TennisEdge's AI model finds no betting value in A. Fery vs O. Virtanen (Wimbledon on 2026-07-02) — the market is efficient, with O. Virtanen 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 | A. Fery | O. Virtanen |
|---|---|---|
| Elo ratings-based, surface-weighted |
27.4% | 72.6% |
| Machine learning 29-feature gradient-boosted model |
27.1% | 72.9% |
| Serve / return point-by-point simulation |
32.6% | 67.4% |
| Blend what the bot actually prices off |
32.2% | 67.8% |
| Market implied de-vigged from the opening odds |
39.5% | 60.5% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| A. Fery | hard | 1722 | 118–49 | 167 |
| O. Virtanen | hard | 1724 | 122–77 | 199 |
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 March 19, 2026.
Level at 1–1 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| July 2, 2026 | Wimbledon | grass | A. Fery | 3 - 1 |
| June 5, 2022 | Nottingham | grass | O. Virtanen | 2 - 1 |
| L | A. Zverev | 0 - 3 | July 10, 2026 |
| W | F. Cobolli | 0 - 3 | July 8, 2026 |
| W | G. Dimitrov | 2 - 3 | July 6, 2026 |
| W | Z. Bergs | 2 - 3 | July 4, 2026 |
| W | O. Virtanen | 3 - 1 | July 2, 2026 |
| L | T. Skatov | 2 - 0 | July 30, 2026 |
| W | F. Romano | 0 - 2 | July 28, 2026 |
| L | D. Dedura | 2 - 1 | July 26, 2026 |
| W | M. Kasnikowski | 0 - 2 | July 25, 2026 |
| W | H. Squire | 0 - 0 | July 24, 2026 |
All July 2, 2026 results & picks →
═══════════════════════════════════════════
A. Fery (A) vs O. Virtanen (B)
Wimbledon · grass · 1/32-finals · 2026-07-02
═══════════════════════════════════════════
MODELS
ELO A 27.4% | B 72.6%
ML A 27.1% | B 72.9%
S/R A 32.6% | B 67.4%
MODEL A 32.2% | B 67.8%
MARKET A 2.40 (impl 39.5%) | B 1.57 (impl 60.5%)
Value pick B, edge +7.3%. Bot: BET 2.5u on B.
-> Every model AND the market agree on B. Full consensus.
FORM (last ~14d, grass)
FERY: W Dzumhur(1440,Weak) 3-6 6-2 6-2 6-1 GRIND->control;
L J.M.Cerundolo(1396); W Burruchaga(1545,Avg) GRIND;
L F.Cerundolo(1606); W Mannarino(1807,Strong) 7-6 6-4 GRIND;
W Samuel(1581) 6-0 6-2 DOM. London QF run. Solid but grind-heavy,
best win = Mannarino (Strong).
VIRTANEN: W Shelton(1896,ELITE) 6-4 3-6 6-7 6-2 7-6 -> 5-SET MARATHON 2d ago;
W Final vs Budkov Kjaer 7d ago; W SF Nardi; W QF Martinez;
L Nottingham Final vs O'Connell 6-7 6-7. Elite-level grass form,
multiple finals, clearly the stronger player.
FATIGUE
Virtanen 1041 court-min / 9 matches / 27 sets / 5 back-to-back (14d)
Fery 608 court-min / 6 matches / 18 sets / 2 back-to-back (14d)
Court-time gap 433 min (>60) -> HEAVY disadvantage Virtanen.
Only 2 days rest after a 5-set war vs Shelton. Real, under-priced load.
NARRATIVE
Virtanen: WEAK let-down (won a Final 7d ago) + fresh Elite 5-setter.
Fery: no Final; steady but unspectacular, lost to weak Cerundolo.
Venue: Virtanen deep Wimbledon/grass momentum — amplifies his talent edge
but cannot offset acute marathon fatigue.
VERDICT — PASS
The full model+market consensus on Virtanen is built on talent, and it is
correct on talent. But the models do not price the 5-set marathon + 433-min
court-time gap + 2-day turnaround. That fatigue erases the claimed +7.3%
value on B at 1.57 — this is NOT a bet on B. Yet Fery's own form (grind wins,
a loss to a weak Cerundolo) is not strong enough to confidently back the
2.40 underdog either. Edge on B killed, no validated edge on A. No play.
═══════════════════════════════════════════