M. Dellavedova vs H. Jones — prediction

The 325-point Elo gap (1768 vs 1443) is the single largest driver of this match's projection. In ITF-level events, gaps of this size typically translate into clear dominance in service holds and break opportunities, even without granular serve/return data available here. This gap alone explains most of the 87% favorite probability.

Because Elo is soft-market calibrated at this level, the number should be read as directional confidence rather than precise metric — but a gap this wide rarely reflects a close match on paper.

Dellavedova has won both prior meetings with Jones, most recently in 2026. A clean 2-0 head-to-head at the same tier reinforces the Elo-based read: this is not just a ratings gap, it's a pattern that has held up on court before.

Two matches is a small sample, so this should support rather than override the level-based projection, but it adds directional confirmation.

The form trajectories are sharply opposed. Dellavedova's last 10 (WWWWWWLWLW) shows a mid-season losing blip but an overall dominant run with a current 1-match win streak. Jones's last 10 (LWLLWLWLLL) reflects a 3-match losing streak and only 3 wins in that span.

This divergence in current form aligns with — and amplifies — the underlying Elo and head-to-head signals, giving little reason to expect an upset based on recent trajectory.

Dellavedova has played 6 matches in the last 14 days versus Jones's single match in the same window, even though both have similar days since their last outing (6 vs 8). Heavier recent workload can be a fatigue factor, though best-of-three ITF matches are less taxing than five-set formats, so this concern should be weighted lightly.

Jones's relative inactivity could cut either way: fresher legs, or possible rust from limited match play — the data doesn't resolve which effect dominates.

At odds of 1.17, the market implies an 85% win probability for Dellavedova, essentially in line with the model's 87% estimate. The resulting 1.5% expected value is marginal and falls within normal noise for an Elo-based estimate on a soft ITF market.

This is a case where the favorite is heavily backed by both the model and market, but there is no meaningful mispricing to exploit. Treat the positive EV as an artifact of estimation, not a confirmed edge — being the favorite here does not equate to value.