Colorado Rockies at Arizona Diamondbacks: Prediction, Odds & Preview
DiamondIQ Model — Win Probability
The model leans AZ (52.5%). DiamondIQ model v2: season records, home-field advantage, the starting-pitcher quality gap (PitchIQ), and a calibration adjustment fit and validated on four seasons of backtest — plus live game state once underway.
The Matchup
This is an early look at a May 22 matchup at Chase Field between the Colorado Rockies and the Arizona Diamondbacks, with probable starters not yet announced. The DiamondIQ model's estimate gives Arizona a 52.5% win probability against Colorado's 47.5%, a modest lean rooted in home-field advantage and the model's starting-pitcher quality gap component, even with both rotations still to be determined. Chase Field carries a DiamondIQ park factor of 1.03, meaning the three-season run environment sits three percent above league average, and with a forecast of 107 degrees at first pitch and an 11 mph west-to-east wind blowing left to right, the conditions figure to play up offense further once the game is actually played. That combination of a hitter-friendly park and extreme heat historically suppresses swing-and-miss and elevates contact, making run prevention a premium.
Without confirmed starters, the pitching picture will sharpen considerably as the week progresses, and it is worth noting that both clubs are carrying meaningful rotation depth concerns on the injured list. Colorado has five pitchers sidelined, including Chase Dollander on the 60-day IL, while Arizona is without Drey Jameson and Michael Soroka. How each club fills its rotation slot for this game could meaningfully shift the model's lean, since the PitchIQ component currently driving Arizona's edge is based on aggregate starter quality and will update once names are confirmed.
One thing to watch as this game approaches is the bullpen disparity. Arizona's relief corps carries a BullpenIQ of 55 out of 100 with six fresh arms and closer Paul Sewald available, while Colorado's bullpen checks in at 44 with six fresh arms and closer Jordan Romano. The DiamondIQ model does not currently factor in bullpen performance, so that gap represents a layer of context the raw win-probability figure does not fully capture, particularly in a high-temperature environment where starters may be managed carefully.