B. Ellis vs D. Markovina — prediction

The 148-point Elo gap (1648 vs 1500) is the foundation of Ellis's favorite status, translating to a 70% model win probability. That gap reflects a meaningfully stronger rating history, and in an ITF-level match with limited other data, it's the single most reliable signal available.

Still, the market prices Ellis even higher, at 78% implied probability against odds of 1.28. That gap between the model's 70% and the market's 78% is the core tension of this matchup: the market is treating the rating edge as more decisive than the model does.

Recent form leans toward Markovina, who is on a 2-match win streak (WLWLLWLLWW) compared to Ellis's current skid, having lost his last match after a mixed WLLWLLWLWL stretch. Momentum alone doesn't override a 148-point Elo gap, but it tempers how comfortably Ellis should be seen as the favorite.

The lone head-to-head meeting, in 2026, also went to Markovina. With only one prior match between them, this carries limited statistical weight, but it does mean the opponent has already shown he can beat Ellis at this level.

Ellis's 61% serve-points-won rate is a real asset — a mark that, if sustained, should let him hold routinely and keep pressure off his own return games. No comparable serve number exists for Markovina, so a direct percentage comparison isn't possible, but Ellis's own serve profile is solid on its own terms.

The offsetting factor is his 35% return rate, which is below what's typically needed to consistently pressure an opponent's service games. That imbalance suggests Ellis's path to winning likely runs through holding serve rather than generating many break opportunities.

Rest cuts in different directions for each player. Ellis has played 3 matches in the last 14 days, more total volume, though he has had 2 days since his last outing. Markovina has played less overall (2 matches in 14 days) but is coming in on just 1 day of rest, the shorter turnaround of the two.

Neither scheduling picture is a clear red flag, but the tighter recovery window for Markovina is worth noting as a minor drag heading into this match.

The math here is straightforward: the model gives Ellis a 70% chance to win, while the market (via 1.28 odds) implies 78%. That gap produces a expected value of -10.3%, meaning that even though Ellis is very likely the better player by rating, backing him at this price is not a value bet by the model's own estimate.

This is worth stating plainly: being the favorite and offering value are two different things. On a soft ITF market where Elo-based edges are unproven in practice, the honest read is that Ellis is favored to win the match, but the current price does not offer a discount worth taking on the numbers presented.