TennisEdge's AI model finds no betting value in S. Bejlek vs L. Siegemund (Eastbourne on 2026-06-22) — the market is efficient, with L. Siegemund favoured at 1.67.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | S. Bejlek | L. Siegemund |
|---|---|---|
| Elo ratings-based, surface-weighted |
13.6% | 86.5% |
| Machine learning 29-feature gradient-boosted model |
49.1% | 50.9% |
| Blend what the bot actually prices off |
31.3% | 68.7% |
| Market implied de-vigged from the opening odds |
43.2% | 56.8% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| S. Bejlek | clay | 1747 | 137–58 | 195 |
| L. Siegemund | hard | 1715 | 70–62 | 132 |
Our own ratings, computed from match results since 2018. Players with fewer than 5 rated matches are omitted rather than shown at a provisional starting value. Last updated July 24, 2026.
Level at 1–1 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| June 23, 2026 | Eastbourne | grass | S. Bejlek | 2 - 1 |
| May 6, 2026 | Rome | clay | L. Siegemund | 2 - 0 |
| W | Xiy. Wang | 1 - 2 | Aug. 2, 2026 |
| L | L. Tagger | 2 - 0 | July 24, 2026 |
| W | M. Timofeeva | 0 - 2 | July 23, 2026 |
| W | A. Blinkova | 1 - 2 | July 21, 2026 |
| L | C. Tauson | 2 - 0 | July 17, 2026 |
| L | E. Mertens | 2 - 0 | June 30, 2026 |
| L | S. Bejlek | 2 - 1 | June 23, 2026 |
| L | A. Anisimova | 0 - 2 | June 10, 2026 |
| W | F. Jones | 2 - 0 | June 9, 2026 |
| L | N. Osaka | 0 - 2 | May 26, 2026 |
All June 22, 2026 results & picks →
The mathematical consensus identifies a marginal 1.6% edge for Siegemund, supported qualitatively by her status as an experienced grass-courter facing a clay specialist (Bejlek) who is 0-2 on the surface. However, the significant divergence between the ELO model (13.6% for Bejlek) and the ML model (49.1%) introduces high variance, suggesting the true probability is uncertain. With a Kelly suggestion of 0.0 units and a low odds profile, the risk/reward ratio is poor.