Bets tested
24,016
every one graded against the market's final verdict
updated just now
The evidence
A large historical test showing the method works, and a live account showing the scanner still works today. They ran on different markets, so each is shown on its own — never combined into a single figure.
Bets tested
24,016
every one graded against the market's final verdict
Return per bet, flat stakes
+4.93%
at the prices actually offered, nothing deducted
one unit a bet — so it compares with the live figure below
Luck detector
6.51
above 3 means very unlikely to be chance
| Market | Bets | Return | Mean CLV | Luck detector |
|---|---|---|---|---|
| 1X2 | 19,835 | +5.09% | +2.89% | 5.90 |
| Over/Under | 2,569 | +3.93% | +3.11% | 2.05 |
| Asian handicap | 1,612 | +4.50% | +1.42% | 2.01 |
It shows the method works. 24,016 bets over 2012–2025 is a large enough sample that luck is not a credible explanation. Every bet counts at the price that was actually offered, and prices above 3.50 were left out. Re-running the test reproduces these exact figures — they are read straight from its output rather than written into the page.
It is not a forecast. The historical prices are the best price across 33 to 50 bookmakers a match — the archive's own count, whose names it never published — while the live scanner compares every bookmaker our odds feed carries. Different books, different countries, different years, so expect a different number. The live account below shows what it is actually doing in today's market.
How our own record is staked. Every stake in the historical test was sized by a fixed rule from that bet's own price and edge, then capped. Against a flat stake over the same 24,016 bets it produced a smaller worst drawdown — 37.4% against 56.9%. This describes how the record was produced; it is not advice on what you should stake.
It is not being compared to zero. Every price carries the bookmaker's cut. Across 7,132 complete match-result markets in our own live feed, the middle one takes 7.29% — so €100 of turnover comes back as €93.21, before you have been right or wrong about anything. The tightest book measured takes 1.89%, the widest 13.57%. That is the starting point a bettor is up against: not break-even, but roughly 6.8% behind.
A high strike rate is not the same as making money. This is the part most records get wrong, and it is arithmetic anyone can check. What matters is not how often a bet wins, but whether it wins more often than its price requires.
| A record that backs… | At a price of | Wins | Needs to win | Actually returns |
|---|---|---|---|---|
| heavy favourites | 1.20 | 80.0% | 83.3% | −4.00% |
| short favourites | 1.50 | 62.0% | 66.7% | −7.00% |
| near coin-flips | 1.90 | 51.0% | 52.6% | −3.10% |
| these signals | varies | 47.5% | 45.2% | +4.93% |
The first three rows win more often than ours and lose money every time. Ours wins 47.5% of the time — and loses 51.8% — which is 2.3 points more often than the prices it was taken at require. That gap is the entire result.
Check the sum yourself. A winning bet returned +1.1909 on average per €1 staked. So 47.45% × +1.1909 − 51.77% × 0.9963 = +4.93%. Every number in that line is on this page.
What it is not. This is a return on each euro staked, not a return per year. What it becomes over a season depends on how many bets a market offers and how much a bookmaker will accept from you — and we will not publish a projected annual figure, because at this stage it would be a guess dressed as a result.
Bets placed live
124
1 still running
+1.68%
across 119 scored — the first number to trust
the backtest earned +3.09% on this same mix of leagues
Return per bet, flat stakes
+3.67%
across 123 settled bets
one unit a bet — the same flat €50 bets as the pre-match account.
Progress to verdict
119/500
our promise: if 500 bets do not show clearly positive CLV, we stop and say so
A strike rate means nothing on its own — it depends entirely on the prices taken. What can be compared is the gap: how often these bets won, against the rate the prices themselves required to break even.
| Wins | Needs to win | Clear by | Bets | |
|---|---|---|---|---|
| Historical — 1X2, Over/Under 2.5, Asian handicap | 47.5% | 45.2% | +2.3 pts | 24,016 |
| Live — 1X2, Over/Under 2.5, Asian handicap | 45.5% | 43.7% | +1.9 pts | 123 |
Both rows cover the same markets. “Needs to win” is the rate the odds taken require: for the live row it comes from the odds of every settled bet, because on a small sample the winning bets alone would overstate it — on the historical bets the two ways agree to within 0.1 points. The gap is what carries across — a strike rate on its own does not.
| Market | Bets | Mean CLV | Return |
|---|---|---|---|
| 1X2 | 114 | +1.71% | +4.15% |
| Over/Under | 3 | -1.22% | +29.00% |
| Asian handicap | 2 | +4.34% | -100.00% |
Profit is buried in luck — you can win for months on a losing strategy, or lose for months on a winning one. CLV is not: across 286,398 historical bets it predicted real profit almost one-for-one. It settles in weeks where profit takes years, so it is the number to watch first on the live account.
Every live bet is recorded at the price the bookmaker offered when we found it, and the backtest is measured the same way. Your own bookmaker may give a slightly different price — slippage — and nothing is deducted for it in our figures. Losing periods stay on this page.
The record above is about prices before kick-off, and it is measured against the market's own final verdict. The live signals are measured on their own record: how often each strategy lands. That strike rate gives every live signal its break-even odds and its value zone — the odds at which taking it is a value bet, as long as the strike rate holds.
What a live signal is: a match where our measured strike rate backs the predicted outcome. What it is not: a price or a moment to bet. In-play odds move every second, and a goal can be scored after the signal and before your bet is placed, so the odds at the moment of the signal are not odds you can count on.
| Signal type | Settled | Landed | Strike rate | Break-even odds | Value zone |
|---|---|---|---|---|---|
| The team leading when the signal arrives wins the match (Underdog wins) | 1684 | 1038 | 61.6% | 1.62 | 1.70+ |
| At least 1 goal in the Second Half (2nd Half) | 1109 | 863 | 77.8% | 1.29 | 1.35+ |
| After an active First Half: Both teams to score + Over 2.5 goals (Active 1st Half)
💡 Recommended: play it as a bet builder — Both teams to score + Over 2.5 goals. Its strike rate and break-even are this bet's. ⏳ Wait for the combined odds to reach 1.70 (if no goal has been scored by then): at 1.70 the bet has value. We publish this strategy to measure how well it works. |
471 | 297 | 63.1% | 1.59 | 1.67+ |
| At least 1 goal in the First Half (1st Half) | 405 | 296 | 73.1% | 1.37 | 1.44+ |
| Both teams to score (BTTS) | 74 | 42 | 56.8% | 1.76 | 1.85+ |
Break-even odds are the price this strike rate needs to come out level. Take the bet only if the odds you are offered are HIGHER than that number — above it the bet has value, at or below it the bet loses over time. It assumes this strike rate holds, and it is measured over the settled signals in the row beside it and nothing else. What is this?
Every signal is listed for subscribers, win or lose, with the odds it carried when it was sent. This is a hit rate, not a profit claim.