TennisEdge's AI model finds no betting value in L. Boisson vs A. Ruzic (Toronto on 2026-08-02) — the market is efficient, with A. Ruzic favoured at 1.73.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | L. Boisson | A. Ruzic |
|---|---|---|
| Elo ratings-based, surface-weighted |
22.2% | 77.8% |
| Machine learning 29-feature gradient-boosted model |
29.9% | 70.1% |
| Serve / return point-by-point simulation |
49.0% | 51.0% |
| Blend what the bot actually prices off |
37.5% | 62.5% |
| Market implied de-vigged from the opening odds |
45.2% | 54.8% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| L. Boisson | clay | 1739 | 82–39 | 121 |
| A. Ruzic | hard | 1784 | 114–50 | 164 |
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 May 26, 2026.
A. Ruzic leads 1–0 over 1 meeting. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 3, 2026 | Toronto | hard | A. Ruzic | 2 - 1 |
| L | A. Ruzic | 1 - 2 | Aug. 3, 2026 |
| W | E. Arango | 0 - 2 | Aug. 1, 2026 |
| L | E. Rybakina | 1 - 2 | June 30, 2026 |
| L | S. Sierra | 1 - 2 | June 10, 2026 |
| L | A. Kalinskaya | 2 - 0 | May 26, 2026 |
| W | L. Boisson | 1 - 2 | Aug. 3, 2026 |
| L | D. Salkova | 0 - 2 | July 20, 2026 |
| L | J. Ostapenko | 2 - 0 | July 1, 2026 |
| W | D. Semenistaja | 2 - 1 | June 29, 2026 |
| L | P. Marcinko | 1 - 2 | June 22, 2026 |
All Aug. 2, 2026 results & picks →
BOISSON (A) vs RUZIC (B) — WTA Toronto, hard, 1/64-finals
═══════════════════════════════════════════════════════
MODELS vs MARKET
ELO(A) 0.222 | ML(A) 0.299 | blended 0.375
Betfair ML: A 2.24 / B 1.46 → devigged ≈ Boisson 39.5% / Ruzic 60.5%
Book quote (A 2.1 / B 1.73) is STALE — the "value_pick B / +7.6%" edge is a
book-vs-exchange artifact, NOT real value. Exchange has Ruzic far shorter (1.46 vs 1.73).
Model (0.375) and devigged exchange (0.395) agree Boisson is a ~38-40% underdog. Consensus.
RECENT FORM (last 5 each)
Boisson: 08-01 WON Final vs Arango(1592) 1-6 6-7tb GRIND; 06-30 L Rybakina(1816 Elite)
4-6 6-1 3-6; 06-10 L Sierra(1611) 4-6 7-5 3-6; 05-26 L Kalinskaya(1719) 6-2 6-2;
05-19 L Mboko(1704) 6-4 6-3. Shape: FADING on tour — the Toronto title is an outlier
after a long losing run; historically weak on hard (ELO 0.22 credible for surface).
Ruzic: 07-20 L Salkova(1577) 6-7 4-6; 07-01 L Ostapenko(1764) 6-2 6-0; 06-29 W
Semenistaja(1536); 06-22 L Marcinko(1458) 6-7 6-4 6-7; 06-16 L McNally(1740) 6-3 6-0.
Shape: FADING (lost 3 of last 4) after strong June grass run (Nottingham+London titles).
PATH / SHARPNESS
Boisson: on hard already, match-sharp, won title 08-01 — BUT letdown/short-rest spot
(won Final yesterday; age 23, baseline recovery; ~2-set win, not extreme fatigue).
Ruzic: last match 07-20 (13d off) + last two events clay/grass → TRANSITION to hard;
mild rust. Nets against Boisson's letdown → roughly cancels.
FATIGUE (Step 2)
Boisson last match 08-01 (76min recorded), Ruzic 07-20. Neither at an extreme spot;
Boisson's next-day title win = possible LETDOWN but she's the priced underdog anyway,
so it cannot manufacture edge on the favourite side.
GATES
G1 edge-origin: FAIL — no numeric model-vs-exchange disagreement pre-narrative.
Back B @1.46: edge -7%. Back A @2.24: need 0.509, est ~0.40 → edge negative. Both fail 5% bar.
G2 consensus: PASS(tier) — full model+exchange consensus on ~60/40 Ruzic; no lone-model split to exploit.
G3 fatal-risk: n/a (ML, no bet)
MARKET CHOICE
alt_markets: handicap/totals null; only betfair_ml usable → ML only.
Back A @2.24 negative edge; back B @1.46 negative edge. No stakeable market.
AH: not quoted
VERDICT: PASS. Book "value_pick B" is a stale-quote mirage; the sharp exchange prices
Ruzic at 1.46 (negative to back) and Boisson at 2.24 (still short of her ~40% true prob).
Model and market converge — nothing to bet.