst, not a guesser.
Are AI football predictions accurate? An honest answer
Can an AI predict football matches accurately? Yes and no. A good model gives calibrated probabilities, not certainties — here's what 'accurate' actually means and how to judge it.
Your club, every week: the brief a football fan actually wants
Pick your club and the model works for you specifically — one weekly email with your team's next analysed fixture and the freshest transfer news. Here's how it works.
One board, every match: run your matchday like an analyst
An analyst doesn't scroll a feed — they work from a board. The desk pins every fixture, each with the model's read and the edge it sees, so you can scan the day in one screen.
Analysis on demand: pick the match, we'll model it
The board covers the matches we choose. On the Edge tier you choose — name the fixtures that matter to you and the model runs a full read on each. Here's how it works.
AI in football, explained: what it actually does
AI in football isn't one thing. It's tracking data, tactics assistants, injury models and outcome prediction — here's the whole map, in plain language.
What xG really measures (and what it doesn't)
Expected goals (xG) is the most quoted stat in modern football analysis. Here's what it actually tells you, where it misleads, and how a model should use it.
TacticAI: what DeepMind's corner-kick assistant means for football
DeepMind and Liverpool built an AI that reads corner kicks and suggests better setups — experts preferred it 90% of the time. Here's how it works and why it matters.
How we model a match: Dixon-Coles, explained
The Dixon-Coles model is the workhorse behind serious football prediction. Here's how it turns team strength into a full scoreline distribution — in plain language.
How AI and data quietly took over football
From tracking cameras to tactics assistants, AI reshaped football in a decade. Here's what actually changed on the pitch, in the dugout and in the analysis.
The modal scoreline: why the most likely score isn't the likely winner
A team can be a clear favourite while the single most probable scoreline is a draw. Here's the counterintuitive maths behind the modal scoreline — and why it's the honest way to read a match.
Can AI predict the correct score? What a score model really tells you
AI can't call the exact score of a match — but it can do something more useful. Here's what a real correct-score model gives you, and why '70% accuracy' claims don't add up.
AI football predictions: how to tell a real model from a mystery box
Anyone can slap 'AI' on a tips page. Here are the seven questions that separate a genuine football model from a black box selling certainty.
Machine learning vs Dixon-Coles: do you need deep learning to predict football?
Neural networks or a 1997 statistical model — which actually predicts football better? The honest answer is less exciting and more useful than the hype.
AI injury prediction in football: what it can and can't see
Clubs use AI to flag injury risk before it happens. Here's how it works, what it genuinely predicts, and why it will never be a crystal ball for hamstrings.