This isn't tipster content, and it never will be — the whole point of The Bankroll Project is showing the reasoning, not just handing over a selection. That reasoning almost always comes back to value. It's a simple idea underneath, even if the maths looks intimidating at first.
Odds are a claim about probability
Every price on Matchbook Exchange is really just a probability wearing a disguise. Odds of 2.0 imply a 50% chance of that outcome happening. Odds of 4.0 imply 25%. Odds of 1.5 imply roughly 67%. The maths is simple: implied probability equals 1 divided by the decimal odds, as a percentage.
A "value bet" is one where the price on offer implies a lower probability than what you believe the true probability actually is. If you think a team has a 45% chance of winning, and the market is offering odds of 2.5 (which implies only 40%), that gap is the value. You're being paid as though the outcome is less likely than it really is.
Value isn't the same as "a good bet that wins." A value bet can lose — often does, if the true probability was only 45% — and a bad-value bet can still win. Value describes the price, not the outcome. That's the entire reason this challenge tracks bets over a full season rather than judging any single week.
Expected value, worked through with real numbers
Expected value (EV) turns that probability gap into an actual number — what a bet is worth, on average, if you could somehow place it hundreds of times. The formula is:
EV = (probability of winning × profit if won) − (probability of losing × stake)
That £2.50 doesn't mean the bet returns £2.50 — it means that if the true probability estimate is right, backing this exact price repeatedly would average out to £2.50 profit per £20 staked. One bet tells you almost nothing. A season of them, tracked honestly, starts to.
Where the estimate actually comes from
The hard part was never the formula — it's the "true probability" input, and nobody has perfect access to that number. It's built from form, underlying performance data, team news, and how the market itself has moved in the hours before kick-off. It's an estimate, not a certainty, which is exactly why the weekly episodes explain the reasoning behind each number rather than presenting it as fact.
Why an exchange makes value easier to find and keep
Value betting and exchange betting reinforce each other, which is a big part of why this challenge runs on Matchbook rather than a bookmaker. Two reasons in particular:
- Sharper prices. Exchange odds are set by other bettors trading against each other, not a bookmaker's margin — so the implied probabilities tend to sit closer to the true ones, leaving smaller gaps to find but more honest ones.
- Less of the edge lost to fees. A found value bet is worth less if commission eats into it. At 2% on net winnings, Matchbook takes a noticeably smaller cut than a bookmaker's built-in margin, so more of any genuine edge survives through to the bankroll.
Getting the probability estimate wrong is the single biggest risk in value betting — and it happens constantly, even to good judgement. That's precisely why results are shown here every single week rather than cherry-picked at season's end.
None of this is a system that guarantees winning bets — it's a way of thinking about which bets are worth making in the first place. Whether it actually adds up to £10,000 by the end of the season is the entire premise of this project, tracked openly on the bets page as it happens.