TENNISEDGE.GG

U. Humbert vs K. Khachanov — Prediction & Analysis

Hamburg · ATP Hamburg - 1/8-finals · May 20, 2026 · 10:00 UTC

TennisEdge's AI model finds no betting value in U. Humbert vs K. Khachanov (Hamburg on 2026-05-20) — the market is efficient, with K. Khachanov favoured at 1.50.

Surface clay Odds U. Humbert 2.60 · K. Khachanov 1.50

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: U. Humbert · K. Khachanov · head-to-head record · Hamburg

Model probabilities

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

Method U. Humbert K. Khachanov
Elo
ratings-based, surface-weighted
28.6% 71.4%
Machine learning
29-feature gradient-boosted model
22.8% 77.2%
Serve / return
point-by-point simulation
49.5% 50.5%
Blend
what the bot actually prices off
33.6% 66.4%
Market implied
de-vigged from the opening odds
36.6% 63.4%

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

Elo ratings

PlayerSurfaceEloRecordMatches
U. Humbert hard1831 131–80211
K. Khachanov hard1781 120–84204

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 July 30, 2026.

Head to head

U. Humbert leads 2–1 over 3 meetings. Full record.

DateEventSurfaceWinnerScore
May 20, 2026HamburgclayU. Humbert2 - 1
Nov. 2, 2024ATP ParishardU. Humbert2 - 1
July 29, 2021Olympic GameshardK. Khachanov2 - 1

Recent form

U. Humbert

L B. Shelton 1 - 2 July 30, 2026
W A. Martin 2 - 0 July 28, 2026
L Z. Bergs 2 - 3 June 30, 2026
L Z. Bergs 2 - 1 June 28, 2026
W J. Draper 0 - 2 June 26, 2026

K. Khachanov

L B. Tomic 2 - 0 July 30, 2026
L F. Cobolli 2 - 3 July 4, 2026
W Y. Hanfmann 3 - 0 July 2, 2026
W B. Harris 3 - 1 June 30, 2026
L M. Damm 0 - 2 June 24, 2026

All May 20, 2026 results & picks →

Full analysis

Math-only prediction: no LLM calls used