The Rotation Is Broken. The Run Prevention Isn't. That Contradiction Is Arizona's Whole Season.

Everyone watching the Diamondbacks right now sees the same thing: a pitching staff held together with electrical tape, thirteen arms on the injured list, Zac Gallen posting a 6.40 ERA on the 60-day IL, Merrill Kelly bleeding runs at a 5.48 clip. The narrative writes itself — Arizona is a lineup propping up a rotation in structural collapse, and the only question is how long the offense can keep outrunning the damage. That narrative is not wrong. It is just incomplete.

Here is the number that complicates it: 3.4 runs allowed per game over the last ten games, against a prior 20-game baseline of 4.0. A 0.6-run improvement, in the middle of a rotation apocalypse, while going 7-3. I have spent most of this season documenting the ways Arizona's pitching has deteriorated. I wrote two weeks ago that the bill had finally come due. And yet the defense is somehow tighter now than it was when the staff was nominally healthier. That is the contradiction at the center of this team, and it deserves more than a footnote.

The explanation lives in the two arms who actually matter. Eduardo Rodriguez at 12-4 with a 2.71 ERA and Brandon Pfaadt at 7-1 with a 2.89 ERA are not just surviving — they are dominating, and their combined workload is carrying a disproportionate share of the innings that actually decide games. Pfaadt's start against Atlanta on August 15 was the clearest illustration: a performance good enough to build a 10-3 win around, the kind of outing that makes 3.4 runs allowed look sustainable rather than lucky. When those two pitch, Arizona has a genuine chance. The 0.6-run improvement is largely a story about what happens when Rodriguez and Pfaadt are in the rotation's driver's seat and the skeleton crew behind them is managed carefully enough to limit the bleeding.

Michael Soroka's return adds a third credible arm to that equation. His 8-4 record and 3.46 ERA since coming back from the IL give Arizona something it lacked for most of the summer — a genuine third option who can keep a game close. That is not nothing. In a race where the elimination number sits at 63 with 60 games left, "keeping games close" is functionally the entire strategy.

The fragility is real, though, and I am not going to pretend otherwise. Ketel Marte exiting against Atlanta with left knee soreness is the kind of news that can unravel a hot stretch in a single afternoon. The DFA of Paul Sewald — 24 saves, but a 6.23 ERA — cleaned up one structural problem in the bullpen while leaving the depth thinner. Geraldo Perdomo's knee issue earlier this week added another name to a roster already stretched to its limit. The 0.6-run improvement in run prevention is meaningful, but it is also fragile in a way that a healthy rotation would not be. It depends on Rodriguez and Pfaadt staying upright, on Soroka continuing to pitch the way he has since his return, and on the defense behind them not having a bad week at the wrong moment.

What I keep coming back to, though, is the direction of travel. This team was supposed to be in free fall by now. The rotation losses accumulated all summer — Burnes, Gallen, Nelson, Saalfrank, the whole catastrophic list — and the conventional read was that Arizona would eventually run out of ways to compensate. Instead, run prevention has improved in the last ten games. The offense is outpacing its season pace by more than a run per game. The record is 54-48, sitting exactly on the wild card line. None of that was supposed to be possible given the structural damage. The box score keeps disagreeing with the vibe, and at some point that disagreement is the story.