TennisEdge's AI model finds no betting value in J. Fearnley vs S. Ofner (Montreal on 2026-08-01) — the market is efficient, with J. Fearnley 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 | J. Fearnley | S. Ofner |
|---|---|---|
| Elo ratings-based, surface-weighted |
46.1% | 53.9% |
| Machine learning 29-feature gradient-boosted model |
62.1% | 37.9% |
| Serve / return point-by-point simulation |
35.2% | 64.8% |
| Blend what the bot actually prices off |
46.1% | 53.9% |
| Market implied de-vigged from the opening odds |
58.9% | 41.1% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| J. Fearnley | hard | 1713 | 72–38 | 110 |
| S. Ofner | clay | 1701 | 100–69 | 169 |
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 Aug. 1, 2026.
J. Fearnley leads 1–0 over 1 meeting. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 1, 2026 | Montreal | hard | J. Fearnley | 2 - 0 |
| W | S. Ofner | 2 - 0 | Aug. 1, 2026 |
| L | A. Michelsen | 2 - 0 | July 26, 2026 |
| W | M. Zheng | 2 - 0 | July 25, 2026 |
| W | J. Wolf | 2 - 1 | July 24, 2026 |
| W | T. Boyer | 2 - 0 | July 23, 2026 |
| L | J. Fearnley | 2 - 0 | Aug. 1, 2026 |
| L | A. Molcan | 2 - 0 | July 21, 2026 |
| L | T. A. Tirante | 0 - 2 | July 14, 2026 |
| L | D. Rincon | 2 - 1 | July 8, 2026 |
| W | N. Gombos | 0 - 2 | July 7, 2026 |
All Aug. 1, 2026 results & picks →
J. Fearnley (A) vs S. Ofner (B) — Montreal, hard, Final
MODELS
ELO/model A 46.1% | ML A 62.0% | market implied A 58.9% (odds 1.57 / 2.25)
Big split: ELO favours Ofner, ML+market favour Fearnley. value_pick B @2.25 (edge ~9-13%) is ELO/model-driven.
ELO vs RANKING / TRUST
Ofner rank 82, Fearnley rank 123 — ELO leans Ofner partly on rank + a clay-built rating.
Ofner is an Austrian clay specialist; his ELO/wins are clay-farmed and he is transitioning clay->hard here.
ELO therefore understates Fearnley on this surface. The B "value" is a contaminated-model artifact, not a real edge.
RECENT FORM (last 5 each)
Fearnley (hard, SHARPENING): Bloomfield Hills Final (L to Michelsen), SF beat Zheng 1803 DOM 6-3 6-4,
QF Wolf, R16 Boyer, R32 Bertola — deep hard-court run + Newport grass title 07-12. In form, match-sharp on hard.
Ofner (clay, FADING): L Molcan (07-21), L Tirante (07-14), L Braunschweig, mixed — recent losses, poor shape,
rusty (11 days since last match), off-surface.
FATIGUE
Fearnley last match 07-26 (6 days ago) — rested now; last week's b2b load irrelevant with 6d gap.
Ofner 11 days off, 1 match/14d — rusty, not tired. No extreme spot either side. Roughly equal.
NARRATIVE / SURFACE
TRANSITION: Ofner clay->hard = depressed level + wider margins, a risk on his side, mild support for Fearnley ML.
No credible letdown/venue signal in data (neither player has Montreal matches this week in the window).
GATES
G1 edge-origin: FAIL — only clean numeric edge is B via ELO/model, but that model is clay-farmed/contaminated;
A-side edge vs book (62% - 59% = ~3%) is below the 5% bar and the 8% book-priced bar. No staked edge survives.
G2 consensus: PASS (stake-tier only) — ML+market+qualitative all favour A = efficient on the winner.
G3 fatal-risk: n/a — ML-scope only; no bet staked.
MARKET CHOICE
alt_markets: none at analysis time
ML book-only, A-side edge ~3% < 8% book bar; B-side edge is a contaminated-ELO mirage. No margin market quoted.
AH: not quoted
VERDICT: PASS. Qualitative read agrees with ML/market that Fearnley is the rightful favourite on hard;
the exchange has no price and the book edge on A is too thin. The apparent B value is manufactured by a
clay-farmed ELO against a surface-transitioning, out-of-form clay specialist — reject, do not fade the favourite.