← Back to matches
hardoutdoor· China· Hangzhou Open

Taro Daniel vs Dane Sweeny

24 Sept, 05:00 · Round 1 · best of 3 · in 2h 2m

45%
55%

Why this is close

Each bar is that input's share of the explanatory signal.

Surface Elo (hard)17.6% · +1.4% toward Taro Daniel
ATP ranking48.0% · -3.8% toward Dane Sweeny
Rest and workload34.4% · -2.7% toward Dane Sweeny

Raw model probability 45%, calibrated to 45% for Taro Daniel. Calibration is fitted on completed matches and pulls confident predictions back toward the centre.

Projected total games

24.0

leans over for a best of 3

Straight sets

50%

chance the match ends without a deciding set

Market comparison

Our model
55%
Market
43%

2.22

Difference
11.9%

Prices exist, but one player is unrated on this surface, so we are not drawing a comparison.

Betano MX · 1.67 / 2.22 · bookmaker margin of 4.9% removed so the two sides sum to 100%. Predictions only — not a profit claim.

Underdog keeps it close

Dane Sweeny is the market underdog, and our model rates them 11.9% higher than the de-vigged price. On top of that, the margin model gives them a 78% chance of finishing within 4.5 games (or winning outright).

Margin model measured on 2,080 settled matches. We carry no handicap prices yet, so this is a statistical read on how close the match should be — not a priced edge.

What is missing here

  • Taro Daniel is unrated on hard: fewer than 5 matches in our history, so his rating here falls back to his overall level. This also blocks any market comparison for this match.
  • Taro Daniel's ATP ranking could not be confirmed against this week's list, so it is used with less confidence.
  • Dane Sweeny has a provisional hard rating (5–19 matches), so it is pulled part-way toward his overall rating rather than used at face value.
Taro DanielDane Sweeny
1627hard Elo1627
1665Overall Elo1648
-37hard vs overall-21
2–1hard record2–3
unratedrating basisprovisional

The model uses an effective rating that pulls a thin surface sample back toward the overall rating: 1659 vs 1643.