TennisEdge's AI model finds no betting value in P. Marcinko vs K. Birrell (Eastbourne on 2026-06-24) — the market is efficient, with K. Birrell favoured at 1.57.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | P. Marcinko | K. Birrell |
|---|---|---|
| Elo ratings-based, surface-weighted |
34.8% | 65.2% |
| Machine learning 29-feature gradient-boosted model |
49.7% | 50.3% |
| Serve / return point-by-point simulation |
57.3% | 42.7% |
| Blend what the bot actually prices off |
47.3% | 52.7% |
| Market implied de-vigged from the opening odds |
39.7% | 60.3% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| P. Marcinko | clay | 1727 | 89–46 | 135 |
| K. Birrell | hard | 1773 | 149–98 | 247 |
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 17, 2026.
Level at 1–1 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| June 24, 2026 | Eastbourne | grass | P. Marcinko | 2 - 0 |
| Feb. 24, 2026 | WTA Austin | hard | K. Birrell | 2 - 1 |
| L | T. Zidansek | 2 - 0 | July 17, 2026 |
| W | E. Avanesyan | 0 - 2 | July 15, 2026 |
| W | N. Podoroska | 1 - 2 | July 13, 2026 |
| L | C. Burel | 1 - 2 | July 7, 2026 |
| L | S. Kenin | 0 - 2 | June 30, 2026 |
| L | V. Golubic | 0 - 2 | Aug. 2, 2026 |
| L | Y. Ma | 2 - 0 | July 29, 2026 |
| L | S. Cirstea | 2 - 0 | July 2, 2026 |
| W | A. Korneeva | 2 - 1 | June 30, 2026 |
| L | P. Marcinko | 2 - 0 | June 24, 2026 |
All June 24, 2026 results & picks →
The model consensus identifies a marginal 2.1% edge on P. Marcinko, primarily driven by the ML model and a market trend indicating that K. Birrell is over-priced due to recency bias. However, Birrell is coming off a high-quality straight-sets victory over B. Krejcikova today, suggesting a 'hot hand' and high confidence level that pure statistical models may not fully capture. While the market trend signals value on the underdog, the combination of a tiny edge (below reliable betting thresholds), conflicting model signals (Elo vs ML), and Birrell's momentum makes this a high-variance spot. The recommended action is to pass, as the risk outweighs the slim mathematical advantage.