TennisEdge's AI model finds no betting value in A. Magadan vs G. Johns (Los Cabos on 2026-07-26) — the market is efficient, with A. Magadan favoured at 1.85.
MATCH: A. Magadan (rank 795, age 25) vs G. Johns (rank 361, age 25) - Los Cabos, hard MODELS vs MARKET - Book: A 1.85 / B 1.92 (implied A 50.9 pct) - Betfair exchange: A 1.02 / B 1.73 - A side effectively empty (1.02 is a broken/illiquid quote); executable price for Johns is 1.73 (1.69 after 5 pct commission) - ELO: Johns 57.4 pct | ML model: Magadan 52.0 pct (agrees with book market) | value_pick B +8.4 pct is computed vs the BOOK quote - ELO vs ranking: Johns 361 vs Magadan 795 - consistent with ELO lean, no farmed-rating flag. But ML sides with the market against ELO: lone-model split. RECENT FORM (last 5 shown; all fetched matches used) Magadan (5 matches in 14d, ITF/Challenger level, many default-1500 opponent ELOs): - 07-22 Bloomfield Hills L vs M. Zheng (1781) - 07-21 Bloomfield Hills W vs S. Kozlov (1545) - GRIND (tiebreak) - 07-20 Bloomfield Hills Final W vs A. Mayo (1533) - GRIND (two tiebreaks) - 07-19 Bloomfield Hills W vs O. DeMuth (1350) - DOM vs Weak - 07-12 Granby L vs B. T. George (1164) Johns (4 matches in 14d): - 07-21 Winnipeg L vs A. Holmgren (1643) - 3-set GRIND loss - 07-20 Winnipeg Final W vs E. Aguiard (1508) - GRIND - 07-19 Winnipeg W vs B. T. George (1195) - routine 3 sets - 07-14 Granby L vs D. Glinka (1618) - 07-09 Newport (grass) L vs Dar. Blanch (1639) Both grind out wins vs Weak/Avg opposition and lose to Strong opponents. No class edge visible either way. FATIGUE (precomputed) - Magadan: 5 matches/14d, 595 court min, 3 back-to-backs, last match 07-22 - Johns: 4 matches/14d, 556 court min, 2 back-to-backs, last match 07-21 - Diff 39 min / 1 match: slight edge Johns at most; both age 25 (baseline recovery). Not material. NARRATIVE - Let-down: BOTH won an ITF final 5d ago (Magadan Bloomfield Hills 07-20, Johns Winnipeg 07-20) - WEAK letdown on both sides, cancels out. - Surface transition: none - both coming off hard courts onto hard. - Venue: no Los Cabos history for either. PRICING DISCIPLINE The book 1.92 on Johns is the source of the +8.4 pct value flag, but the exchange prices Johns at 1.73 (eff. 1.69). Edge at the executable price even using ELO 57.4 pct: 0.574 - 1/1.69 = -1.7 pct. Negative. The exchange contradicting a soft book quote kills the bet - stale quote, not value. Backing Magadan is impossible: the A side of the exchange shows 1.02 (dead market). GATES G1 edge-origin: FAIL - vs the executable Betfair price (1.73, eff 1.69) the numerical edge for Johns is negative before any qualitative adjustment; the apparent edge exists only vs a stale book quote G2 consensus: FAIL - ML model and exchange market both side against the ELO pick; lone ELO split with no hard independent evidence (no H2H, no live-event class signal) G3 fatal-risk: PASS - no thesis-negating risk, but moot given G1/G2 VERDICT: PASS. Lone ELO signal against ML + exchange consensus, and the value evaporates at the executable price. Data quality also weak (many default-1500 opponent ELOs, scrambled round ordering at ITF level).