TennisEdge's AI model finds no betting value in L. Klimovicova vs E. Yaneva (Warsaw on 2026-08-03) — the market is efficient, with L. Klimovicova favoured at 1.91.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | L. Klimovicova | E. Yaneva |
|---|---|---|
| Elo ratings-based, surface-weighted |
47.3% | 52.7% |
| Machine learning 29-feature gradient-boosted model |
52.8% | 47.2% |
| Serve / return point-by-point simulation |
54.1% | 45.9% |
| Blend what the bot actually prices off |
47.3% | 52.7% |
| Market implied de-vigged from the opening odds |
50.0% | 50.0% |
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. Klimovicova | hard | 1647 | 130–59 | 189 |
| E. Yaneva | hard | 1703 | 38–15 | 53 |
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. 4, 2026.
E. Yaneva leads 1–0 over 1 meeting. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 4, 2026 | Warsaw | hard | E. Yaneva | 2 - 1 |
| L | E. Yaneva | 1 - 2 | Aug. 4, 2026 |
| L | R. Saigo | 0 - 2 | Aug. 1, 2026 |
| W | V. Chaudhari | 2 - 0 | July 31, 2026 |
| W | C. Vlasselaer | 2 - 0 | July 30, 2026 |
| W | N. Sato | 2 - 0 | July 28, 2026 |
| W | L. Klimovicova | 1 - 2 | Aug. 4, 2026 |
| L | F. Urgesi | 2 - 1 | July 20, 2026 |
| L | M. Trevisan | 1 - 2 | July 17, 2026 |
| W | S. Lansere | 2 - 0 | July 15, 2026 |
| W | O. Selekhmeteva | 0 - 2 | July 13, 2026 |
All Aug. 3, 2026 results & picks →
Klimovicova (A) vs Yaneva (B) — Warsaw, hard.
MODELS vs MARKET
ELO A 47.3% | ML A 52.8% | sim A 47.3% — models cluster ~50/50.
Betfair ML: A 1.66 (~55% devig), B 2.0 (~50% raw). Market makes A a
clear favourite, MORE bullish on A than ANY model. Book quote 1.91/1.91
is stale placeholder; exchange is truth.
RECENT FORM
A (Klimovicova, 22, #138): playing W35 Aldershot ITF on HARD this week —
W 6-0 6-3 (1466), W 6-1 6-3 (1383), W 6-3 7-6 (1388), then L 3-6 6-7 SF
vs Saigo (1479) yesterday 08-01. DOM over very weak fields (ELO 1383-1479),
then lost the SF. Match-sharp on the surface but the level faced is bottom-
tier ITF; shape STABLE-to-FADING (lost yesterday to a 1479).
B (Yaneva, 19, #224): higher-level path — WTA Rome/Palermo, beat Lansere
6-2 7-6, lost Trevisan/Urgesi; last match 07-20 Palermo (clay). Stronger
opposition but all on CLAY/grass. 13 days off = rust risk, and clay->hard
TRANSITION (looser timing, depressed level in match one).
FATIGUE (Step 2 calibration)
A: 4 matches/7d, 283min, back_to_back 2, last match yesterday — but it was
a 2-SET LOSS (no 3-set-hangover haircut, that applies only to yesterday's
3-set WINNERS). A is 22 -> recovers fast, soften. Cumulative load is
CONTEXT not an edge; not an extreme spot. Roughly neutral.
B: 13 days idle, well-rested but rusty.
ELO vs RANKING
A #138 yet ELO makes her the underdog (47%) — ELO does NOT overrate her
despite ranking. B #224 yet ELO gives her 53% (built vs strong clay
fields). ELO credible; no farmed-rating red flag driving the pick.
NARRATIVE
No Final in window (letdown None). Surface transition on B (clay->hard) —
a negative for B, mild support for A. No venue/title signals.
GATES
G1 edge-origin: FAIL — no >=5% exchange edge on either side. Back B @2.0
eff 1.95 (impl 51.3%): best model B 52.7% -> edge ~1.4%, ML B 47.2% ->
negative. Back A @1.66 eff 1.627 (impl 61.5%): all models <=52.8% ->
negative. Numeric edge absent before any narrative.
G2 consensus: n/a (stake tier) — no bet.
G3 fatal-risk: n/a — ML market, no bet.
MARKET CHOICE
alt_markets: none at analysis time (handicap/totals null; only betfair_ml).
AH: not quoted.
No priced margin market; ML only. Neither ML side clears its min-odds bar.
VERDICT
Model-vs-market DO disagree directionally (models ~50/50, market prices A
at 60%), but the only side that disagreement points to (B) does not clear
the 5% exchange-edge bar (~1.4% best case, negative on ML), and the
qualitative read (B rusty 13d + clay->hard transition; A match-sharp on
hard) supports the market's A-lean, not a B bet. No executable value
either way. PASS.