To be clear, this is something of a devil’s advocate piece. I think it’s quite likely the team will bring Merrill Kelly back, and it’s not a scenario to which I would completely, or even particularly, be opposed. The team desperately needs starting pitchers. Right now, the rotation has one man with an ERA+ over 86 last year (Ryne Nelson), and a pair who were mediocre (Brandon Pfaadt and Eduardo Rodriguez). There is literally nobody else healthy in the organization who made a start for Arizona last year. Zac Gallen, Kelly, Anthony DiSclafani, Nabil Crismatt and Jalen Beeks (!) are free agents, while Corbin Burnes will be on the IL until well into 2026.
Kelly was dealt to the Rangers at the trade deadline last year, but even before that happened, Merrill expressed a willingness to come back to the D-backs. “The front office knows how fond I am of being here and being home. I know my wife is fond of being here and being home. Regardless of what happens after the 31st, being here is never off the table.” That loyalty to the organization is understandable. After all, it was the Diamondbacks who plucked Kelly from the relative wilderness of Korean baseball in December 2018 and brought him back to the majors, culminating in a World Series start for Merrill in 2023.
It has been worked for both sides. Kelly has earned $38.5 million through the end of this season. But he has been worth 16.6 bWAR, and ranks top five there in franchise history, as well as starts, innings pitched and wins. In April 2022, he signed an extension with Arizona, which ended up being a three-year deal for $24 million. There have been hopes Kelly might give the team a home-town discount. But earlier this month Nick Piecoro wrote there are indications Kelly “is looking for a deal more commensurate with his market value.” If that’s true, then the risks become higher – and that’s what I am here to discuss. Let’s begin with the most obvious issue.
Merrill Kelly is already 37. He may not seem that old, because he he did not make his MLB debut until 2019, when he was in his thirties: Ketel Marte was in his fifth major-league season that same year. But the reality is, few pitchers of that age continue to be successful as starting pitchers. Only one man in their age 37+ season this year, put up even two bWAR: Jacob deGrom. That’s not an aberration: there were none in 2024. Interestingly, relievers seem to age better. In each of the last two seasons, the most productive old pitcher worked out of the bullpen, in Kirby Yates and Aroldis Chapman respectively.
But for rotation members, the aging curve at this point is steep. The chart below plots age against number of 1+ win seasons by starting pitchers, for the five post-pandemic years, with a total sample size of 530 player-seasons.
Over the last five seasons, only eighteen starting pitchers aged 37 or older, have posted even 1.0 bWAR or better. It’s partially a matter of simple wear and tear. Of those eighteen, only a handful of men made more than 30 starts, indicating they enjoyed an entirely healthy campaign. The problem is if the “market value” for which Kelly is looking, is determined for him by past performance, not his likely future production. Because the latter is likely to be significantly less. He’s also going to be looking for a multi-year contract, and the problem gets compounded the further we go into his future. However, I sincerely doubt any team will offer him more than two years.
Jeff Zimmerman looked at pitcher aging curves in 2020, and almost all components reflect a decline: strikeout, walk and home-run rates all clearly get worse (for the pitcher) as they age. We are seeing Kelly’s average fastball velocity go down, one of the factors involved in the overall decline for results. After peaking at 92.6 mph in 2022, Merrill’s velo has declined each of the past three seasons, to a figure of 91.8 mph this year. It’s interesting to note that’s about the speed he had when he first returned from Korea, and Kelly initially struggled. For example, in 2021, his fastball averaged 91.7 mph, and he had an ERA of 4.44 across 158 innings.
The above is Kelly’s Baseball Savant page for this year, and there’s rather too much blue on it for my tastes. The two areas where he has been significantly better than average are chase percentage (the rate at which batters swing at pitches outside the strike-zone) and walk rate. On the other hand, xERA in particular has a strong dislike for Merrill. This is a number basically based off exit velocity and launch angle of the batted balls a pitcher allows, and is generally close to ERA. Kelly has beaten it of late, with his xERA figure for each of the past three seasons significantly higher than his actual ERA, by between 0.64 and 0.88 runs.
Kelly’s batting average on balls in play (BABIP) has also been consistently low. Over the past four years, it has ranged between .264 and .279 – MLB average is .293. Among the 97 pitchers to have thrown at least four hundred innings in that time, Kelly’s overall BABIP of .270 is the eleventh lowest. That isn’t the result of him allowing fewer line drives – his LD% is 43rd among the same sample – or hard-hit balls, factors which can play into a reduced BABIP. Indeed, by the latter, he has actually been worse than average over that time, Kelly ranking 68th of 97 by that metric. This suggests he “should” be allowing more hits and runs than he has.
Are these sustainable? That’s likely going to determine the success or otherwise of Kelly’s next contract.
ESPN’s Kelly McDaniel projected a two-year, $35m contract for Kelly, writing the player “doesn’t have much margin for error to keep his ERA in the 3s, but he has the feel to eat up innings regardless of his raw stuff.” Over on MLB Trade Rumors, they were not too far away: two years and $36m. They noted “Kelly could help just about any postseason hopeful looking to solidify the middle of its rotation,” a broad market which would increase the player’s value. Those kind of contracts would likely be too much for Arizona: as Jack wrote, “While there might be a team out there that is willing to offer Kelly a two-year, $30 million contract, I doubt that team would be the Diamondbacks.”
In the same piece, he suggested a one-year, $12 million deal, with a team option at the same price for a second season, plus a $3m buyout. With just the one guaranteed year, and a significantly lower value than the projections above, it would require Merrill to take a significant hometown discount to accept such an offer. Having arguably already done so with the previous contract extension, you wonder how willing Kelly will be to do so again. How much is his wife “fond of being here”? A million here or there might be okay: but the five or six million suggested, would go a long way to buying out any disruption.