WIMBLEDON Final and SF prediction evaluation

PREDICTION: Sinner to beat Zverev, and Sinner to win one set 6–3,

and the match ended 7–6, 6–7, 6–3, 6–4 in Sinner's favor, then I'd say it was a strong prediction, especially the set score.

Here's why:

Picking Sinner was a reasonable forecast. He came into the final as world No. 1, defending champion, and had dominated the recent head-to-head with Zverev. So while not trivial, it was the outcome many rating systems would have favored. Predicting an exact 6–3 set is much more specific. Even when a player is favored, there are many plausible winning set scores: 6–0, 6–1, 6–2, 6–3, 6–4, 7–5, and 7–6. Hitting one of those exactly is a much lower-probability event than simply predicting the match winner. The actual match also featured two tiebreak sets, making the correctly predicted 6–3 third set stand out even more.

So I would rate the prediction like this:

Winner: ★★★★☆ (good, but in line with expectations for a favorite).

Correctly forecasting a 6–3 set: ★★★★★ (impressive level of detail).

Overall (SF and FINAL)

Taken together, I'd rate the three-match prediction set at about 9.5/10.

The reasons are:

✅ Correctly identified the winner in all three matches.

✅ Correctly identified a straight-sets victory.

✅ Correctly predicted two exact set scores (7–6 and 6–3), which are much harder than simply picking the winner.

The only reason I wouldn't call it a perfect 10/10 is that a model's quality is judged over many predictions, not just three. Even outstanding models can have a perfect weekend, and average models can occasionally get lucky.

If TenELOs produces this level of accuracy consistently over an entire Grand Slam or season, then it would deserve to be considered a genuinely elite tennis forecasting model.