xgEdge

Model vs market

We model football probabilities, then test them against the market.

The model estimates each outcome's probability, compares it with the bookmaker's margin-free probability, and tracks what happens when the two disagree.

matches priced next 72 h 25 leagues
184
signals across 106 matches
187
newest price each price carries its own time
Research model · Not betting advice

Not betting advice. Estimates from a statistical model, published for reading. A positive edge is a statement about an average over many comparable prices, not about the match in front of you.

Current evidence · Retrospective record · Rule v2

No profitable edge demonstrated

Lost money on this record (interval below zero)

ROI
−17.7%
95% interval
−25.3% to −10.2%
Sample
1,697 signals 1,023 matches

Prospective validation

Not started

No signal published live has settled yet: 0 of 400 signals, 0 of 250 matches.

Settled signals 0 of 400
Settled matches 0 of 250

Counting model version 2 since 19 November 2026. Earlier versions' signals are kept apart, not pooled.

Model version 2, an interim, took effect on 19 November 2026: less compressed than version 1, and still failing two of four honesty conditions (goal-difference slope 1.23 against 0.85 to 1.15; goals predicted 3.4% under those scored, against within 3%). No profitable edge is demonstrated under any version.

Earlier model series

  • Model version 1, 26 August 2026 to 19 November 2026 Closed

    315 signals written, 190 settled on 84 matches.

    Final result: −36.01u units, ROI −19.0%. 95% interval −41.8% to +5.9%.

How this is computed
Units
−300.49u
Rule
v2

Today's rule, version 2, applied to each historical selection at its last evaluation before kickoff. These were not necessarily signals published live at the time, and a quoted historical price may not have been available to you. This record can hold the verdict down; it can never raise it. Only signals published live, in the forward ledger, can.

The 95% interval resamples matches rather than selections, 2,000 times from seed 20261102, so selections on one match move together.

11 voids are staked and counted. 929 settled rows are left out of every figure: their price was captured after the evaluation it was paired with.

Retrospective: today's rule asked about each selection at its last evaluation before kickoff, not a list published at the time. A quoted historical price may not have been available to you. Computed .

How this record is calculated

Computed . Explore the evidence

Today's model–market disagreements

Each match's largest price that meets the current signal rule, with the model's probability set against the market's. A signal is not a recommendation.

Disagreement Model-implied edge

Live validation: 0 of 400 signals settled

  1. Serie A Juventus – Venezia Venezia to win · 1X2
    Model probability
    16.5%
    Market probability
    11.3%
    Disagreement
    +5.2 pp
    Model price
    6.08
    Market price
    8.50 Bet365
    Model-implied edge
    +39.9%
    Details : Juventus – Venezia

    Why the model differs

    Why the model favours Juventus

    16.5% Venezia to win

    Attack
    7th vs 14th
    Juventus attack v Venezia defence
    Defence
    1st vs 9th
    Defence v their attack
    Recent xG form
    +0.44 vs -0.19
    xGD per match, last 6
    Projected goals
    1.89 – 0.99
    Home v away

    These are the model's inputs for this match, not a causal attribution, and not advice.

    Price at which a 5% model-implied edge begins
    6.38
    Bookmaker margin
    4.1%
    Raw implied probability (1 / odds)
    11.8%
    Margin removal
    Multiplicative, over the complete book
    Odds captured
    Price age
    9 h 20 min
    Time to kickoff
    2 d 12 h

    Similar past signals (retrospective, model-implied edge 30–40%): Insufficient sample (45)

    Versions
    Model v2 · rule v2 · model2-rule2-hierarchy0-freshness0-markets0-calibration1-evaluation0

    Full match analysis →: Juventus – Venezia

  2. Super League Greece PAOK – AE Kifisia PAOK at most 1 goal · Team goals
    Model probability
    50.7%
    Market probability
    33.7%
    Disagreement
    +16.9 pp
    Model price
    1.97
    Market price
    2.75 Bet365
    Model-implied edge
    +39.3%
    Details : PAOK – AE Kifisia

    Why the model expects PAOK to score at most once

    How the model gets this chance

    1. Projected goals 1.61 – 1.07
    2. Score matrix Binomial · Dixon–Coles
    3. This selection 50.7% Derived prediction

    PAOK's projected goals and AE Kifisia's set a score matrix of two moment-matched binomial goal counts with the Dixon–Coles low-score adjustment; the scorelines this selection covers add up to 50.7%.

    Projected goalsDerived prediction
    1.61
    PAOK's scoring rate in the score matrix the probability is read from
    VenueModel input
    Home
    The model uses the competition's home scoring rate
    Ranks, recent window and sample
    Attack v defenceModel input
    4th vs 4th
    PAOK's attack against AE Kifisia's defence, by fitted rating
    Recent xGContext
    1.94 vs 0.92
    Not a model input: PAOK's xG per match against AE Kifisia's xG conceded, last 6 matches

    These are the model's inputs for this match, not a causal attribution, and not advice.

    Price at which a 5% model-implied edge begins
    2.07
    Bookmaker margin
    7.8%
    Raw implied probability (1 / odds)
    36.4%
    Margin removal
    Multiplicative, over the complete book
    Odds captured
    Price age
    3 h 21 min
    Time to kickoff
    2 d 18 h

    Similar past signals (retrospective, model-implied edge 30–40%): Insufficient sample (31)

    Versions
    Model v2 · rule v2 · model2-rule2-hierarchy0-freshness0-markets0-calibration1-evaluation0

    Full match analysis →: PAOK – AE Kifisia

  3. Ukrainian Premier League Dynamo Kyiv – Obolon Kyiv Dynamo Kyiv not to score · Team goals
    Model probability
    18.6%
    Market probability
    12.5%
    Disagreement
    +6.1 pp
    Model price
    5.39
    Market price
    7.50 Bet365
    Model-implied edge
    +39.2%
    Details : Dynamo Kyiv – Obolon Kyiv

    Why the model rates Dynamo Kyiv not to score above the market

    How the model gets this chance

    1. Projected goals 1.55 – 0.59
    2. Score matrix Binomial · Dixon–Coles
    3. This selection 18.6% Derived prediction

    Dynamo Kyiv's projected goals and Obolon Kyiv's set a score matrix of two moment-matched binomial goal counts with the Dixon–Coles low-score adjustment; the scorelines this selection covers add up to 18.6%.

    Projected goalsDerived prediction
    1.55
    Dynamo Kyiv's scoring rate in the score matrix the probability is read from
    VenueModel input
    Home
    The model uses the competition's home scoring rate
    Ranks, recent window and sample
    Attack v defenceModel input
    5th vs 8th
    Dynamo Kyiv's attack against Obolon Kyiv's defence, by fitted rating
    Recent xGContext
    1.45 vs 1.70
    Not a model input: Dynamo Kyiv's xG per match against Obolon Kyiv's xG conceded, last 6 matches

    These are the model's inputs for this match, not a causal attribution, and not advice.

    Price at which a 5% model-implied edge begins
    5.66
    Bookmaker margin
    6.7%
    Raw implied probability (1 / odds)
    13.3%
    Margin removal
    Multiplicative, over the complete book
    Odds captured
    Price age
    12 h 20 min
    Time to kickoff
    2 d 11 h

    Similar past signals (retrospective, model-implied edge 30–40%): Insufficient sample (25)

    Versions
    Model v2 · rule v2 · model2-rule2-hierarchy0-freshness0-markets0-calibration1-evaluation0

    Full match analysis →: Dynamo Kyiv – Obolon Kyiv

  4. Premier League Aston Villa – Sunderland Sunderland 2+ goals · Team goals
    Model probability
    42.7%
    Market probability
    29.1%
    Disagreement
    +13.6 pp
    Model price
    2.34
    Market price
    3.25 Bet365
    Model-implied edge
    +38.8%
    Details : Aston Villa – Sunderland

    Why the model rates Sunderland 2+ goals above the market

    How the model gets this chance

    1. Projected goals 1.52 – 1.42
    2. Score matrix Binomial · Dixon–Coles
    3. This selection 42.7% Derived prediction

    Sunderland's projected goals and Aston Villa's set a score matrix of two moment-matched binomial goal counts with the Dixon–Coles low-score adjustment; the scorelines this selection covers add up to 42.7%.

    Projected goalsDerived prediction
    1.42
    Sunderland's scoring rate in the score matrix the probability is read from
    VenueModel input
    Away
    The model uses the competition's away scoring rate
    Ranks, recent window and sample
    Attack v defenceModel input
    13th vs 17th
    Sunderland's attack against Aston Villa's defence, by fitted rating
    Recent xGContext
    1.57 vs 1.83
    Not a model input: Sunderland's xG per match against Aston Villa's xG conceded, last 6 matches

    These are the model's inputs for this match, not a causal attribution, and not advice.

    Price at which a 5% model-implied edge begins
    2.46
    Bookmaker margin
    5.8%
    Raw implied probability (1 / odds)
    30.8%
    Margin removal
    Multiplicative, over the complete book
    Odds captured
    Price age
    3 h 20 min
    Time to kickoff
    39 h 38 min

    Similar past signals (retrospective, model-implied edge 30–40%): Insufficient sample (38)

    Versions
    Model v2 · rule v2 · model2-rule2-hierarchy0-freshness0-markets0-calibration1-evaluation0

    Full match analysis →: Aston Villa – Sunderland

  5. SuperLiga (Serbia) FK IMT Beograd – FK Partizan FK IMT Beograd to win · 1X2
    Model probability
    40.8%
    Market probability
    27.2%
    Disagreement
    +13.5 pp
    Model price
    2.45
    Market price
    3.40 Bet365
    Model-implied edge
    +38.7%
    Details : FK IMT Beograd – FK Partizan

    Why the model differs

    Why the model favours FK IMT Beograd

    40.8% FK IMT Beograd to win

    Attack
    8th vs 4th
    FK IMT Beograd attack v FK Partizan defence
    Defence
    1st vs 3rd
    Defence v their attack
    Recent xG form
    +0.06 vs +0.93
    xGD per match, last 6
    Projected goals
    1.32 – 1.05
    Home v away

    These are the model's inputs for this match, not a causal attribution, and not advice.

    Price at which a 5% model-implied edge begins
    2.57
    Bookmaker margin
    8.0%
    Raw implied probability (1 / odds)
    29.4%
    Margin removal
    Multiplicative, over the complete book
    Odds captured
    Price age
    9 h 21 min
    Time to kickoff
    2 d 14 h

    Similar past signals (retrospective, model-implied edge 30–40%): Insufficient sample (48)

    Versions
    Model v2 · rule v2 · model2-rule2-hierarchy0-freshness0-markets0-calibration1-evaluation0

    Full match analysis →: FK IMT Beograd – FK Partizan

See all 187 signals in the table →

All prices, next 72 hours

Every priced outcome against the model, largest disagreement first. The table opens on the signals at this moment; All prices shows every evaluated price in the window.

1–25 of 209 prices

  • Signal: every check of the rule passed, a model-implied edge of 5% or more
  • Stale price, limited data or doesn't beat base rate: never a signal
  • Model probability
  • Market probability
  1. Shakhtar Donetsk – FC Livyi Bereh Kyiv Ukrainian Premier League Shakhtar Donetsk 2+ goals Team goals
    Details: Shakhtar Donetsk 2+ goals
    Model
    49.7%
    Market
    64.5%
    Disagreement
    −14.8 pp
    Model-implied edge
    −28.2%

    Model price 2.01 Market price 1.44 Bet365

  2. FC Bayern München – 1. FC Köln Bundesliga 1. FC Köln not to score Team goals
    Details: 1. FC Köln not to score
    Model
    32.6%
    Market
    46.3%
    Disagreement
    −13.8 pp
    Model-implied edge
    −34.8%

    Model price 3.07 Market price 2.00 Bet365

  3. FC Bayern München – 1. FC Köln Bundesliga Both teams not to score BTTS
    Details: Both teams not to score
    Model
    34.4%
    Market
    48.0%
    Disagreement
    −13.6 pp
    Model-implied edge
    −32.8%

    Model price 2.90 Market price 1.95 Bet365

  4. Shakhtar Donetsk – FC Livyi Bereh Kyiv Ukrainian Premier League Over 2.5 goals Over/Under
    Details: Over 2.5 goals
    Model
    41.4%
    Market
    53.9%
    Disagreement
    −12.5 pp
    Model-implied edge
    −27.5%

    Model price 2.41 Market price 1.75 Bet365

  5. Shakhtar Donetsk – FC Livyi Bereh Kyiv Ukrainian Premier League Shakhtar Donetsk to win 1X2
    Details: Shakhtar Donetsk to win
    Model
    59.2%
    Market
    71.3%
    Disagreement
    −12.1 pp
    Model-implied edge
    −26.0%

    Model price 1.69 Market price 1.25 Bet365

  6. FC Bayern München – 1. FC Köln Bundesliga FC Bayern München to win 1X2
    Details: FC Bayern München to win
    Model
    78.2%
    Market
    89.1%
    Disagreement
    −10.9 pp
    Model-implied edge
    −16.3%

    Model price 1.28 Market price 1.07 Bet365

  7. Molde FK – Bodø/Glimt Eliteserien Bodø/Glimt 2+ goals Team goals
    Details: Bodø/Glimt 2+ goals
    Model
    58.5%
    Market
    68.7%
    Disagreement
    −10.2 pp
    Model-implied edge
    −20.2%

    Model price 1.71 Market price 1.36 Bet365

  8. Troyes – Le Havre Ligue 1 Both teams not to score BTTS
    Details: Both teams not to score
    Model
    35.4%
    Market
    45.3%
    Disagreement
    −10.0 pp
    Model-implied edge
    −27.5%

    Model price 2.83 Market price 2.05 Bet365

  9. AGF – Lyngby Superliga (Denmark) Both teams not to score BTTS
    Details: Both teams not to score
    Model
    32.7%
    Market
    42.3%
    Disagreement
    −9.6 pp
    Model-implied edge
    −28.0%

    Model price 3.06 Market price 2.20 Bet365

  10. Athletic Club – Espanyol La Liga Athletic Club 2+ goals Team goals
    Details: Athletic Club 2+ goals
    Model
    49.5%
    Market
    58.9%
    Disagreement
    −9.4 pp
    Model-implied edge
    −22.3%

    Model price 2.02 Market price 1.57 Bet365

  11. Athletic Club – Espanyol La Liga Athletic Club to win 1X2
    Details: Athletic Club to win
    Model
    54.0%
    Market
    63.4%
    Disagreement
    −9.4 pp
    Model-implied edge
    −18.9%

    Model price 1.85 Market price 1.50 Bet365

  12. Troyes – Le Havre Ligue 1 Under 1.5 goals Over/Under
    Details: Under 1.5 goals
    Model
    16.3%
    Market
    25.5%
    Disagreement
    −9.2 pp
    Model-implied edge
    −38.7%

    Model price 6.12 Market price 3.75 Bet365

  13. Troyes – Le Havre Ligue 1 Under 2.5 goals Over/Under
    Details: Under 2.5 goals
    Model
    41.0%
    Market
    50.0%
    Disagreement
    −9.0 pp
    Model-implied edge
    −21.7%

    Model price 2.44 Market price 1.91 Bet365

  14. AGF – Lyngby Superliga (Denmark) Lyngby not to score Team goals
    Details: Lyngby not to score
    Model
    24.8%
    Market
    33.7%
    Disagreement
    −8.9 pp
    Model-implied edge
    −31.7%

    Model price 4.03 Market price 2.75 Bet365

  15. FC Bayern München – 1. FC Köln Bundesliga 1. FC Köln at most 1 goal Team goals
    Details: 1. FC Köln at most 1 goal
    Model
    72.2%
    Market
    81.1%
    Disagreement
    −8.9 pp
    Model-implied edge
    −15.7%

    Model price 1.38 Market price 1.17 Bet365

  16. Shakhtar Donetsk – FC Livyi Bereh Kyiv Ukrainian Premier League FC Livyi Bereh Kyiv not to score Team goals
    Details: FC Livyi Bereh Kyiv not to score
    Model
    47.2%
    Market
    55.7%
    Disagreement
    −8.5 pp
    Model-implied edge
    −21.3%

    Model price 2.12 Market price 1.67 Bet365

  17. Pogoń Szczecin – Śląsk Wrocław Ekstraklasa Pogoń Szczecin 2+ goals Team goals
    Details: Pogoń Szczecin 2+ goals
    Model
    49.4%
    Market
    57.7%
    Disagreement
    −8.3 pp
    Model-implied edge
    −20.3%

    Model price 2.03 Market price 1.62 Bet365

  18. Troyes – Le Havre Ligue 1 Troyes at most 1 goal Team goals
    Details: Troyes at most 1 goal
    Model
    47.7%
    Market
    55.7%
    Disagreement
    −8.0 pp
    Model-implied edge
    −20.5%

    Model price 2.10 Market price 1.67 Bet365

  19. Pogoń Szczecin – Śląsk Wrocław Ekstraklasa Pogoń Szczecin to win 1X2
    Details: Pogoń Szczecin to win
    Model
    44.8%
    Market
    52.7%
    Disagreement
    −7.8 pp
    Model-implied edge
    −22.6%

    Model price 2.23 Market price 1.73 Bet365

  20. Molde FK – Bodø/Glimt Eliteserien Bodø/Glimt to win 1X2
    Details: Bodø/Glimt to win
    Model
    44.6%
    Market
    52.2%
    Disagreement
    −7.7 pp
    Model-implied edge
    −19.7%

    Model price 2.24 Market price 1.80 Bet365

  21. FC Botoșani – FC Voluntari SuperLiga (Romania) Both teams not to score BTTS
    Details: Both teams not to score
    Model
    33.5%
    Market
    41.1%
    Disagreement
    −7.6 pp
    Model-implied edge
    −24.6%

    Model price 2.98 Market price 2.25 Bet365

  22. Troyes – Le Havre Ligue 1 Troyes not to score Team goals
    Details: Troyes not to score
    Model
    15.9%
    Market
    23.4%
    Disagreement
    −7.5 pp
    Model-implied edge
    −36.4%

    Model price 6.29 Market price 4.00 Bet365

  23. FK Septemvri Sofia – Lokomotiv Plovdiv Parva Liga (Bulgaria) Over 2.5 goals Over/Under
    Details: Over 2.5 goals
    Model
    32.6%
    Market
    40.1%
    Disagreement
    −7.4 pp
    Model-implied edge
    −23.3%

    Model price 3.07 Market price 2.35 Bet365

  24. FC Farul Constanța – FC Universitatea Cluj SuperLiga (Romania) FC Universitatea Cluj to win 1X2
    Details: FC Universitatea Cluj to win
    Model
    31.7%
    Market
    39.0%
    Disagreement
    −7.4 pp
    Model-implied edge
    −25.6%

    Model price 3.16 Market price 2.35 Bet365

  25. Piast Gliwice – MKS Korona Kielce Ekstraklasa Both teams not to score BTTS
    Details: Both teams not to score
    Model
    34.0%
    Market
    41.1%
    Disagreement
    −7.1 pp
    Model-implied edge
    −23.5%

    Model price 2.94 Market price 2.25 Bet365

Graded results for everything this board ranked: Results