Mike Hazen has a bullpen problem

GM Mike Hazen deserves a lot of credit for what he has done in his time with the Diamondbacks. Most obviously, he has rebuilt a farm system from one largely bereft of major-league talent, to one ranked by MLB.com as the third-best in the majors coming into this season - the first time in MLB Pipeline history it has been that high. Right now, the team would be in the 2023 post-season if it started today, and is making a good fist of competing in the NL West against the much better-resourced Padres and Dodgers. But it hasn’t all been roses. Free-agent signings have been much more of a mixed bag i.e. Madison Bumgarner, and there’s one area where Hazen has persistently struggled.

If you watched the Diamondbacks over the weekend, you will be painfully aware of what that is. On successive nights, our two “best” relievers, Andrew Chafin and Miguel Castro, helped blow win probabilities for Arizona which peaked at 99.6% and 95.3%. Going into Tuesday night’s game, the D-backs’ bullpen ranked in the following spots across the majors:

• ERA: 4.89 (25th)
• FIP: 4.91 (27th)
• fWAR: -0.3 (26th)
• HR/9: 1.49 (29th)

It’s a sensitive topic among fans, especially after the nightmare of last season, where Mark Melancon and Ian Kennedy combined to go 7-17 with an ERA of 4.99, while earning $10.5 million between them. This year, Chafin and Melancon will cost $11.5 million. Chafin has a 4.26 ERA, ranked 19th of the 22 pitchers with 5+ saves; Melancon hasn’t thrown a single pitch due to a shoulder injury, and nothing has been heard of him for more than a month.

Over the time Hazen has been here, various theories have been tried out. Let’s get relievers who have closing experience -the just not recently so we avoid the closer’s cost. This brought in the likes of Joakim Soria, Hector Rondon, Greg Holland and Brad Boxberger. Let’s actually bite the bullet and pay market value (most recently, Melancon and Chafin). Let’s stockpile power arms, throw them at the wall and see who sticks (this year that means Carlos Vargas, Luis Frias and Jose Ruiz, with the last-named the only one to have reached five MLB innings). The uniting factors among all these ideas is that I could see how they might make sense... and they were all failures.

For if it seems like forever since the D-backs had a good bullpen, it is. By fWAR, you have to go all the way back to 2019 to find the last time Arizona’s relievers were even replacement level, a modest 2.4 fWAR which ranked 17th. The only time in the Hazen era the bullpen has been above average was his first year, when they came 13th at 4.3 fWAR. And much of that 2017 was produced by players inherited from the previous regime, such as Archie Bradley and Chafin. Fernando Rodney was the sole successful example of the “previous closer” thing, and we’ve been chasing that dragon ever since. From 2017-23, six different pitchers recorded 10+ saves for Arizona. Bradley is the only one with an ERA for us below 4.20.

Over the entirety of Hazen’s tenure, the team ranks 29th for relief fWAR, their 4.1 fWAR being ahead only of the Marlins (2.8). And since the start of 2020, the picture is even grimmer: Arizona is dead-last at -2.8 fWAR, almost five wins below n the 29th-ranked Athletics. Including the shortened 2020 season, this period covers a total of about 420 games. It works out at about 4.3 wins over a normal 162-game season the team has been worse, simply compared to an average bullpen. The best bullpen over that time in the majors is the Dodgers, at 20.3 fWAR. And it’s not all about resources, because second place on the list goes to the Rays.

Below is a chart showing where the Diamondbacks have ranked in the majors, for each of the four key bullpen metrics, since Mike Hazen took over before the 2017 season.

If there is a sign of hope, it’s the way the yellow line has jumped up this season. That’s xFIP, or expected Fielding Independent ERA. Home-runs are generally regarded as a combination of fly-balls allowed and your home park, so xFIP “assumes” that a team will allow an average number of home-runs per outfield fly-ball. The D-backs bullpen hasn’t. Better than one in six fly-balls they’ve given up, haven’t come back, at 17.8%. That’s abnormal: from 2017-22, the figure was only 12.8%, so they’re 39% above that average rate this season. While those home-runs allowed aren’t coming back, that number should regress back towards normal going forward, and if it does, then the other metrics should improve with it.

That aside, is there anything else that can be done? There’s the $64,000 question. Bullpen arms are always going to be volatile. It’s the nature of the beast, where you’re typically talking about 50-odd innings over the course of a season. That’s the equivalent of a couple of months of work for a starter. Hell, Madison Bumgarner had a 3.31 ERA after his outing on June 1 last year, having thrown 54.1 innings. So any reliever can fluke their way into a good (or bad) season. All you can really do is pick pitchers who you think have the necessary tools to succeed.

High strikeouts are helpful, and Arizona currently are mid-level there, with just under a K per inning (126 in 127 IP, 19th in the majors). They are walking a few too many batters, with a rate of 3.83 per nine innings, =10th most. The D-backs’ bullpen is actually slightly better than average at generating ground-balls. As noted above, the main problem is that a horrendous number of the fly-balls are coming down in the bleachers. The Marlins are the only NL team to have given up a higher number of relief home-runs, and they’ve worked considerably more innings than the D-backs’ pen. Figuring out whether anything the bullpen is doing to lead to these homers, is probably the best thing pitching coach Brent Strom can do.