| League | Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| DEN1 | ViborgvAarhus | Aarhus A | 47.0% | 2.13 | 2.84 | +33.3% | 3.00% |
| BEL1 | Cercle BruggevSt. Truiden | St. Truiden A | 39.4% | 2.54 | 2.95 | +16.1% | 2.07% |
| NOR1 | RosenborgvViking | Viking A | 54.1% | 1.85 | 2.08 | +12.6% | 2.91% |
| P1 | Sporting CPvGuimaraes | Draw | 14.2% | 7.05 | 7.03 | -0.3% | — |
| AUT1 | Lask LinzvRied | Lask Linz H | 66.7% | 1.50 | 1.49 | -0.6% | — |
| P1 | Sporting CPvGuimaraes | Sporting CP H | 78.2% | 1.28 | 1.18 | -7.7% | — |
| AUT1 | Lask LinzvRied | Draw | 20.8% | 4.82 | 4.23 | -12.1% | — |
| BEL1 | Cercle BruggevSt. Truiden | Draw | 24.8% | 4.03 | 3.54 | -12.3% | — |
| DEN1 | ViborgvAarhus | Draw | 24.2% | 4.14 | 3.55 | -14.3% | — |
| BEL1 | Cercle BruggevSt. Truiden | Cercle Brugge H | 35.8% | 2.79 | 2.24 | -19.9% | — |
| NOR1 | RosenborgvViking | Draw | 21.4% | 4.67 | 3.71 | -20.7% | — |
| AUT1 | Lask LinzvRied | Ried A | 12.5% | 8.00 | 6.16 | -23.0% | — |
| NOR1 | RosenborgvViking | Rosenborg H | 24.5% | 4.08 | 3.12 | -23.5% | — |
| DEN1 | ViborgvAarhus | Viborg H | 28.9% | 3.46 | 2.28 | -34.3% | — |
| P1 | Sporting CPvGuimaraes | Guimaraes A longshot | 7.6% | 13.15 | 13.66 | +3.9% | — |
| PSL | StellenboschvMarumo Gallants | Stellenbosch H | 53.2% | 1.88 | — | — | — |
| ITCUP | ParmavCatania | Parma H thin data | 51.5% | 1.94 | — | — | — |
| J1 | Tokyo VerdyvKashiwa Reysol | Kashiwa Reysol A | 48.3% | 2.07 | — | — | — |
| N1 | TelstarvSparta Rotterdam | Telstar H | 44.9% | 2.23 | — | — | — |
| ITCUP | MonzavAvellino | Monza H thin data | 43.6% | 2.30 | — | — | — |
| N1 | TelstarvSparta Rotterdam | Sparta Rotterdam A | 29.2% | 3.42 | — | — | — |
| ITCUP | MonzavAvellino | Avellino A thin data | 28.9% | 3.46 | — | — | — |
| PSL | StellenboschvMarumo Gallants | Draw | 28.6% | 3.49 | — | — | — |
| J1 | Tokyo VerdyvKashiwa Reysol | Draw | 27.9% | 3.59 | — | — | — |
| ITCUP | MonzavAvellino | Draw thin data | 27.5% | 3.63 | — | — | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
Kept separate from the match-winner table because they are a different bet: independent of who wins, driven by the total-goals distribution rather than the difference between the two sides. Worth a look for two reasons — a goal model is being asked the question it was actually built to answer, and a totals line is far less correlated with the rest of a slip than another favourite. Against that, books charge more margin here than on the match winner, so a given edge has further to travel.
| League | Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|---|
| NOR1 | RosenborgvViking | Totals | Over3.5 | 45.3% | 2.21 | 6.00 | +171.8% | — |
| NOR1 | RosenborgvViking | Totals | Over2.5 | 66.2% | 1.51 | 3.69 | +143.8% | — |
| NOR1 | RosenborgvViking | BTTS | Yes— | 62.8% | 1.59 | 3.73 | +134.1% | — |
| DEN1 | ViborgvAarhus | Totals | Over3.5 | 33.2% | 3.01 | 6.88 | +128.8% | 3.00% |
| BEL1 | Cercle BruggevSt. Truiden | BTTS | Yes— | 56.1% | 1.78 | 3.94 | +121.3% | — |
| BEL1 | Cercle BruggevSt. Truiden | Totals | Over3.5 | 33.3% | 3.00 | 6.59 | +119.8% | 3.00% |
| P1 | Sporting CPvGuimaraes | Totals | Over3.5 | 42.8% | 2.33 | 5.09 | +118.2% | 3.00% |
| DEN1 | ViborgvAarhus | Totals | Over2.5 | 54.7% | 1.83 | 3.21 | +75.8% | 3.00% |
| P1 | Sporting CPvGuimaraes | Totals | Over2.5 | 64.0% | 1.56 | 2.73 | +74.7% | — |
| AUT1 | Lask LinzvRied | BTTS | Yes— | 51.4% | 1.95 | 3.40 | +74.7% | 3.00% |
| BEL1 | Cercle BruggevSt. Truiden | Totals | Over2.5 | 54.8% | 1.82 | 3.18 | +74.1% | 3.00% |
| BEL1 | Cercle BruggevSt. Truiden | Totals | Over1.5 | 77.8% | 1.28 | 2.22 | +72.9% | — |
| DEN1 | ViborgvAarhus | BTTS | Yes— | 56.1% | 1.78 | 3.00 | +68.1% | 3.00% |
| NOR1 | RosenborgvViking | Totals | Over1.5 | 85.2% | 1.17 | 1.95 | +65.9% | — |
| DEN1 | ViborgvAarhus | Totals | Over1.5 | 77.8% | 1.29 | 2.05 | +59.7% | — |
| P1 | Sporting CPvGuimaraes | BTTS | Yes— | 49.3% | 2.03 | 3.21 | +58.4% | 3.00% |
| P1 | Sporting CPvGuimaraes | Totals | Over1.5 | 83.8% | 1.19 | 1.81 | +51.9% | — |
| AUT1 | Lask LinzvRied | Totals | Under1.5 | 23.0% | 4.34 | 4.33 | -0.2% | — |
| AUT1 | Lask LinzvRied | Totals | Under3.5 | 67.7% | 1.48 | 1.44 | -2.4% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
The vs naive column is how far a bookmaker would be off if they priced the combination by multiplying the two legs together. A large positive number means the events happen together more often than independence implies — draws and low-scoring games, for instance — so a multiplied price is too generous.
| Match | Combination | Model | Fair | vs naive | Book | EV |
|---|---|---|---|---|---|---|
| StellenboschvMarumo Gallants | Away + Away clean sheet | 10.7% | 9.35 | +185% | — | — |
| TelstarvSparta Rotterdam | Away + Away clean sheet | 12.7% | 7.88 | +153% | — | — |
| RosenborgvViking | Home + Home clean sheet | 13.1% | 7.65 | +121% | — | — |
| RosenborgvViking | BTTS No + Under 2.5 | 23.3% | 4.29 | +118% | — | — |
| ParmavCatania | Away + Away clean sheet | 17.1% | 5.85 | +114% | — | — |
| Cercle BruggevSt. Truiden | Away + Away clean sheet | 16.0% | 6.27 | +113% | — | — |
| ViborgvAarhus | Home + Home clean sheet | 16.1% | 6.22 | +112% | — | — |
| Sporting CPvGuimaraes | Draw + Under 2.5 | 11.1% | 9.00 | +105% | — | — |
| MonzavAvellino | Away + Away clean sheet | 18.9% | 5.28 | +100% | — | — |
| Tokyo VerdyvKashiwa Reysol | Home + Home clean sheet | 19.6% | 5.11 | +98% | — | — |
| RosenborgvViking | Away + Away clean sheet | 15.6% | 6.41 | +97% | — | — |
| Tokyo VerdyvKashiwa Reysol | BTTS Yes + Over 2.5 | 27.8% | 3.59 | +95% | — | — |
| ViborgvAarhus | Away + Away clean sheet | 19.2% | 5.20 | +89% | — | — |
| Cercle BruggevSt. Truiden | Home + Home clean sheet | 19.3% | 5.18 | +89% | — | — |
| MonzavAvellino | Home + Home clean sheet | 21.2% | 4.73 | +86% | — | — |
| StellenboschvMarumo Gallants | BTTS Yes + Over 2.5 | 26.8% | 3.73 | +86% | — | — |
| Cercle BruggevSt. Truiden | BTTS No + Under 2.5 | 32.9% | 3.04 | +77% | — | — |
| ViborgvAarhus | BTTS No + Under 2.5 | 33.0% | 3.03 | +77% | — | — |
| RosenborgvViking | Draw + Under 2.5 | 13.5% | 7.42 | +75% | — | — |
| ParmavCatania | BTTS Yes + Over 2.5 | 34.0% | 2.94 | +72% | — | — |
Legs from different matches are independent, so the probabilities multiply — and so does the bookmaker's margin. The final column is the share of fair value lost to compounding vig at 5% per leg, before your model is even wrong about anything. Accumulators are on this page because people want them, not because they are a good idea.
| Legs | N | Model | Fair | Book | EV | Margin cost |
|---|---|---|---|---|---|---|
| Rosenborg v Viking — over 2.5 goals Cercle Brugge v St. Truiden — both teams to score Viborg v Aarhus — over 2.5 goals | 3 | 20.3% | 4.92 | 46.70 | +848.6% | −14% |
| Rosenborg v Viking — over 2.5 goals Cercle Brugge v St. Truiden — both teams to score Sporting CP v Guimaraes — over 2.5 goals | 3 | 23.8% | 4.21 | 39.67 | +842.5% | −14% |
| Rosenborg v Viking — both teams to score Viborg v Aarhus — over 2.5 goals Sporting CP v Guimaraes — over 2.5 goals | 3 | 22.0% | 4.55 | 32.73 | +619.0% | −14% |
| Rosenborg v Viking — both teams to score Lask Linz v Ried — both teams to score Cercle Brugge v St. Truiden — over 2.5 goals | 3 | 17.7% | 5.66 | 40.29 | +612.0% | −14% |
| Lask Linz v Ried — both teams to score Cercle Brugge v St. Truiden — over 2.5 goals Viborg v Aarhus — both teams to score | 3 | 15.8% | 6.33 | 32.38 | +411.3% | −14% |
| Viborg v Aarhus — both teams to score Sporting CP v Guimaraes — both teams to score | 2 | 27.6% | 3.62 | 9.63 | +166.3% | −9% |
| Sporting CP v Guimaraes — both teams to score Lask Linz v Ried — Lask Linz to win (H) | 2 | 32.9% | 3.04 | 4.79 | +57.5% | −9% |
| Tokyo Verdy v Kashiwa Reysol — under 3.5 goals Stellenbosch v Marumo Gallants — Stellenbosch or draw (H) | 2 | 68.4% | 1.46 | — | — | −9% |
| Tokyo Verdy v Kashiwa Reysol — under 3.5 goals Stellenbosch v Marumo Gallants — under 3.5 goals | 2 | 68.0% | 1.47 | — | — | −9% |
| Stellenbosch v Marumo Gallants — Stellenbosch or draw (H) Parma v Catania — Parma or draw (H) thin data | 2 | 64.0% | 1.56 | — | — | −9% |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| Lask LinzvRied | Lask Linz H | 66.7% | 1.50 | 1.49 | -0.6% | — |
| Lask LinzvRied | Draw | 20.8% | 4.82 | 4.23 | -12.1% | — |
| Lask LinzvRied | Ried A | 12.5% | 8.00 | 6.16 | -23.0% | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| Lask LinzvRied | BTTS | Yes— | 51.4% | 1.95 | 3.40 | +74.7% | 3.00% |
| Lask LinzvRied | Totals | Under1.5 | 23.0% | 4.34 | 4.33 | -0.2% | — |
| Lask LinzvRied | Totals | Under3.5 | 67.7% | 1.48 | 1.44 | -2.4% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
| Model | Market | Book odds | Edge | |
|---|---|---|---|---|
| Lask Linz | 66.7% | 64.3% | 1.49 | +2.5pp |
| Draw | 20.8% | 21.5% | 4.23 | -0.7pp |
| Ried | 12.5% | 14.3% | 6.16 | -1.7pp |
| Over 1.5 | 77.0% | 78.3% | 1.20 | -1.3pp |
| Under 1.5 | 23.0% | 21.7% | 4.33 | +1.3pp |
| Over 3.5 | 32.3% | 35.5% | 2.62 | -3.2pp |
| Under 3.5 | 67.7% | 64.5% | 1.44 | +3.2pp |
| BTTS Yes | 51.4% | 27.7% | 3.40 | +23.7pp |
| BTTS No | 48.6% | 72.3% | 1.30 | -23.7pp |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| Cercle BruggevSt. Truiden | St. Truiden A | 39.4% | 2.54 | 2.95 | +16.1% | 2.07% |
| Cercle BruggevSt. Truiden | Draw | 24.8% | 4.03 | 3.54 | -12.3% | — |
| Cercle BruggevSt. Truiden | Cercle Brugge H | 35.8% | 2.79 | 2.24 | -19.9% | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| Cercle BruggevSt. Truiden | BTTS | Yes— | 56.1% | 1.78 | 3.94 | +121.3% | — |
| Cercle BruggevSt. Truiden | Totals | Over3.5 | 33.3% | 3.00 | 6.59 | +119.8% | 3.00% |
| Cercle BruggevSt. Truiden | Totals | Over2.5 | 54.8% | 1.82 | 3.18 | +74.1% | 3.00% |
| Cercle BruggevSt. Truiden | Totals | Over1.5 | 77.8% | 1.28 | 2.22 | +72.9% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
| Model | Market | Book odds | Edge | |
|---|---|---|---|---|
| Cercle Brugge | 35.8% | 42.2% | 2.24 | -6.4pp |
| Draw | 24.8% | 26.1% | 3.54 | -1.3pp |
| St. Truiden | 39.4% | 31.6% | 2.95 | +7.7pp |
| Over 1.5 | 77.8% | 42.5% | 2.22 | +35.3pp |
| Under 1.5 | 22.2% | 57.5% | 1.64 | -35.3pp |
| Over 2.5 | 54.8% | 33.3% | 3.18 | +21.5pp |
| Under 2.5 | 45.2% | 66.7% | 1.58 | -21.5pp |
| Over 3.5 | 33.3% | 15.9% | 6.59 | +17.4pp |
| Under 3.5 | 66.7% | 84.1% | 1.25 | -17.4pp |
| BTTS Yes | 56.1% | 27.4% | 3.94 | +28.8pp |
| BTTS No | 43.9% | 72.6% | 1.49 | -28.8pp |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| ViborgvAarhus | Aarhus A | 47.0% | 2.13 | 2.84 | +33.3% | 3.00% |
| ViborgvAarhus | Draw | 24.2% | 4.14 | 3.55 | -14.3% | — |
| ViborgvAarhus | Viborg H | 28.9% | 3.46 | 2.28 | -34.3% | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| ViborgvAarhus | Totals | Over3.5 | 33.2% | 3.01 | 6.88 | +128.8% | 3.00% |
| ViborgvAarhus | Totals | Over2.5 | 54.7% | 1.83 | 3.21 | +75.8% | 3.00% |
| ViborgvAarhus | BTTS | Yes— | 56.1% | 1.78 | 3.00 | +68.1% | 3.00% |
| ViborgvAarhus | Totals | Over1.5 | 77.8% | 1.29 | 2.05 | +59.7% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
| Model | Market | Book odds | Edge | |
|---|---|---|---|---|
| Viborg | 28.9% | 41.3% | 2.28 | -12.5pp |
| Draw | 24.2% | 25.9% | 3.55 | -1.7pp |
| Aarhus | 47.0% | 32.8% | 2.84 | +14.2pp |
| Over 1.5 | 77.8% | 45.9% | 2.05 | +31.9pp |
| Under 1.5 | 22.2% | 54.1% | 1.74 | -31.9pp |
| Over 2.5 | 54.7% | 32.1% | 3.21 | +22.6pp |
| Under 2.5 | 45.3% | 67.9% | 1.52 | -22.6pp |
| Over 3.5 | 33.2% | 15.0% | 6.88 | +18.3pp |
| Under 3.5 | 66.8% | 85.0% | 1.21 | -18.3pp |
| BTTS Yes | 56.1% | 36.6% | 3.00 | +19.4pp |
| BTTS No | 43.9% | 63.4% | 1.73 | -19.4pp |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| ParmavCatania | Parma H thin data | 51.5% | 1.94 | — | — | — |
| MonzavAvellino | Monza H thin data | 43.6% | 2.30 | — | — | — |
| MonzavAvellino | Avellino A thin data | 28.9% | 3.46 | — | — | — |
| MonzavAvellino | Draw thin data | 27.5% | 3.63 | — | — | — |
| ParmavCatania | Draw thin data | 26.7% | 3.75 | — | — | — |
| ParmavCatania | Catania A thin data | 21.8% | 4.58 | — | — | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| Tokyo VerdyvKashiwa Reysol | Kashiwa Reysol A | 48.3% | 2.07 | — | — | — |
| Tokyo VerdyvKashiwa Reysol | Draw | 27.9% | 3.59 | — | — | — |
| Tokyo VerdyvKashiwa Reysol | Tokyo Verdy H | 23.9% | 4.19 | — | — | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| TelstarvSparta Rotterdam | Telstar H | 44.9% | 2.23 | — | — | — |
| TelstarvSparta Rotterdam | Sparta Rotterdam A | 29.2% | 3.42 | — | — | — |
| TelstarvSparta Rotterdam | Draw | 25.9% | 3.87 | — | — | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| RosenborgvViking | Viking A | 54.1% | 1.85 | 2.08 | +12.6% | 2.91% |
| RosenborgvViking | Draw | 21.4% | 4.67 | 3.71 | -20.7% | — |
| RosenborgvViking | Rosenborg H | 24.5% | 4.08 | 3.12 | -23.5% | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| RosenborgvViking | Totals | Over3.5 | 45.3% | 2.21 | 6.00 | +171.8% | — |
| RosenborgvViking | Totals | Over2.5 | 66.2% | 1.51 | 3.69 | +143.8% | — |
| RosenborgvViking | BTTS | Yes— | 62.8% | 1.59 | 3.73 | +134.1% | — |
| RosenborgvViking | Totals | Over1.5 | 85.2% | 1.17 | 1.95 | +65.9% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
| Model | Market | Book odds | Edge | |
|---|---|---|---|---|
| Rosenborg | 24.5% | 29.8% | 3.12 | -5.2pp |
| Draw | 21.4% | 24.8% | 3.71 | -3.4pp |
| Viking | 54.1% | 45.5% | 2.08 | +8.6pp |
| Over 1.5 | 85.2% | 48.2% | 1.95 | +37.1pp |
| Under 1.5 | 14.8% | 51.8% | 1.81 | -37.1pp |
| Over 2.5 | 66.2% | 27.9% | 3.69 | +38.3pp |
| Under 2.5 | 33.8% | 72.1% | 1.43 | -38.3pp |
| Over 3.5 | 45.3% | 18.0% | 6.00 | +27.3pp |
| Under 3.5 | 54.7% | 82.0% | 1.32 | -27.3pp |
| BTTS Yes | 62.8% | 29.8% | 3.73 | +32.9pp |
| BTTS No | 37.2% | 70.2% | 1.58 | -32.9pp |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| Sporting CPvGuimaraes | Draw | 14.2% | 7.05 | 7.03 | -0.3% | — |
| Sporting CPvGuimaraes | Sporting CP H | 78.2% | 1.28 | 1.18 | -7.7% | — |
| Sporting CPvGuimaraes | Guimaraes A longshot | 7.6% | 13.15 | 13.66 | +3.9% | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.
| Match | Market | Pick | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|---|
| Sporting CPvGuimaraes | Totals | Over3.5 | 42.8% | 2.33 | 5.09 | +118.2% | 3.00% |
| Sporting CPvGuimaraes | Totals | Over2.5 | 64.0% | 1.56 | 2.73 | +74.7% | — |
| Sporting CPvGuimaraes | BTTS | Yes— | 49.3% | 2.03 | 3.21 | +58.4% | 3.00% |
| Sporting CPvGuimaraes | Totals | Over1.5 | 83.8% | 1.19 | 1.81 | +51.9% | — |
One row per market per match: the side the model prefers. The two sides of a two-way line are near mirror images, so showing both would double the table without adding anything. Edges are measured against each pair de-vigged on its own, not against the match-winner margin — books load goals markets more heavily, and borrowing the 1X2 overround here would flatter every row.
| Model | Market | Book odds | Edge | |
|---|---|---|---|---|
| Sporting CP | 78.2% | 81.9% | 1.18 | -3.7pp |
| Draw | 14.2% | 12.4% | 7.03 | +1.8pp |
| Guimaraes | 7.6% | 5.7% | 13.66 | +1.9pp |
| Over 1.5 | 83.8% | 52.1% | 1.81 | +31.7pp |
| Under 1.5 | 16.2% | 47.9% | 1.97 | -31.7pp |
| Over 2.5 | 64.0% | 38.5% | 2.73 | +25.5pp |
| Under 2.5 | 36.0% | 61.5% | 1.71 | -25.5pp |
| Over 3.5 | 42.8% | 21.8% | 5.09 | +21.1pp |
| Under 3.5 | 57.2% | 78.2% | 1.42 | -21.1pp |
| BTTS Yes | 49.3% | 29.6% | 3.21 | +19.7pp |
| BTTS No | 50.7% | 70.4% | 1.35 | -19.7pp |
| Match | Selection | Model | Fair | Book | EV | Stake |
|---|---|---|---|---|---|---|
| StellenboschvMarumo Gallants | Stellenbosch H | 53.2% | 1.88 | — | — | — |
| StellenboschvMarumo Gallants | Draw | 28.6% | 3.49 | — | — | — |
| StellenboschvMarumo Gallants | Marumo Gallants A | 18.1% | 5.51 | — | — | — |
Stake is blank unless a selection clears every gate: positive expected value, at least 5 percentage points of edge over the de-vigged line, and not more than 25 (a disagreement that large is usually the model missing something, not an opportunity). Most rows will be blank most days. Selections marked longshot are below 10% — their EV figure divides by a very small number and should not be trusted. thin data means one side has too few matches in the model's history for its rating to mean anything; the huge edge shown is the model's ignorance, not the market's mistake.