TENNISEDGE.GG

A. Ito vs K. Sebov — Prediction & Analysis

Toronto · · Aug. 2, 2026 · 18:00 UTC

TennisEdge's AI model finds no betting value in A. Ito vs K. Sebov (Toronto on 2026-08-02) — the market is efficient, with A. Ito favoured at 1.59.

Surface hard Odds A. Ito 1.59 · K. Sebov 2.30

Verdict: no bet

The model found no value at current prices — the market has this one priced about right. That verdict is part of the public record too.
Profiles: A. Ito · K. Sebov · head-to-head record · Toronto

Model probabilities

Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.

Method A. Ito K. Sebov
Elo
ratings-based, surface-weighted
 63.6% 36.4%
Machine learning
29-feature gradient-boosted model
 45.8% 54.2%
Serve / return
point-by-point simulation
 54.0% 46.0%
Blend
what the bot actually prices off
 63.6% 36.4%
Market implied
de-vigged from the opening odds
 59.1% 40.9%

We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?

Elo ratings

PlayerSurfaceEloRecordMatches
A. Ito hard1631 140–78218
K. Sebov hard1569 140–95235

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. 2, 2026.

Head to head

K. Sebov leads 1–0 over 1 meeting. Full record.

DateEventSurfaceWinnerScore
Aug. 2, 2026TorontohardK. Sebov2 - 0

Recent form

A. Ito

L K. Sebov 0 - 2 Aug. 2, 2026
W O. Selekhmeteva 1 - 2 Aug. 1, 2026
L Y. Ma 2 - 1 July 26, 2026
W A. Sharma 1 - 2 July 25, 2026
L G. Knutson 0 - 2 July 18, 2026

K. Sebov

W A. Ito 0 - 2 Aug. 2, 2026
L E. Kalieva 0 - 2 July 28, 2026
W M. A. Teixido Garcia 2 - 0 July 26, 2026
W C. Soares 2 - 0 July 25, 2026
L T. Maria 0 - 2 July 9, 2026

All Aug. 2, 2026 results & picks →

Full analysis

A. Ito (22, #86) vs K. Sebov (27, #225) — Toronto, hard.

MODELS vs MARKET
  ELO: A 63.6% | ML: A 45.8% (lone outlier) | model 63.6%
  Book A 1.59 / B 2.30 ; Betfair A 1.54 / B 2.42 (exchange SHORTENED A vs book — money is coming FOR the favourite).
  ELO + ranking (86 vs 225) + exchange all agree A ~63%. ML is the sole dissenter (break-even ~46%).

RECENT FORM (last 5 each)
  Ito: 08-01 Won Toronto Final vs Selekhmeteva (1701 Strong) 3-set GRIND; 07-26 Lost Memphis Final (Ma 1500); 07-25 Won Memphis SF (Sharma 1500, grind); 07-18 Lost Prague SF (Knutson 1425, clay); 07-14 Lost Athens (Korneeva 1500) 2-6 1-6. Volatile — title run yesterday but bad losses to weak fields. Shape STABLE/mixed.
  Sebov: 07-28 Lost Memphis (Kalieva 1500); 07-26 Won Memphis Final (Teixido 1500) 6-1 7-5; 07-25 Won Memphis SF (Soares 1500) 6-1 6-0 DOM; 07-09 Lost Newport (Maria 1803 Elite); 07-08 Won Newport (Brace 1417) DOM. Winning at ITF/quali level. Shape STABLE.

PATH & SHARPNESS
  Ito match-sharp (deep Toronto run, played 08-01). Sebov last match 07-28 (5d, within window).

FATIGUE / NARRATIVE
  Ito: 3-set title win YESTERDAY (326 court-min/7d, back_to_back). Extreme spot + LETDOWN (won Final 1d ago) + 3-set hangover −4.5pp. Ito 22 → recovers fast, haircut softened to ~−2/−3pp. Net A ~60-61%.
  Sebov more rested but no independent edge basis.
  ELO/ranking sanity: Ito high ELO AND high rank — agreement, ELO credible; ML is the discount candidate, not ELO.

PRICING (exchange truth)
  A: model ~63.6% (~60% post-letdown) vs Betfair 1.54 → breakeven 66.1%, NEGATIVE edge; min_odds 1.65 > 1.54 → auto-PASS.
  B: my honest B estimate (ELO+market+ranking consensus, letdown-adjusted) ~38-40% vs Betfair 2.42 breakeven 42.6% → NO edge. Backing B would require trusting the lone ML outlier over ELO+market+ranking triple-consensus, while exchange money is shortening A — classic manufactured contrarian edge.
  alt_markets: none at analysis time (handicap/totals null).
  AH: not quoted

GATES
  G1 edge-origin: FAIL — no numeric model-vs-market edge before narrative on either side (A negative on exchange; B edge exists only if the lone ML is believed over ELO+market+ranking). Letdown/fatigue cannot create the edge.
  G2 consensus:   stake-tier note — ELO+market+ranking align on A; ML lone dissent → contaminated/lone, cap tier only (moot, no bet).
  G3 fatal-risk:  n/a — ML market, G3 does not block; no bet regardless.

VERDICT: PASS. Favourite unbackable on exchange (price shorter than fair + below min-odds); underdog has no edge unless the noisy lone-ML is trusted against triple consensus with money flowing to A. Letdown is real but market has not drifted (it shortened A), so no priced edge to capture.

More from Toronto