TennisEdge's AI model finds no betting value in L. Fernandez vs A. Eala (Washington on 2026-07-29) — the market is efficient, with A. Eala favoured at 1.62.
═══ L. Fernandez (A) vs A. Eala (B) — WTA Washington, hard, 1/8-finals ═══
MODELS vs MARKET
ELO : A 47.0% ML : A 50.1% Blend : A 47.0%
Book : A 2.30 / B 1.62 (context only)
BETFAIR (executable): A 2.24 / B 1.79 → normalised A 44.4% / B 55.6%
Commission-adj: A eff 2.178 (breakeven 45.9%) | B eff 1.751 (breakeven 57.1%)
Edge on A: blend +1.1%, ELO +1.1%, lone-ML best case +4.2% — ALL below the 5% bar.
Edge on B: my Eala read 53–56% < 57.1% breakeven → negative. No side prices.
Min odds for A at 47%: 2.234 eff (raw 2.30). Betfair 2.24 is BELOW the minimum.
Note: the brief's value_edge 5.7% on A is computed vs the soft BOOK quote 2.30.
The exchange contradicts it at 2.24 — that kills the bet, it does not confirm it.
RECENT FORM (last 5 each)
L. Fernandez (23, rank 22)
07-27 W Washington hard vs M. Linette (1775, Strong) 6-1 6-4 — DOM
06-29 L Wimbledon grass vs J. Tjen (1464, Weak) 1-6 6-7 — bad loss
06-23 L Bad Homburg grass vs Xin. Wang (1800) 3-6 4-6
06-21 W Bad Homburg grass vs K. Boulter (1765, Strong) 7-6 6-3 6-3 — GRIND
06-16 L Nottingham grass vs Z. Sonmez (1709) 4-6 6-7
A. Eala (21, rank 42)
07-28 W Washington hard vs Q. Zheng (1920, ELITE) 4-6 6-4 6-1 — GRIND, comeback
07-06 L Wimbledon grass vs J. Paolini (1817) 4-6 6-4 3-6
07-04 W Wimbledon grass vs I. Swiatek (1965, ELITE) 7-6 6-2
07-02 W Wimbledon grass vs M. Joint (1616) 6-3 6-2 6-0
06-30 W Wimbledon grass vs R. Zarazua (1564) 6-1 6-2
Deeper window: Eala also beat Rybakina (1838) 7-5 6-4 and Svitolina (1840) in Berlin,
and won the Birmingham title (06-07, Final over Bartunkova 7-5 6-3 5-7).
SHAPE / PATH (Step 1b)
Fernandez: MIXED / recovering. One genuine over-performance (DOM over a Strong opponent
07-27), but it sits on top of a poor grass block including a straight-sets loss to a
1464 (badly under-performing) and a losing clay stretch. Not a SHARPENING profile —
one good match after six weeks of drift.
Eala: SHARPENING, and the clearest read on the board. Wins over the Elite bracket
(Swiatek, Zheng) plus Rybakina/Svitolina and a title — repeatedly beating players
above her ranking bracket. Margins vs expectation are widening, not tightening.
Path: both 1 match at this venue, both match-sharp on the surface. No transition risk
on either side (both last played hard court here). Neither carries rust.
ELO / RANKING SANITY
Eala ranked 42 but ELO-favoured — the divergence is the RIGHT direction: her rating is
built on Swiatek/Rybakina/Zheng/Svitolina scalps, not farmed at weak events. Ranking
lags her level. That means ELO likely UNDERSTATES Eala, i.e. the models' 47% on
Fernandez is if anything generous to the side the brief wants to back.
FATIGUE (Step 2)
Fernandez: 1 match/14d, 64 min, last 07-27 → two days' rest. Fresh.
Eala: 1 match/14d, 102 min, last 07-28 → WON a 3-set match yesterday. The calibrated
3-set-hangover haircut (−4.5pp, surface-agnostic) applies. Age 21 (≤22, recovers fast)
→ soften one tier, call it ~−3pp. Applied once, not stacked.
Post-haircut Eala ≈ 50–53%, so the price/prob picture tightens but does NOT flip:
Fernandez still lands ~47–50% against a 45.9% breakeven — inside noise, not an edge.
Per Step 2, this is exactly the shape where the market over-prices a fatigue story; the
haircut is real but it is not large enough to manufacture 5%.
NARRATIVE (Step 3)
Let-down: none inside window for either (Eala's Birmingham Final was 06-07, 52d — no spot).
Venue momentum: neither has history at Washington in the window.
Title recency: Eala TITLE Birmingham 52d ago (grass) — stale, live form is the better read.
Surface transition: none. Both already played this hard court this week.
GATES
G1 edge-origin: FAIL — pre-narrative exchange edge on Fernandez is +1.1% (blend/ELO),
+4.2% on the lone ML read. Both under the 5% bar, and Betfair 2.24 is below the
2.30 raw minimum implied by min_odds = 1.05/0.47. G1 is the absolute ML blocker.
G2 consensus: n/a (stake-tier only, never blocks) — would have been capped tier
anyway: only ML backs Fernandez, ELO and market both side with Eala.
G3 fatal-risk: n/a for ML (never blocks; no margin market priced — handicap/totals null).
VERDICT — PASS
The value flag is a stale-book artifact: 2.30 at the book, 2.24 on the exchange, and the
edge evaporates against the price you can actually take. Beyond the pricing, the
qualitative read points AT the market's side, not against it: Eala is the SHARPENING
player beating the Elite bracket repeatedly, Fernandez has one good match glued onto a
poor two-month block. The one factor favouring Fernandez — Eala's 3-set win yesterday —
is a ~3pp haircut on a 21-year-old, not a thesis. Backing a side the market, ELO and the
form read all disagree with, at below the minimum price, on a lone-ML split, is the
definition of a manufactured edge. No stake.