The MLB trade deadline is two months away, a benchmark with more finality this season because, for the first time, there will be only one trade deadline. No more August waiver deals. As usual, everyone needs relief pitching. Everyone. Contenders. Pretenders. Suspenders. Maybe not that last one.
While trades involving relief pitchers, especially non-closers, generally don't generate headlines in 72-point font, there are always plenty of them during a typical season. As talent flows from win-later to win-now teams, relievers are the most common currency. A good bullpen on a bad team is a senseless luxury when those productive arms can be converted into prospect depth. Bullpens, as the thinking so often goes, are the final pieces in the title-pursuing puzzle.
I am very curious to see how this scenario plays out this summer. For over a month, I've been beating the drum over what I consider to be a shortage of quality relief pitching.
One reason for that is that there is simply more asked of relievers writ large in terms of innings. Relievers as a group aren't performing as well as starters this season by ERA, something that would have been considered borderline unthinkable even a few years ago. Part of this is a consequence of limiting the exposure of starters. Innings a tiring starter might have once tried, and failed, to fight through are now being handled by fresh-armed relievers.
However, a fresh arm does not necessarily equate to an effective arm, and even if raw velocity is more prevalent than ever, the sheer number of relievers needed to cover the innings demanded by new pitching-staff models has created a supply issue. And while all of that velocity might keep strikeout rates high, it also means it's harder to keep enough arms in working order. Thus injured lists all over baseball are overstuffed with ailing relievers.
For sellers at this year's deadline, that creates opportunity. Let's use Alex Colome of the Chicago White Sox as an example. Colome has been the most effective reliever for Chicago this season, posting a 1.59 ERA, an elite .190 wOBA allowed, albeit with a non-elite 8.3 strikeouts per nine innings, and 11 saves. He's still under team control next year, his final season of arbitration eligibility, and is making a team-friendly-ish $7.325 million this season.
Given the bullpen needs of contenders from Boston to Los Angeles to Milwaukee to Philadelphia to Atlanta to the other end of Chicago, the White Sox could reasonably expect to leverage interested parties against one another to yield a solid prospect or two.
However, the White Sox have factors to consider in addition to simply determining the best offer, because the best option might be to take no offer at all. If they believe Colome can be a dependable high-leverage reliever next season, when the White Sox are more likely to move into serious contention, the current reliever marketplace might look barren enough that hanging on to Colome is the way to go. This evaluation of the reliever market between now and a rebuilding team's next push for contention takes on heightened importance in the current landscape.
All of these emerging forces could mean that the movement of relievers, and the prospects they yield, will be one of the most interesting and revealing aspects of this season's trade deadline. Here's another factor to consider: Trading for relief pitchers, even those who have been throwing well, is far from a sure bet to boost a team's pennant chances.
Using FanGraphs.com, I put together a database of every pitcher who pitched at least 20 innings for two different teams in a season, going back to 2000 to keep it contemporary. There were 193 such pitchers. Sixty of them -- nearly a third -- produced below replacement level after being dealt.
Part of this might stem from the fact that a lot of teams have sold high on relievers, whose career patterns often look like Pinocchio's polygraph results. For some, that is surely the case. Seventy-seven relievers of my group of 193 pitched worse (in terms of fWAR per 60 innings) after being traded, producing an aggregate average of 0.2 fWAR. That means 116 pitchers -- 60 percent -- actually pitched better, and that group averaged 0.45 fWAR after being dealt.
While there is nothing in the data to suggest that you should avoid outside acquisitions of relievers altogether, the success rate of those additions does make it a bit of a crapshoot. It's worth looking at some of the characteristics of the relievers who helped their new teams, and of the ones who have not.
First, there is clearly a good bit of regression in play. The post-trade helpers managed strikeout rates of 110 percent of league average for their original teams, and 118 percent after. The non-helpers went from 114 percent to 101 percent. Same story with walks: Helpers rose from league average to 126 percent; non-helpers fell from 115 percent to 103.
It's like that with home run rates, too, but the effect is a lot more pronounced. Helpers went from 90 percent of league average to 145 percent; non-helpers fell from 133 percent to 87. So teams that can look past outlier home run rates, or rates of homers per fly balls, will go some way in sorting wheat from chaff.
But here's another thing in the data that stands out:
This looks at the recent track records of the pitchers. One-year WAR is simply the fWAR they produced as a reliever the previous season. Two-year WAR is the previous two seasons, and so on. The results are counterintuitive. The relievers who produced more in previous seasons were the ones who helped the least after being traded.
The way I read this is that when it comes to relief pitchers, track record is not entirely helpful. More valuable is recent performance and, presumably, a fresher arm. The track-record guys tend to be the ones with the most recognizable names, but in the bullpen market, brand recognition might not be the way to go.
Obviously there are no magic bullets when it comes to targeting bullpen help in the trade market. Teams will look at a much deeper suite of analytics than this and pair those observations with what they get from their scouts. The tendencies noted here are more about group dynamics than anything else, but we can still use them to narrow down some of this season's possible bullpen trade targets.
Without getting too deep into the methodology, here's a quick primer: I identified 11 teams as likely sellers. Then I created a rating for those teams' relievers based on likely regression, pre-2019 track record (with recent past performance weighted heavily) and 2019 fWAR. I mashed those ratings into one using z-scores, then ranked the relievers for their desirability as a trade target. I did not worry about contract or current injury status.
In list form, here are the results, which I didn't generate until I had already written my Colome example above, so I feel pretty good about choosing him:
TOP BULLPEN TRADE TARGETS
1. Alex Colome, White Sox
2. Ty Buttrey, Angels
3. Greg Holland, Diamondbacks
4. Kirby Yates, Padres
5. Ken Giles, Blue Jays
5. Shane Greene, Tigers
7. Adrian Sampson, Rangers
8. Cam Bedrosian, Angels
9. Jake Diekman, Royals
10. Aaron Bummer, White Sox
11. Daniel Stumpf, Tigers
11. Shawn Kelley, Rangers
11. Tony Watson, Giants
14. Reyes Moronta, Giants
15. Sam Dyson, Giants
16. Will Smith, Giants
17. Roenis Elias, Mariners
18. Mark Melancon, Giants
19. Juan Minaya, White Sox
20. Ariel Jurado, Rangers
BULLPEN TRADE TARGETS TO AVOID
1. Zac Reininger, Tigers
2. Wei-Yin Chen, Marlins
3. Miguel Castro, Orioles
4. Adam Warren, Padres
5. Drew Steckenrider, Marlins
6. Brad Wieck, Padres
7. Kyle Dowdy, Rangers
8. Josh Osich, White Sox
9. Matt Wisler, Padres
10. Noe Ramirez, Angels
11. Victor Alcantara, Tigers
12. Branden Kline, Orioles
13. Elvis Luciano, Blue Jays
14. Derek Holland, Giants
15. Ryan Tepera, Blue Jays
16. Matt Andriese, Diamondbacks
17. Luis Perdomo, Padres
18. Chris Martin, Rangers
19. Brad Boxberger, Royals
20. Jeanmar Gomez, Rangers
Extra innings
1. The Rays' Austin Meadows had another big night against the red-hot Twins on Thursday. Even before that game, Meadows had moved into 10th in the American League by fWAR, with 2.1. That's still a way from Mike Trout's league-leading 3.3, but Meadows is drawing closer to the group behind Trout even though he's played in just 38 games.
Meadows ranks third in the AL in FanGraphs' version of win probability added, so his contributions have mattered. And since he plays on the contending Rays, Meadows leads all AL players in championship win probability added, per TheBaseballGauge.com.
Meanwhile, Tampa Bay starter Tyler Glasnow is on the 60-day injured list and probably won't return until after the All-Star Game. However, he currently ranks seventh in AL fWAR among pitchers, 12th in win probability added and sixth in championship probability added. And Glasnow hasn't thrown an inning since May 10.
You can see where this is headed, right? If Glasnow had not gotten hurt, the Rays could have had leading contenders for both AL MVP and Cy Young in two of the three players they got last season from the Pittsburgh Pirates in exchange for Chris Archer.
Archer, by the way, has pitched at exactly replacement level this season. And the other player the Rays acquired in that trade -- pitching prospect Shane Baz -- had a 1.80 ERA and 25-5 strikeout-to-walk ratio over four early outings in Class-A.
I've said it before and I'll say it again: Do not trade with the Rays.
2. Talk about extremes: As I, and quite a few others by now, had noticed, this year's Astros continue to sport one of the top offenses of all time, per the FanGraphs metric wRC+. Through Wednesday, Houston's mark (125) had slipped behind the 1927 Yankees (126) into second place on the historical list. That number could start to slide given the injuries the Astros suddenly are dealing with in relation to their potent lineup.
Just because Houston's hitters are humming at a historic level doesn't mean that there should be another offense doing the same but at the opposite end of the spectrum. I mean, this is not a zero-sum kind of category. But there is such a team: the Miami Marlins.
The Marlins have been playing better of late, but nevertheless, through Wednesday's action, Miami had a 71 wRC+ for the season. As you've probably guessed, it's one of the worst figures of all time. Only four teams have done worse for a full season: the 1920 Philadelphia Athletics (67), 1963 New York Mets (69), 1909 Boston Doves (70) and 1952 Pittsburgh Pirates (70).
3. Last month, I paid my first visit to Coors Field, doing so in a non-working capacity. (Which basically means that I get to drink beer during the game. But I still keep score.) I liked the park very much, though I won't do an in-depth write-up on it here. I love that it's located in and is so well integrated with downtown Denver. In fact, if you want to see a tremendous example of a vibrant urban core that's been refashioned over the years, go to Denver.
About a half-block from Coors Field, right there on Blake Street, is the National Ballpark Museum. This is what I have really been meaning to mention. The name is pretty grandiose, but at least it gives the place room to grow. For now, it's a modest storefront museum, with an impressive array of baseball arcanum.
For a ballpark aficionado -- which I claim to be -- there are some interesting experiences to be had. Here's the litmus test: Would it be a gas for you to walk through an actual turnstile from Shibe Park? If the answer is yes, look this place up if you're in Denver.