Aberdeen and Kilmarnock played out a goalless draw at Pittodrie, leaving both sides with a point apiece. It was a tight contest that failed to produce the goals either team will have fancied themselves to score.

Our model came into the match expecting a more open affair. We'd projected a 2-1 Aberdeen win, backed by a 51% win probability for the home side and only a 23% chance of a draw. The underlying form suggested Aberdeen had the edge—their expected goals came in at 1.77 compared to Kilmarnock's 1.12—but that advantage didn't translate on the day. A goalless stalemate was always in play at 23% probability, though it wasn't the likeliest outcome we'd called. The match simply didn't follow the expected pattern: chances that looked set to fall either didn't arrive or went begging.

Both sides will probably feel they could've won it. Aberdeen dominated the xG stakes before kickoff and carried that expectation into the game, but Kilmarnock proved stubborn and neither keeper was seriously beaten. It's one of those afternoons where the tactical setup or individual moments inside the box made all the difference. Our model leaned toward an Aberdeen win on the back of their form advantage, but the reality was tighter and more cautious than anticipated. A draw wasn't a wild miss—it was always a reasonable outcome—but it does show why single-match predictions, even when they're built on solid underlying metrics, can't account for the fine margins that decide tight games.