The same production stack priced under 4 course readings, against the bookmaker's pre-stage line. Each scenario is one opinion about what kind of day this is; the results section settles the argument.
The fork
Course profile is an input the model can't see from a map — it's set from reference races (or overridden by hand). Different readings redistribute the same probability mass across rider types.
The board
Blend win probability per scenario (cascade × LambdaRank, DNF-adjusted at the pricing layer), next to the devigged Kambi (Unibet) win line captured the morning before the stage. "Max edge" compares the most favourable scenario against the book — treat it as scenario-conditional, not a free bet.
Read "Max edge" as directional, not as a price. The blend is a set of per-rider marginals and does not sum to one book — across this field each reading totals Hard-men sprint 135%, Sprint survives 124%, Hindsight (post-hoc) 124% — while the devigged book column sums to exactly 100%. Putting the two side by side flatters the model, and Max edge is inflated by about that factor: divide a rider's probability by his reading's total before calling anything an edge. Of our heads only the LambdaRank one is natively book-preserving.
| Rider | Team | Rouleur break | Hard-men sprint | Sprint survives | Hindsight (post-hoc) | Book odds | Book devig | Max edge |
|---|---|---|---|---|---|---|---|---|
| Kooij Olav | Decathlon CMA CGM Team | 0.1% | 16.4% | 48.4% | 0.4% | 9.00 | 7.2% | +41.1pp |
| Pogacar Tadej | UAE Team Emirates-XRG | – | 11.0% | 3.2% | 36.6% | 301.00 | 0.2% | +36.4pp |
| Philipsen Jasper | Alpecin-Premier Tech | 0.2% | 23.8% | 30.5% | 2.0% | 7.00 | 9.3% | +21.2pp |
| Pedersen Mads | Lidl-Trek | 1.8% | 11.4% | 3.4% | 3.6% | 5.00 | 13.0% | -1.6pp |
| van der Poel Mathieu | Alpecin-Premier Tech | 3.7% | 10.8% | 1.8% | 6.2% | 7.00 | 9.3% | +1.5pp |
| del Toro Isaac | UAE Team Emirates-XRG | – | 0.9% | 0.3% | 6.9% | 301.00 | 0.2% | +6.7pp |
| Seixas Paul | Decathlon CMA CGM Team | – | 1.2% | 0.0% | 6.8% | n/a | – | – |
| Girmay Biniam | NSN Cycling Team | 0.3% | 6.1% | 5.7% | 1.3% | 21.00 | 3.1% | +3.0pp |
| Kanter Max | XDS Astana Team | 0.1% | 3.9% | 6.1% | 0.3% | 41.00 | 1.6% | +4.5pp |
| Evenepoel Remco | Red Bull-BORA-hansgrohe | – | 2.0% | 0.1% | 5.2% | 301.00 | 0.2% | +5.0pp |
| Pidcock Tom | Pinarello-Q36.5 Pro Cycling Team | – | 3.5% | 0.2% | 5.2% | 34.00 | 1.9% | +3.2pp |
| Ayuso Juan | Lidl-Trek | – | 0.8% | 0.0% | 4.6% | n/a | – | – |
| Ackermann Pascal | Team Jayco-AlUla | 0.0% | 0.8% | 4.3% | 0.1% | 251.00 | 0.3% | +4.0pp |
| Wærenskjold Søren | Uno-X Mobility | 0.9% | 3.8% | 2.9% | 1.5% | 41.00 | 1.6% | +2.2pp |
| Grégoire Romain | Groupama-FDJ United | 2.7% | 1.3% | 0.0% | 3.5% | 126.00 | 0.5% | +3.0pp |
| Oldani Stefano | Caja Rural-Seguros RGA | 0.4% | 2.8% | 0.4% | 0.5% | 301.00 | 0.2% | +2.6pp |
| Fretin Milan | Cofidis | 0.0% | 1.0% | 2.7% | 0.1% | 101.00 | 0.6% | +2.1pp |
| Matthews Michael | Team Jayco-AlUla | 0.4% | 2.5% | 1.8% | 0.8% | 41.00 | 1.6% | +1.0pp |
| Bauhaus Phil | Bahrain Victorious | 0.0% | 0.5% | 2.4% | 0.0% | 251.00 | 0.3% | +2.1pp |
| Martinez Lenny | Bahrain Victorious | – | 0.2% | 0.0% | 2.2% | n/a | – | – |
| Askey Lewis | NSN Cycling Team | 0.4% | 2.2% | 0.7% | 0.3% | 301.00 | 0.2% | +2.0pp |
| Johannessen Tobias Halland | Uno-X Mobility | 4.4% | 0.9% | 0.0% | 2.2% | 301.00 | 0.2% | +2.0pp |
| Vacek Mathias | Lidl-Trek | 1.6% | 2.1% | 0.4% | 1.3% | 34.00 | 1.9% | +0.2pp |
| Carapaz Richard | EF Education-EasyPost | 2.3% | 0.4% | 0.0% | 2.1% | 301.00 | 0.2% | +1.9pp |
| # | Rider | Blend | Cascade | LambdaRank | Podium | Top 15 | Book odds |
|---|---|---|---|---|---|---|---|
| 1 | Kuss Sepp | 1.8% | 5.9% | 1.1% | 10.9% | 23.7% | n/a |
| 2 | Johannessen Tobias Halland | 1.7% | 0.7% | 2.2% | 2.3% | 18.6% | 301.00 |
| 3 | Riccitello Matthew | 1.6% | 4.5% | 1.2% | 7.4% | 31.2% | n/a |
| 4 | Arensman Thymen | 1.6% | 1.5% | 1.9% | 4.8% | 18.2% | 201.00 |
| 5 | van der Poel Mathieu | 1.4% | 2.7% | 1.3% | 6.5% | 18.9% | 7.00 |
| 6 | Gee Derek | 1.2% | 1.1% | 1.4% | 4.7% | 27.5% | 401.00 |
| 7 | Van Wilder Ilan | 1.0% | 1.2% | 1.2% | 3.2% | 22.7% | 501.00 |
| 8 | Grégoire Romain | 1.0% | 1.2% | 1.1% | 4.0% | 18.3% | 126.00 |
| 9 | Hindley Jai | 0.9% | 0.5% | 1.1% | 4.5% | 20.7% | n/a |
| 10 | Carapaz Richard | 0.9% | 1.0% | 1.0% | 3.2% | 20.6% | 301.00 |
| 11 | Vauquelin Kévin | 0.9% | 0.5% | 1.1% | 2.9% | 21.1% | 251.00 |
| 12 | Piganzoli Davide | 0.8% | 0.4% | 1.1% | 4.1% | 29.5% | 801.00 |
| 13 | Jorgenson Matteo | 0.7% | 0.7% | 0.9% | 2.5% | 15.8% | 201.00 |
| 14 | Pedersen Mads | 0.7% | 1.1% | 0.7% | 5.3% | 15.6% | 5.00 |
| 15 | McNulty Brandon | 0.6% | 0.8% | 0.7% | 2.2% | 17.2% | 126.00 |
| 16 | Aranburu Alex | 0.6% | 1.0% | 0.6% | 3.1% | 20.1% | 81.00 |
| 17 | Vacek Mathias | 0.6% | 0.9% | 0.6% | 3.7% | 19.0% | 34.00 |
| 18 | Jegat Jordan | 0.6% | 1.0% | 0.6% | 5.8% | 32.1% | 601.00 |
| 19 | Higuita Sergio | 0.6% | 0.4% | 0.7% | 3.3% | 24.4% | n/a |
| 20 | Bernal Egan | 0.6% | 0.5% | 0.7% | 2.5% | 16.6% | n/a |
| 21 | Healy Ben | 0.6% | 1.3% | 0.5% | 4.0% | 17.1% | 201.00 |
| 22 | Schmid Mauro | 0.6% | 1.1% | 0.5% | 3.4% | 15.8% | 41.00 |
| 23 | Castrillo Pablo | 0.6% | 0.8% | 0.6% | 3.4% | 17.0% | 401.00 |
| 24 | Izagirre Ion | 0.5% | 0.2% | 0.7% | 2.8% | 17.5% | 601.00 |
| 25 | Baudin Alex | 0.5% | 0.5% | 0.5% | 2.4% | 20.4% | 501.00 |
| 26 | Storer Michael | 0.5% | 0.9% | 0.5% | 4.4% | 25.1% | n/a |
| 27 | Tejada Harold | 0.5% | 0.9% | 0.4% | 3.2% | 17.7% | 601.00 |
| 28 | Yates Adam | 0.5% | 0.5% | 0.5% | 1.4% | 16.1% | n/a |
| 29 | Plapp Lucas | 0.4% | 0.9% | 0.4% | 3.0% | 13.5% | n/a |
| 30 | Van Gils Maxim | 0.4% | 0.7% | 0.5% | 2.4% | 12.2% | 101.00 |
| # | Rider | Blend | Cascade | LambdaRank | Podium | Top 15 | Book odds |
|---|---|---|---|---|---|---|---|
| 1 | Philipsen Jasper | 23.8% | 23.5% | 17.6% | 42.2% | 58.3% | 7.00 |
| 2 | Kooij Olav | 16.4% | 17.5% | 11.9% | 28.2% | 38.7% | 9.00 |
| 3 | Pedersen Mads | 11.4% | 16.0% | 7.3% | 29.9% | 58.0% | 5.00 |
| 4 | Pogacar Tadej | 11.0% | 8.8% | 8.7% | 22.4% | 53.5% | 301.00 |
| 5 | van der Poel Mathieu | 10.8% | 16.0% | 6.7% | 37.9% | 63.0% | 7.00 |
| 6 | Girmay Biniam | 6.1% | 5.1% | 4.8% | 17.0% | 45.4% | 21.00 |
| 7 | Kanter Max | 3.9% | 5.0% | 2.6% | 20.3% | 46.1% | 41.00 |
| 8 | Wærenskjold Søren | 3.8% | 3.7% | 2.8% | 11.6% | 32.6% | 41.00 |
| 9 | Pidcock Tom | 3.5% | 0.6% | 3.3% | 7.5% | 41.4% | 34.00 |
| 10 | Oldani Stefano | 2.8% | 0.7% | 2.6% | 4.5% | 34.8% | 301.00 |
| 11 | Matthews Michael | 2.5% | 3.4% | 1.7% | 10.6% | 42.1% | 41.00 |
| 12 | Askey Lewis | 2.2% | 1.6% | 1.8% | 8.6% | 25.1% | 301.00 |
| 13 | Vacek Mathias | 2.1% | 2.0% | 1.6% | 7.5% | 31.6% | 34.00 |
| 14 | Aranburu Alex | 2.0% | 1.5% | 1.6% | 6.8% | 40.5% | 81.00 |
| 15 | Evenepoel Remco | 2.0% | 1.0% | 1.7% | 5.6% | 23.5% | 301.00 |
| 16 | Van Gils Maxim | 1.6% | 0.9% | 1.4% | 7.4% | 27.0% | 101.00 |
| 17 | Wright Fred | 1.4% | 1.1% | 1.1% | 9.2% | 34.0% | 41.00 |
| 18 | Grégoire Romain | 1.3% | 0.5% | 1.2% | 3.3% | 19.2% | 126.00 |
| 19 | Seixas Paul | 1.2% | 2.1% | 0.7% | 6.4% | 27.9% | n/a |
| 20 | Vauquelin Kévin | 1.0% | 0.3% | 0.9% | 2.3% | 11.7% | 251.00 |
| 21 | Fretin Milan | 1.0% | 2.7% | 0.3% | 5.4% | 29.5% | 101.00 |
| 22 | Stuyven Jasper | 0.9% | 0.3% | 0.8% | 2.1% | 27.8% | 34.00 |
| 23 | Godon Dorian | 0.9% | 1.3% | 0.5% | 5.8% | 30.1% | 41.00 |
| 24 | del Toro Isaac | 0.9% | 0.5% | 0.7% | 1.9% | 9.9% | 301.00 |
| 25 | Johannessen Tobias Halland | 0.9% | 0.6% | 0.7% | 3.2% | 20.3% | 301.00 |
| 26 | Ayuso Juan | 0.8% | 0.3% | 0.7% | 2.9% | 12.2% | n/a |
| 27 | Ackermann Pascal | 0.8% | 1.3% | 0.5% | 4.2% | 26.9% | 251.00 |
| 28 | Pluimers Rick | 0.8% | 0.4% | 0.6% | 3.6% | 34.2% | 201.00 |
| 29 | Skjelmose Mattias | 0.7% | 0.8% | 0.5% | 2.2% | 14.4% | 501.00 |
| 30 | Bittner Pavel | 0.6% | 1.5% | 0.2% | 6.0% | 21.7% | 201.00 |
| # | Rider | Blend | Cascade | LambdaRank | Podium | Top 15 | Book odds |
|---|---|---|---|---|---|---|---|
| 1 | Kooij Olav | 48.4% | 48.5% | 38.9% | 73.4% | 87.5% | 9.00 |
| 2 | Philipsen Jasper | 30.5% | 29.9% | 24.7% | 56.2% | 86.3% | 7.00 |
| 3 | Kanter Max | 6.1% | 4.6% | 5.3% | 21.1% | 82.0% | 41.00 |
| 4 | Girmay Biniam | 5.7% | 4.5% | 4.9% | 18.0% | 82.5% | 21.00 |
| 5 | Ackermann Pascal | 4.3% | 1.8% | 4.1% | 10.0% | 75.2% | 251.00 |
| 6 | Pedersen Mads | 3.4% | 3.3% | 2.7% | 14.2% | 57.8% | 5.00 |
| 7 | Pogacar Tadej | 3.2% | 8.2% | 1.4% | 11.7% | 31.6% | 301.00 |
| 8 | Wærenskjold Søren | 2.9% | 2.7% | 2.4% | 9.3% | 73.6% | 41.00 |
| 9 | Fretin Milan | 2.7% | 2.9% | 2.2% | 11.7% | 77.4% | 101.00 |
| 10 | Bauhaus Phil | 2.4% | 1.2% | 2.2% | 11.4% | 59.9% | 251.00 |
| 11 | van der Poel Mathieu | 1.8% | 2.3% | 1.3% | 9.7% | 46.1% | 7.00 |
| 12 | Matthews Michael | 1.8% | 1.9% | 1.4% | 7.5% | 62.4% | 41.00 |
| 13 | Ballerini Davide | 1.6% | 0.9% | 1.4% | 8.4% | 46.7% | 126.00 |
| 14 | Bittner Pavel | 1.5% | 1.8% | 1.1% | 8.1% | 74.5% | 201.00 |
| 15 | Wright Fred | 0.7% | 1.0% | 0.5% | 6.1% | 31.5% | 41.00 |
| 16 | Askey Lewis | 0.7% | 1.1% | 0.4% | 5.8% | 41.6% | 301.00 |
| 17 | van Dijke Tim | 0.6% | 0.6% | 0.5% | 2.9% | 37.1% | 401.00 |
| 18 | Vacek Mathias | 0.4% | 0.3% | 0.4% | 2.4% | 15.5% | 34.00 |
| 19 | Oldani Stefano | 0.4% | 0.1% | 0.4% | 3.4% | 41.5% | 301.00 |
| 20 | Bol Cees | 0.3% | 0.3% | 0.2% | 1.9% | 27.4% | 301.00 |
| 21 | Stuyven Jasper | 0.3% | 0.1% | 0.3% | 1.3% | 31.5% | 34.00 |
| 22 | del Toro Isaac | 0.3% | 0.3% | 0.2% | 0.7% | 4.2% | 301.00 |
| 23 | Cort Magnus | 0.3% | 0.2% | 0.2% | 1.4% | 31.0% | 23.00 |
| 24 | Artz Huub | 0.2% | 0.3% | 0.1% | 3.5% | 28.8% | 151.00 |
| 25 | Pidcock Tom | 0.2% | 0.1% | 0.2% | 1.1% | 6.7% | 34.00 |
| 26 | Teunissen Mike | 0.2% | 0.1% | 0.2% | 1.2% | 38.4% | 301.00 |
| 27 | Godon Dorian | 0.2% | 0.1% | 0.2% | 1.1% | 22.9% | 41.00 |
| 28 | Russo Clément | 0.2% | 0.2% | 0.1% | 0.6% | 35.8% | 301.00 |
| 29 | Pluimers Rick | 0.2% | 0.2% | 0.1% | 1.0% | 18.7% | 201.00 |
| 30 | Aranburu Alex | 0.2% | 0.1% | 0.1% | 2.0% | 10.1% | 81.00 |
| # | Rider | Blend | Cascade | LambdaRank | Podium | Top 15 | Book odds |
|---|---|---|---|---|---|---|---|
| 1 | Pogacar Tadej | 36.6% | 32.5% | 30.6% | 62.6% | 81.7% | 301.00 |
| 2 | del Toro Isaac | 6.9% | 2.2% | 6.7% | 12.8% | 56.2% | 301.00 |
| 3 | Seixas Paul | 6.8% | 6.1% | 5.7% | 14.2% | 59.7% | n/a |
| 4 | van der Poel Mathieu | 6.2% | 12.0% | 3.5% | 25.7% | 45.1% | 7.00 |
| 5 | Evenepoel Remco | 5.2% | 2.7% | 4.8% | 14.1% | 52.6% | 301.00 |
| 6 | Pidcock Tom | 5.2% | 2.0% | 5.0% | 7.0% | 54.4% | 34.00 |
| 7 | Ayuso Juan | 4.6% | 1.5% | 4.5% | 7.4% | 48.6% | n/a |
| 8 | Pedersen Mads | 3.6% | 5.4% | 2.5% | 13.6% | 31.5% | 5.00 |
| 9 | Grégoire Romain | 3.5% | 4.0% | 2.7% | 10.7% | 39.6% | 126.00 |
| 10 | Martinez Lenny | 2.2% | 2.4% | 1.8% | 6.4% | 43.1% | n/a |
| 11 | Johannessen Tobias Halland | 2.2% | 1.3% | 2.0% | 9.0% | 38.9% | 301.00 |
| 12 | Carapaz Richard | 2.1% | 1.3% | 1.9% | 4.9% | 36.5% | 301.00 |
| 13 | Philipsen Jasper | 2.0% | 2.8% | 1.4% | 8.2% | 21.3% | 7.00 |
| 14 | Kuss Sepp | 1.9% | 6.8% | 0.3% | 7.9% | 17.2% | n/a |
| 15 | Skjelmose Mattias | 1.7% | 1.2% | 1.5% | 5.2% | 41.7% | 501.00 |
| 16 | Jorgenson Matteo | 1.6% | 1.2% | 1.4% | 4.0% | 27.4% | 201.00 |
| 17 | Aranburu Alex | 1.5% | 1.4% | 1.2% | 7.0% | 29.8% | 81.00 |
| 18 | Wærenskjold Søren | 1.5% | 3.5% | 0.7% | 7.7% | 18.8% | 41.00 |
| 19 | Vacek Mathias | 1.3% | 1.6% | 1.0% | 6.6% | 22.7% | 34.00 |
| 20 | Girmay Biniam | 1.3% | 1.9% | 0.9% | 5.4% | 16.6% | 21.00 |
| 21 | Vauquelin Kévin | 1.2% | 0.7% | 1.0% | 4.5% | 16.8% | 251.00 |
| 22 | Higuita Sergio | 1.1% | 0.5% | 1.0% | 2.2% | 17.1% | n/a |
| 23 | Hindley Jai | 1.0% | 0.2% | 1.0% | 2.0% | 12.4% | n/a |
| 24 | Riccitello Matthew | 1.0% | 1.9% | 0.5% | 3.4% | 14.7% | n/a |
| 25 | Bernal Egan | 0.9% | 0.6% | 0.8% | 2.6% | 20.0% | n/a |
| 26 | Van Wilder Ilan | 0.9% | 1.0% | 0.7% | 2.6% | 28.0% | 501.00 |
| 27 | Piganzoli Davide | 0.8% | 0.6% | 0.7% | 4.2% | 40.0% | 801.00 |
| 28 | Matthews Michael | 0.8% | 1.3% | 0.5% | 6.9% | 30.5% | 41.00 |
| 29 | Van Gils Maxim | 0.7% | 0.9% | 0.6% | 3.1% | 14.0% | 101.00 |
| 30 | Gee Derek | 0.7% | 0.4% | 0.7% | 3.4% | 14.8% | 401.00 |
Angle
The top 8 on GC after stage 16 removed — everyone below that is far enough back to be given freedom. Remaining riders under the 'Rouleur break' read; the conditional column renormalizes their win probability over this pool (P(win | a non-GC rider wins), pool holds 38% of the unconditional book). The model can't see who actually makes the break — read this as a fitness-for-the-day ranking, not break-formation odds.
| Rider | Team | Rouleur break win p | P(win | break wins) | Book odds | Book devig |
|---|---|---|---|---|---|
| Kuss Sepp | Team Visma | Lease a Bike | 1.8% | 4.6% | n/a | – |
| Johannessen Tobias Halland | Uno-X Mobility | 1.7% | 4.4% | 301 | 0.2% |
| Riccitello Matthew | Decathlon CMA CGM Team | 1.6% | 4.2% | n/a | – |
| Arensman Thymen | Netcompany INEOS | 1.6% | 4.1% | 201 | 0.3% |
| van der Poel Mathieu | Alpecin-Premier Tech | 1.4% | 3.7% | 7 | 9.3% |
| Gee Derek | Lidl-Trek | 1.2% | 3.0% | 401 | 0.2% |
| Van Wilder Ilan | Soudal Quick-Step | 1.0% | 2.7% | 501 | 0.1% |
| Grégoire Romain | Groupama-FDJ United | 1.0% | 2.7% | 126 | 0.5% |
| Hindley Jai | Red Bull-BORA-hansgrohe | 0.9% | 2.3% | n/a | – |
| Carapaz Richard | EF Education-EasyPost | 0.9% | 2.3% | 301 | 0.2% |
| Vauquelin Kévin | Netcompany INEOS | 0.9% | 2.3% | 251 | 0.3% |
| Piganzoli Davide | Team Visma | Lease a Bike | 0.8% | 2.2% | 801 | 0.1% |
| Jorgenson Matteo | Team Visma | Lease a Bike | 0.7% | 1.9% | 201 | 0.3% |
| Pedersen Mads | Lidl-Trek | 0.7% | 1.8% | 5 | 13.0% |
Angle
Shortest book prices for the stage, with our blend win probability (and rank) under each scenario. 'free' = outside the excluded GC pool (top 8 on GC after stage 16), i.e. the type the peloton might let up the road.
| Rider | Book odds | Devig | Rouleur break | Hard-men sprint | Sprint survives | GC status |
|---|---|---|---|---|---|---|
| Pedersen Mads | 5.00 | 13.0% | 0.7% (#22) | 11.4% (#3) | 3.4% (#6) | free |
| van der Poel Mathieu | 7.00 | 9.3% | 1.4% (#11) | 10.8% (#5) | 1.8% (#11) | free |
| Philipsen Jasper | 7.00 | 9.3% | 0.1% (#99) | 23.8% (#1) | 30.5% (#2) | free |
| Kooij Olav | 9.00 | 7.2% | 0.1% (#106) | 16.4% (#2) | 48.4% (#1) | free |
| Girmay Biniam | 21.00 | 3.1% | 0.1% (#81) | 6.1% (#6) | 5.7% (#4) | free |
| Ganna Filippo | 23.00 | 2.8% | 0.2% (#64) | 0.1% (#72) | 0.1% (#42) | free |
| Cort Magnus | 23.00 | 2.8% | 0.0% (#118) | 0.2% (#59) | 0.3% (#23) | free |
| Simmons Quinn | 29.00 | 2.2% | 0.2% (#59) | 0.2% (#64) | 0.0% (#53) | free |
| Asgreen Kasper | 29.00 | 2.2% | 0.2% (#61) | 0.2% (#56) | 0.1% (#31) | free |
| Vacek Mathias | 34.00 | 1.9% | 0.6% (#25) | 2.1% (#13) | 0.4% (#18) | free |
Angle
The stage-4 lesson as a filter: no terrain reading means 'selective enough to shed the fast finishers but won by a punchy non-climber', so this table synthesizes it. We drop the GC pool (top 8 on GC after stage 16) and every pure sprinter, then renormalize each reading over the survivors — P(win | the winner is a punchy non-climber). Pure sprinter is the model's own verdict, not a hand list: probability under a flat-sprint anchor of this exact field >= 1% while retaining under 8% of it when the day turns selective ('break_day' read) — that collapse is what separates a Merlier from a Pedersen. Classified out: Ackermann Pascal, Ballerini Davide, Bauhaus Phil, Bittner Pavel, Fretin Milan, Girmay Biniam, Kanter Max, Kooij Olav, Philipsen Jasper. Pool holds Rouleur break 38% / Hard-men sprint 61% / Sprint survives 17% of each reading's unconditional book.
| Rider | Team | P(win | pool) Rouleur break | P(win | pool) Hard-men sprint | P(win | pool) Sprint survives | Book odds | Book devig |
|---|---|---|---|---|---|---|
| Pedersen Mads | Lidl-Trek | 1.9% | 18.7% | 19.6% | 5 | 13.0% |
| van der Poel Mathieu | Alpecin-Premier Tech | 3.7% | 17.7% | 10.5% | 7 | 9.3% |
| Wærenskjold Søren | Uno-X Mobility | 0.9% | 6.2% | 16.7% | 41 | 1.6% |
| Matthews Michael | Team Jayco-AlUla | 0.4% | 4.2% | 10.5% | 41 | 1.6% |
| Kuss Sepp | Team Visma | Lease a Bike | 4.7% | 0.8% | 0.1% | n/a | – |
| Oldani Stefano | Caja Rural-Seguros RGA | 0.4% | 4.6% | 2.2% | 301 | 0.2% |
| Johannessen Tobias Halland | Uno-X Mobility | 4.5% | 1.4% | 0.2% | 301 | 0.2% |
| Riccitello Matthew | Decathlon CMA CGM Team | 4.2% | 0.2% | 0.1% | n/a | – |
| Arensman Thymen | Netcompany INEOS | 4.2% | 0.2% | 0.1% | 201 | 0.3% |
| Wright Fred | Pinarello-Q36.5 Pro Cycling Team | 0.5% | 2.3% | 4.0% | 41 | 1.6% |
| Askey Lewis | NSN Cycling Team | 0.4% | 3.7% | 3.9% | 301 | 0.2% |
| Vacek Mathias | Lidl-Trek | 1.6% | 3.5% | 2.3% | 34 | 1.9% |
| van Dijke Tim | Red Bull-BORA-hansgrohe | 0.1% | 0.6% | 3.4% | 401 | 0.2% |
| Aranburu Alex | Cofidis | 1.6% | 3.3% | 0.9% | 81 | 0.8% |
| Gee Derek | Lidl-Trek | 3.0% | 0.4% | 0.1% | 401 | 0.2% |
| Van Wilder Ilan | Soudal Quick-Step | 2.7% | 0.5% | 0.1% | 501 | 0.1% |
After the stage
| Pos | Rider | Team | Rouleur break win p | Hard-men sprint win p | Sprint survives win p | Hindsight (post-hoc) win p |
|---|---|---|---|---|---|---|
| 1 | Philipsen Jasper | Alpecin-Premier Tech | 0.1% | 23.8% | 30.5% | 2.0% |
| 2 | Schmid Mauro | Team Jayco-AlUla | 0.6% | 0.2% | 0.0% | 0.3% |
| 3 | Kooij Olav | Decathlon CMA CGM Team | 0.1% | 16.4% | 48.4% | 0.4% |
| 4 | Askey Lewis | NSN Cycling Team | 0.1% | 2.2% | 0.7% | 0.3% |
| 5 | Pluimers Rick | Tudor Pro Cycling Team | 0.1% | 0.8% | 0.2% | 0.2% |
| 6 | Russo Clément | Groupama-FDJ United | 0.0% | 0.2% | 0.2% | 0.1% |
| 7 | Artz Huub | Lotto-Intermarché | 0.1% | 0.3% | 0.2% | 0.2% |
| 8 | Pedersen Mads | Lidl-Trek | 0.7% | 11.4% | 3.4% | 3.6% |
| 9 | Matthews Michael | Team Jayco-AlUla | 0.1% | 2.5% | 1.8% | 0.8% |
| 10 | Gate Aaron | XDS Astana Team | 0.1% | 0.2% | 0.1% | 0.1% |
| Scenario | Blend win p | Book devig |
|---|---|---|
| Rouleur break | 0.06% | 9.29% |
| Hard-men sprint | 23.76% | 9.29% |
| Sprint survives | 30.47% | 9.29% |
| Hindsight (post-hoc) | 2.03% | 9.29% |
Reflection
The road split the difference between our three reads and landed cleanly on none of them. A 26-rider front group — not the pure rouleur escape we drew, but a strong mixed break carrying the fast climbers-sprinters (Philipsen, Kooij, Pedersen, van der Poel, Matthews) alongside the rouleurs — went clear over the early Savoy climbs and stayed to Voiron, where Philipsen won the reduced sprint. The peloton shattered behind (only 34 riders within two minutes of the win) and the entire GC group sat up, rolling in 8:46 down. GC is frozen going into the first Alpine summit: identical top 8, Pogačar still in yellow, Evenepoel +4:32. So it was neither the rouleur break (the fast men made it) nor a full bunch sprint (the field was blown apart in between).
The hard-men sprint read was the best board of the three and it beat the book on the actual winner: it made Philipsen its outright #1 at 17.6% honest (23.8% raw blend) against the book's 9.3% — the model's single top pick, and it won. The sprint-survives read priced Philipsen almost as richly (24.6% honest) but ranked the purer sprinter Kooij ahead at 39%; Kooij finished third, and it was exactly the reduced-group nature of the day — which rewards the punchier Philipsen over Kooij — that hard-men-sprint captured by putting Philipsen on top. The rouleur break was the wrong branch (Philipsen 0.1%): its standings cut correctly freed the rouleurs, but its terrain profile buried the fast men, and the fast men are who won. The book's own 5.0 favourite Pedersen came eighth.
Course knowledge was not the constraint here — it actively hurt. The hindsight run, built from the five nearest finish-curve neighbours (van Aert's 2022 huge-break stage, van der Poel 2025, Cort's Paris-Nice, Brennan and Gilmore at Catalunya) and correctly reconstructing a rolling-breakaway profile (rolling .25 / breakaway .49 / mountain .16 / hilly .10), gives Philipsen just 1.6% (#13) and crowns Pogačar at 29.6% — a man who never once contested the stage. On a literal rolling-breakaway profile the model ranks by rider-course fit and the GC star fits best, so knowing the true terrain made the board worse, not better. The winning move was to read the day as a reduced SPRINT (which holds the fast men up), not the breakaway it literally was.
Carry-forward. (1) This is the GT hard-day hole once more: the model cannot see that the GC teams will grant a big break, nor that the break can swallow the fastest men, so on a transition stage the honest board is the reduced-sprint archetype and not the post-hoc terrain — we weighted the hard-men-sprint read as the base case and it paid. (2) The punchy filter mis-fired: it classified Philipsen and Kooij as pure sprinters that 'collapse' when the day turns selective, yet they made the break and finished 1st and 3rd. The collapse logic assumes the sprinters get dropped; on a day the break carries them up the road, that assumption inverts — a watch item for the next transition stage. (3) GC is untouched, so the stage-16 standings still define tomorrow's break-cut into the Alps.