FootyTipsFootball Value Picks
This board shows where our model most disagrees with the betting market. Divergence is a prompt to investigate, not a betting instruction — and our settled record below shows which bands have actually beaten the closing price.
market outcomes with odds
divergences beyond ±4pp
latest odds snapshot
bookmaker sources
Citable definition: divergence = FootyTips model probability minus de-vigged market implied probability from decimal odds, shown in both directions. Rows only appear when real odds snapshots are imported and at least three books quote a complete 1X2 market.
18+ · Please gamble responsibly. Predictions are probabilistic, not guarantees. Availability of betting content may be restricted in your region.
Biggest Model–Market Divergences
Every 1X2 outcome where we hold both a model probability and a de-vigged market consensus, ranked by how far apart they are — in both directions. Best price shown across 6 crypto sportsbooks. Research input, not staking instructions; odds move and probabilities are not guarantees.
Has this signal actually beaten the market?
We settle this honestly. Across 90 settled matches with usable closing prices, we measure what each divergence band was worth against the market's closing price. Larger model-above-market gaps have not paid — if anything they have been the worst band.
| Divergence band | Matches | Closing-line value | Excess vs market favourite |
|---|---|---|---|
| Market more confident | 22 | +1.17pp | +0.00pp |
| Aligned | 23 | -0.95pp | +0.21pp |
| Model above market | 23 | -1.41pp | -0.02pp |
| Model far above market | 22 | -1.13pp | -0.14pp |
Closing-line value = the best price we showed 24h out, priced against the de-vigged market consensus at kickoff. "Excess" subtracts what you would have got by simply backing the market favourite every time — without that subtraction, normal favourite drift looks like skill. Sample is 90/150 settled matches (median closing snapshot 2.69h before kickoff); bands hold 16–20 matches each, so differences of a few tenths of a point are noise. Closing-line value is only measurable for the model's own top pick, so rows on other outcomes carry no historical evidence.
| Match | Pick | Signal | Model vs market | Edge | Best odds | EV if model is right* | Action | |
|---|---|---|---|---|---|---|---|---|
| Bournemouth vs Liverpool Sun, 20 Sep 2026 13:00 GMT | Liverpool win | Market more confident history: +1.17pp CLV over 22 Confirm near kickoff | 35.3% vs 44.1% | -8.8pp | 2.21 duelbits ↗ | odds value -22% | 1-0 | View analysis → |
| Bournemouth vs Liverpool Sun, 20 Sep 2026 13:00 GMT | Bournemouth win | Model above market history: -1.41pp CLV over 23 Confirm near kickoff | 38.4% vs 30.7% | +7.7pp | 3.15 vave stake 3.1 ↗ | odds value +21% | 1-0 | View analysis → |
| Leeds United vs Crystal Palace Sun, 20 Sep 2026 13:00 GMT | Leeds United win | Market more confident history: +1.17pp CLV over 22 Confirm near kickoff | 50.9% vs 56.0% | -5.1pp | 1.74 vave duelbits 1.72 ↗ | odds value -11% | 1-0 | View analysis → |
| Leeds United vs Crystal Palace Sun, 20 Sep 2026 13:00 GMT | Crystal Palace win | Model above market history: -1.41pp CLV over 23 Confirm near kickoff | 24.4% vs 20.0% | +4.4pp | 4.85 cloudbet ↗ | odds value +19% | 1-0 | View analysis → |
| Fulham vs Manchester United Sun, 20 Sep 2026 15:30 GMT | Manchester United win | Market more confident history: +1.17pp CLV over 22 Confirm near kickoff | 44.2% vs 48.3% | -4.1pp | 2 duelbits ↗ | odds value -12% | 0-1 | View analysis → |
| Manchester City vs Sunderland Sun, 20 Sep 2026 13:00 GMT | Manchester City win | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 75.4% vs 72.4% | +3.0pp | 1.36 vave stake 1.34 ↗ | odds value +3% | 2-0 | View analysis → |
| Fulham vs Manchester United Sun, 20 Sep 2026 15:30 GMT | Fulham win | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 29.9% vs 26.9% | +3.0pp | 3.6 vave stake 3.55 ↗ | odds value +8% | 0-1 | View analysis → |
| Manchester City vs Sunderland Sun, 20 Sep 2026 13:00 GMT | Sunderland win | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 8.9% vs 10.9% | -2.0pp | 9.45 cloudbet ↗ | odds value -16% | 2-0 | View analysis → |
| Manchester City vs Sunderland Sun, 20 Sep 2026 13:00 GMT | Draw | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 15.7% vs 17.0% | -1.3pp | 5.85 duelbits ↗ | odds value -8% | 2-0 | View analysis → |
| Fulham vs Manchester United Sun, 20 Sep 2026 15:30 GMT | Draw | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 25.8% vs 24.7% | +1.1pp | 4 vave stake 3.9 ↗ | odds value +3% | 0-1 | View analysis → |
| Bournemouth vs Liverpool Sun, 20 Sep 2026 13:00 GMT | Draw | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 26.3% vs 25.3% | +1.0pp | 3.85 vave stake 3.8 ↗ | odds value +1% | 1-0 | View analysis → |
| Leeds United vs Crystal Palace Sun, 20 Sep 2026 13:00 GMT | Draw | Aligned history: -0.95pp CLV over 23 Confirm near kickoff | 24.7% vs 24.1% | +0.6pp | 4.1 vave duelbits 4.05 ↗ | odds value +1% | 1-0 | View analysis → |
*EV = model probability × best odds − 1. This is what the bet would be worth **if the model probability were correct** — our settled closing-line record indicates it often is not. Best price = highest decimal odds across stake, bcgame, cloudbet, roobet, vave, duelbits (via OddsPapi). Snapshot: 2026-09-20T08:03:38.483Z. Measured overround & best-price data per sportsbook →
How To Read This Page
Decimal odds are converted into raw implied probability with 1 / odds. For each 1X2 market, the three implied probabilities are then normalized so the market totals 100%. Edge is calculated as model probability - market probability.
Read each row as a prompt to investigate, not a value signal. Our settled record shows the model has no demonstrated edge at picking sides, and the bands where it most exceeds the market have had negative closing-line value.
FAQ
What is a market edge?
A market edge compares FootyTips model probability with the de-vigged market implied probability from decimal odds. A positive edge means the model is higher than the market, not that the result is guaranteed.
Are these betting tips?
No. This page is a probability comparison tool. Odds can move, markets include margin, and football outcomes remain uncertain.
Why might no value picks appear?
If no licensed odds feed or private odds snapshot is connected, the page will not invent market edges. It only shows signals when real odds data is available.
Cite This Data
Writing about this? Copy this citation (attribution appreciated):