T. Berkieta vs D. Kellovsky — prediction

Elo ratings place Berkieta narrowly ahead, 1604 vs 1582, translating into a 53%-47% model split that barely favors the favorite rather than confirming clear superiority.

Recent form echoes this modest gap: Berkieta is 5-5 in his last 10 matches with a short one-match losing streak, while Kellovsky sits at 4-6 and is riding a three-match losing streak. Neither player's underlying level looks dominant, but the trend tilts slightly toward Berkieta.

Rest is the sharpest contrast in this matchup. Berkieta arrives on just 3 days off after playing 6 matches in the last two weeks, a workload that can wear down legs and focus late in matches.

Kellovsky, by contrast, has had 11 days to recover and only 1 match in the same span. This scheduling gap favors the opponent and could offset Berkieta's small Elo advantage if the match runs long.

Berkieta backs his ranking with concrete serve numbers: 65% of service points won and 45% on return, a balanced profile that suggests reliable holds and some pressure on Kellovsky's service games.

No comparable serve or return data exists for Kellovsky, so this should be read as a standalone positive for Berkieta rather than a direct comparison.

The two have met once, in 2025, with Kellovsky winning. A single match is a thin sample and shouldn't be weighted heavily, but it does remove any psychological edge Berkieta might otherwise claim over this opponent.

The market prices Berkieta at 1.27, implying a 79% win probability, while the Elo-based model gives him only 53%. That gap is large, and the resulting expected value of -32.5% signals the price is not supported by the model's read of the match.

Even allowing that Challenger/ITF Elo is a softer, less-tested method, a roughly 26-point gap between market and model is too wide to dismiss as noise. This is not a price worth backing; the model sees a close contest, not the near-lock the odds imply.