FanGraphs Weekly Mailbag: June 21, 2025

By most measures, the Rafael Devers trade happened suddenly. It came without advance notice of his availability, and the Red Sox reportedly weren’t shopping him around. Immediately, it drew comparisons to the Luka Dončić-Anthony Davis trade in the NBA, because hardly ever in our scoops-driven media landscape, where even the tiniest rumor is treated as currency, does a transaction involving a superstar catch us by surprise.

And yet, now that the shock has worn off, trading Devers feels like a logical outcome to the saga that began in March, when the Red Sox signed Alex Bregman to play third base without giving the incumbent a heads up. The details of the ensuing rift have been covered at great length, at FanGraphs and elsewhere, so I won’t go into them here. A lot of the reporting since the trade has described the situation in Boston as untenable, and the damage done to the relationship between Devers and the team as irreparable. But based on how badly the Red Sox botched their initial response to the conflict, and then kept bungling their subsequent attempts at reconciliation, from my perspective, it seems like they didn’t make repairing it much of a priority.

We’ll tackle your questions about the Devers trade and so much more in this week’s FanGraphs mailbag. But first, I’d like to remind all of you that while anyone can submit a question, this mailbag is exclusive to FanGraphs Members. If you aren’t yet a Member and would like to keep reading, you can sign up for a Membership here. It’s the best way to both experience the site and support our staff, and it comes with a bunch of other great benefits. Also, if you’d like to ask a question for next week’s mailbag, send me an email at mailbag@fangraphs.com.

We received two questions about the Devers trade this week, so we’ll begin there. The first one comes from my former Sports Illustrated editor and good friend Connor Grossman, who is also a huge Giants fan and the writer of the Giants Postcards newsletter. The second question is from Jeffrey Gonzalez, who also asked a Hall of Fame question that Jay Jaffe answers later in this mailbag. I’ve included both Devers questions below because they’re similar enough. Let’s get to it:

__

How onerous is the Devers contract in reality? The Giants have been missing a slugger since Barry Bonds left almost two decades ago. And it’s been longer than that since anyone has hit 30 home runs in their uniform. The contract shouldn’t matter, right? — CG

In the aftermath of the Devers trade, a number of commentators (not so much on FanGraphs, but elsewhere) have focused on the defensive limitations of Devers relative to his large contract. I’ve heard a lot about the “clogging” of the DH spot; on a baseball podcast recently, a host mentioned that if, say, Heliot Ramos has a bad ankle and needs to DH to get on the field, Devers blocks him.

My issue with this point is simple: Isn’t Devers the best hitter on the Giants? Won’t he likely be the best hitter on the Giants for the next four or five years? How is it a problem to have him “clogging” a position that prevents him from hurting the team defensively? — Jeff

It seems like you are both on the same page I am. The contract doesn’t matter to the Giants, even if Devers never plays the field again. They appear more than willing to pay him the more than $254 million he is owed over the eight-plus seasons left on his deal because for a big-market team in need of offense, no cost should be too high for one of the best hitters in the sport. Some discussion of the trade has cited San Francisco’s unsuccessful attempts to sign Aaron Judge, Shohei Ohtani, and Juan Soto in consecutive offseasons to explain why the Giants were so desperate to acquire Devers — as if those failures compelled the team to overpay for a flawed player. I don’t think that’s the right framing here. The Giants agreed to take on the rest of Devers’ contract while also giving up giving up Jordan Hicks, Kyle Harrison, James Tibbs III, and Jose Bello for the same reason they offered Judge $360 million and Ohtani $700 million: They haven’t had a bona fide slugger since Bonds retired after the 2007 season. When Judge, Ohtani, and Soto signed elsewhere, the Giants still had the same hole in their lineup. Similarly, if the Giants decide to have Devers play first base next season, as a way to shore up their weak offensive output from that position and to free up the DH spot, I don’t think they would consider the contract any worse. He might provide negative defensive value there, but he’d more than make up for it with the bat — the same way he was making up for his negative fielding value at third base in previous seasons. And if freeing up the DH spot let’s the Giants get another strong hitter in the lineup, the contract will seem even more valuable. The point is, where Devers plays should have nothing to do with how we evaluate the contract. He’s there to mash.

To the Red Sox, on the other hand, the contract mattered a great deal. Surely, it didn’t when they first signed him to the 10-year, $313.5 million extension before the 2023 season (it went into effect in 2024), or else they wouldn’t have offered it to him. But things change. One of the biggest differences is in the baseball operations department. Chaim Bloom, who was in charge when the Red Sox extended Devers, was fired in September 2023, before the contract even kicked in. Boston hired current chief baseball officer Craig Breslow that October. After signing Bregman, the Red Sox were still willing to pay Devers to “clog” their DH spot, even though they initially planned for him to be their third baseman of the future, because they knew the value of his contract would come from his bat, not his glove. I think that’s at least part of why they were so uncompromising with his move to DH, saying he wouldn’t play third base even on Bregman’s days off or take reps at first base; they were fine, even happy, to have him only DH. It was only after Devers became disgruntled and the dispute spilled out into the public that the contract started to matter. Even so, Devers acquiesced and became the full-time DH before Opening Day, and once he caught fire following his rough start, the disagreement seemed to be behind them. That is, until Boston requested that Devers learn to play first base when Triston Casas went down with a season-ending injury and he refused to change positions again.

“The team’s feeling was that a $313.5M contract comes with responsibilities to do what is right for the team and that Devers did not live up to those responsibilities,” Peter Abraham of The Boston Globe reported the night of the trade, citing Red Sox sources. “They had enough and they traded him.”

As was the case with San Francisco, Boston’s side of things comes down to roster construction as well. The Red Sox have a surplus of talented young position players, and they were going to need to solve the roster logjam at some point. Trading Devers doesn’t clear everything up — they still have more outfielders than they can use — but his absence does unclog the DH spot. Of course, nobody on the team is as productive at the plate as Devers, but his contract became expendable in a way it wasn’t before.

––

Does MLB require any sort of validation, like a doctor’s note, for players to be put on the bereavement list, and are there any sort of limitations on what qualifies in the CBA? I wonder if it ever gets abused like a short-term IL holding place. — Randy Kaufmann

As far as I know, the bereavement list has never been abused, nor has anyone tried to abuse it. This makes sense, because it would be incredibly difficult for a team to lie about the death of a player’s immediate family member. The actual roster designation here is the “Major League Bereavement/Family Medical Emergency List,” but while I suppose it would be easier to fake a family medical emergency, I can’t imagine anyone would try it.

I know professional athletes are more competitive than most people, and baseball history is littered with examples of players breaking the rules — sometimes in ways that none of us could have predicted — in order to win, but I don’t think anyone would fake a health crisis involving a player’s family member just to free up a roster spot, especially when the phantom IL is still a fairly easy loophole to exploit.

Even so, I checked The Official Professional Baseball Rules Book — which is different from the baseball rulebook, and governs non-game issues like the draft, reserve lists, waivers, and optional assignments — just to see how the process works. According to Rule 2(c)(2)(A), during the regular season, a player can be placed on the Bereavement/Family Medical Emergency List only when “such player is unable to render services because of the serious or severe illness or death of a member of such player’s immediate family (e.g., spouse, parent, grandparent, sibling, child or grandchild) or a member of such player’s spouse’s immediate family.” The rule doesn’t say anything about whether specific documentation is needed to prove the legitimacy of the death or illness, but the team has to submit a written application to the commissioner’s office requesting to place the player on the list. If approved, the team must then notify the player in writing of the placement.

That’s for the regular season. The process is slightly different in the postseason. Rule 41(a)(5), which covers the Postseason Bereavement/Family Medical Emergency List, includes this additional sentence:

“The Office of the Commissioner shall carefully scrutinize such application, including any documentation submitted by the Club in furtherance thereof, and reserves the right to deny such placement in the absence of appropriate evidence to support such placement.”

The postseason rule mentions documentation, and its tone is more stern, as if to dissuade anyone from trying to abuse it, but it remains light on specifics. What documentation? What or how much evidence would be considered appropriate? More than anything, I think the postseason rule is written in a way that provides the league some sort of cover if it ever needs to deny an application when the stakes are the highest. There is no record I could find of the commissioner’s office ever denying a player’s placement on the bereavement/family emergency list, in either the regular season or the playoffs, and I don’t expect this is ever going to happen. Think of the public relations nightmare that would come from rejecting, or even merely questioning, a player’s right to grieve or deal with a family emergency.

Sure, a cold-hearted club could theoretically see this as a reason to try to exploit the list — the league wouldn’t dare refuse our request — but in reality, I think the opposite is true. Similar to the Paternity List, if players know they won’t face repercussions for stepping away from their teams to be with their families during matters of life or death, they’ll be more likely to take the time they need. And just as I don’t think the commissioner’s office would ever be insensitive enough to suggest someone were faking a family emergency, I don’t believe any of the players or teams would ever be so crass that they would lie about such a tragic situation.

__

I’m curious as to why the sabermetric community uses BABIP as a tool, but not BACON. I haven’t checked myself, but I would imagine that past BACON has more correlation to future BACON than past BABIP does to future BABIP. In some cases, we look at BABIP to get a feeling if a player has been lucky or unlucky, but I imagine that BACON is a more useful tool for the same purpose using the logic that if you hit the ball hard enough to get a home run, you probably “deserved” that hit. Just curious what you think! Thanks. — Joe Meyer, FanGraphs Member Since 2016

Ben Clemens: Sizzling question. Let me see if I can strip the answer down for you. You’re totally right! I ran the numbers for hitters who played in 2023 and 2024, and BABIP had a 0.39 correlation to next-year BABIP, as compared to 0.50 for next-year BACON. The reason for this is pretty simple: Home runs are much stickier than other hits. If you smoke the ball in the air frequently, you’re going to end up with a lot of them. If you don’t, you won’t.

(A quick refresher for everyone reading, because as Joe noted, BACON isn’t a metric the sabermetric community uses much. It stands for Batting Average on Contact, meaning it includes home runs, whereas the denominator for BABIP is batted balls in the field of play.)

The question, then, is whether we want this. I do. Who doesn’t want BACON? But I want more than just BACON! I like to use wOBACON, because if we’re going to try to add in homers, we should cure the greatest weakness of batting average: It treats a single and a homer as the same. By weighting them based on run values, we drain off some of the randomness. I think wOBACON is a great statistic, and it’s more reliable than either BABIP or BACON, with a 0.64 correlation to next-year wOBACON. By amplifying the signal you get from extra-base hits and turning down the signal you get from singles, you do better at finding what batters are likely to do in the future.

Why use a ham-fisted stat like BABIP over an elegant one like BACON? You can think of BABIP as the fat we slice off the delicious wOBACON. Not every hitter has the same BABIP – but nearly every hitter’s BABIP is noisy. Take Luis Arraez. A career .332 BABIP hitter, he posted a .362 BABIP in 2023 and has a .276 mark this year as of Friday morning. Rendering a batted ball into a single? Tricky and random. A homer? Lots of skill involved. Thus, I like to treat BABIP like an error term – I look at it to get an idea of whether that batted-balls-to-singles noise is helping or hurting a player at the moment. If I want stability and “deserved” stats, I’ll either look at a longer time horizon for BABIP (more observations smooths the variance) or wOBACON. And don’t even get me started on Canadian wOBACON, which would be an excellent nickname for Addison Barger.

__

As someone who works in analytics (the business and actuarial sort rather than the baseball sort) I am continuously fascinated by season projections for teams (ala Win Totals and Playoff Odds). Player projections are cool too, but too much random variance for my tastes. So my questions are about the mechanics of the models, their interpretation, and tuning.

(1) At what point is actual in-season data more credible than the thousands (or potentially hundreds of thousands) of simulations run preseason? Are some factors, scenarios, or sections of the schedule more or less impactful to projections than others? Possible examples could be a team with a high projected winning percentage that is temporarily not at full strength, or a challenging 15-20 game stretch against comparable teams (coin-flip scenarios) where performance is materially above or below expectations.

(2) How is simulated performance sensitivity contemplated in end-of-season projections as the season progresses? Specifically, are there some factors from a team’s preseason simulations that may make them more or less resistant to an adjustment in projections as the season progresses? I am thinking about outcome variability in preseason projections. For example, over/under performance of a very good team like the Dodgers, that (I am assuming) has a very narrow outcome distribution vs. teams like the Rays or the A’s that have a roster construction that may result in larger simulated-outcome volatility. Are in-season projections similarly impacted?

(3) How much is contingent probability (apologies if my layperson understanding is resulting in the wrong term) considered in playoffs and World Series odds? I’d imagine this is contemplated as the chances of making the WS is contingent on getting to and through the playoffs (not a level comparison due to the two leagues and the playoff process/seeding). How are the other teams in the mix contemplated? Are there any circumstances where making an assumption of one constraint (i.e., a team makes the playoffs or achieves a bye) materially impacts their projected odds? While it’s been said (and understood) that the playoffs are a crap shoot (now more than ever), understanding the impact of contingent factors on outcome probability is interesting in my day job too. — Tony

Dan Szymborski: Hi Tony, thanks for reading!

To answer the first point, if you’re talking on a team level, there’s never a point where the team’s actual record is a better rest-of-season predictor than the preseason projected record, even with the preseason projections having no idea about the current roster.

In fact, as weird as this may seem, you would still expect the 2025 preseason projected standings to be better predictors of 2026 than the actual 2025 full-season win-loss record! As it turns out, the negatives of recency bias are huge. This is observed in 20 years of ZiPS, and I imagine it holds the same for PECOTA or other methods that project in-season standings.

I can’t tell precisely how our site does it as I’m not involved in the coding of our various things such as this, but I can tell you how it works when I do a ZiPS run.

ZiPS calculates the variance in team projections by starting off first with an initial simulation that estimates the distribution of playing time. So when the Yankees are projected in a full ZiPS run, sometimes Aaron Judge will be very healthy, sometimes he won’t be, and very occasionally he’ll be out with a serious injury. ZiPS then replaces those plate appearances/innings with other players on the depth chart (subject to their own injury risk).

For the playoffs, because ZiPS simulates the season a million times for the full-season standings and has the playoff matchup rules built in, it can use the actual playoff opponents in each simulation. It also has the first three tiebreakers coded in so it can get the matchups correct; if there’s a tie beyond that point, every team has a little random number seed to break the tie. This is a useful kludge in that getting that far in the tiebreakers is rare, and a random answer is just fine for a mass simulation.

__

This question is for Jay: Have you discerned any patterns in the post-BBWAA voting groups? In other words, do you feel comfortable predicting whether anyone who has fallen short in their 10 years will get in via a different committee? My money’s on Torii Hunter. — Jeff

Jay Jaffe: Thanks for your question! There are indeed patterns in the post-BBWAA groups. Unfortunately, the most overarching trend is a dubious one that has carried over from the Veterans Committee to the Era Committees. Allow me to begin by quoting myself, from my 2017 book, The Cooperstown Casebook:

No element of the Hall of Fame has opened it up to more complaint and ridicule over the years than the Veterans Committee, which served as the side door for membership from 1953–2010, flanked on either end by committees serving a similar purpose: electing those outside the purview of the Baseball Writers Association of America. Comprised of between 10 and 20 executives, writers, and Hall of Fame players, the VC fleetingly rescued the Hall from great embarrassment, functioning as a righter of historical wrongs by electing players overlooked by the writers. But all too often, the group scraped the bottom of the barrel by electing substandard candidates, many with clearly traceable connections to committee members, opening it up to charges of cronyism.

That’s the leadoff for a whole chapter, “The Hall of Cronyism,” with a particular focus on the damage done by the 1967-76 VCs led by Frankie Frisch (1967–72), Bill Terry (1971–76), and Waite Hoyt (1971–76), who together elected a whole bunch of former teammates who rank among the worst players in the Hall at their positions according to JAWS. Generally they had shorter careers than the BBWAA-elected players, and their superficially shiny offensive statistics were inflated by the high-scoring environments of the 1920s and ’30s:

Four of Frisch’s former teammates were elected during his tenure: shortstop Dave Bancroft (1971) and right fielder Ross Youngs (’72) from the Giants, and pitcher Jesse Haines (’70) and left fielder Chick Hafey (’71) from the Cardinals. He had managed a fifth, center fielder Lloyd Waner (’67) in 1940–41 with the Pirates. Terry, whose career ran from 1923–36 with the Giants, and Hoyt, whose career spanned 1918–38 with the Yankees, Pirates, and five other teams, each had four teammates from among this group elected, with some crossover between the two.”

Five of the players the VC elected during this period (Hafey, Haines, Waner, Freddie Lindstrom, and George “High Pockets” Kelly, a former teammate of both Frisch and Terry) rank dead last in JAWS at their positions.

After numerous changes in the configuration of the Veterans Committee process — currently divided into three Era Committees, one for pre-1980 players and non-players, one for post-1980 players, and one for post-1980 managers, executives, and umpires — cronyism remains a problem. The Hall has actually baked it into the process by selecting managers, executives and former teammates of certain candidates in such a way that I’ve had to incorporate the pre-election announcement of the committee into my coverage. When Harold Baines was elected on the 2019 Today’s Game Era Committee ballot, the committee included Jerry Reinsdorf, for whose White Sox Baines played during three separate stints (the team actually retired his number while he was still active!); Tony La Russa, who managed Baines in Chicago and Oakland; and Pat Gillick, who as general manager acquired and then re-signed Baines for the Orioles. Lee Smith, also elected on that ballot, played under committee member Joe Torre in St. Louis and was teammates with committee members Greg Maddux and Ozzie Smith. When Fred McGriff was elected (unanimously) on the 2023 Contemporary Baseball ballot, the panel included former teammates Maddux and Kenny Williams, and former Blue Jays executive Paul Beeston. Dave Parker was elected by a 2025 Classic Baseball panel that included former teammates Tony Perez and Paul Molitor as well as former A’s executive Sandy Alderson.

The deck-stacking works both ways. Former Major League Baseball Players Association executive director Marvin Miller was stonewalled during his lifetime by panels that included many executives (and their descendants, also executives) whose lunch he ate again and again during labor negotiations. The McGriff ballot, which also included Barry Bonds and Roger Clemens, featured outspoken critics of performance-enhancing drug usage, such as Frank Thomas, Ryne Sandberg, and Jack Morris.

But let’s set that unsavory aspect of the process aside and consider what else we’ve learned from the committee process. For one, because the BBWAA gets first crack at most candidates, the quality tends to be higher than those elected by the committees — and by a wide margin (all WAR figures I’m citing are from Baseball Reference):

That’s about a 20-WAR gap between BBWAA and non-BBWAA-elected catchers, and likewise for starting pitchers, widening to over 30 WAR for non-catching position players; in terms of JAWS, the BBWAA-elected players range from 10.8 to 15.6 points higher. I didn’t bother to do this for relievers, as Smith is the only committee-elected one out of eight.

For the above table I didn’t include any players honored primarily or entirely for their time in the Negro Leagues, such as Satchel Paige, Willard Brown, and Monte Irvin, nor did I count 19th Century pitcher-turned-shortstop (and labor leader) John Ward. None of those players are in the JAWS set, but the non-BBWAA honorees within it include 19th Century players who played shorter seasons and, for pitchers, totaled many more innings in a season. (I didn’t use S-JAWS here.) Still, it’s a pretty jarring discrepancy between the two classes of honorees.

When the late Joe Gordon was elected to the Hall via the 2009 Veterans Committee ballot, it marked the first time a player had been elected by the VC since 2001. He’s the first of 15 players elected during what we might call the Reconfiguration Era, a period during which the institution has changed voting formats every few years, mainly to redraw the boundaries for the periods under consideration on a given ballot:

There’s no overarching trend that accounts for all of these guys. Some were reasonably or even exceptionally well-supported by the writers but fell short of 75%, others were criminally undersupported, including Simmons, to date the only player to fall victim to the Five Percent Rule and still gain eventual entry.

At best, we can group these players into subclasses. Allen, Gordon, Oliva, and Santo — a group that includes the best of the bunch in terms of WAR and JAWS — were short-career stars who didn’t stick around long enough to reach the kind of milestones that tend to guarantee induction. Gordon’s military service, Oliva’s knee injuries, Santo’s diabetes, and Allen’s general iconoclasm all contributed to the brevity of their careers, and it took a long time to honor them. Of that group, only Santo is above the career WAR standard at his position, while all but Oliva are above the seven-year peak standard.

At the other end of the spectrum, Baines, Hodges, Kaat, McGriff, Morris, Parker, Simmons, Smith, and Trammell were compilers to one degree or another. They had long and distinguished careers, but only Simmons and Trammell have peak scores in line with the standards at their position. Of that group, only Parker won an MVP or Cy Young award, with Trammell the only other one who came close. Baines, McGriff, and Kaat fell short of the major milestones that grant near-automatic entry, with voters doing the equivalent of crediting them with the phantom hits, homers, or wins that would have made them automatic. (Notably, they did not do this for Tommy John, who has more wins than Kaat as well as the medical marvel hook.) None of those guys ranks well in JAWS, while Miñoso, Simmons, and Trammell join Allen, Gordon and Santo among the top 20 at their positions.

Which brings us to Hunter, who in 19 years in the majors amassed 50.6 WAR while collecting 2,452 hits, 353 homers, and nine Gold Gloves. He’s just 35th among center fielders in JAWS (30.8), and so far has never received more than 9.5% of the vote, the share he debuted with in 2021. On the 2025 ballot, he received 5.1%; one fewer vote would have meant the end of his run. There’s no obvious analogue to him among the recently elected players above; his ballot tenure has been Baines-like (Baines fell off after his fifth year) but his WAR and JAWS are closer to McGriff. He outranks only five of the 19 non-Negro Leagues center fielders in Cooperstown, and also trails Andruw Jones and Carlos Beltrán, both of whom are likely to gain entry within a few years.

Hunter’s credentials are likely to have more appeal to a small committee that’s roughly half former players than to the writers, but even so, he doesn’t have an obvious hook; several candidates with similarly high Gold Glove totals remain outside the Hall. What’s more, even a spot on the ballot is hardly assured, as he’s likely to wind up in a pool that will include some combination of Bonds, Clemens, Albert Belle, Dwight Evans, Jeff Kent, Don Mattingly, Dale Murphy, Rafael Palmeiro, Curt Schilling, Gary Sheffield, Sammy Sosa, and Lou Whitaker. Most of those players received better support from the writers than Hunter. The Hall would love to cull the PED-linked players from that herd, and introduced a rule in March that’s designed to winnow it down if they don’t get enough support; again, the institution can wield a heavy hand in that depending on whom it picks for the committee.

Because of the anti-PED bent, I’m reluctant to suggest Bonds, Clemens, Palmeiro, and Sosa as likely honorees; Sheffield, who maxed out at 63.9%, is probably the best bet of that bunch, as his connection to the drugs is tenuous. Among those I listed, I think the most likely to get in are Kent and Whitaker, both of whom have big counting stats for middle infielders, and the latter of whom (who went one-and-done on his only BBWAA ballot) ranks 13th in JAWS and should have ridden Trammell’s coattails into Cooperstown. Particularly with a two-time MVP center fielder (Murphy) in this pool, I don’t see an easy path for Hunter. Then again, it’s not hard to imagine a future committee that features former teammate Joe Mauer, manager Jim Leyland, and GM Dave Dombrowski, all of whom have clear connections to Hunter. The first two are already in Cooperstown, and I believe the third — who’s served on a couple committees in the past, including the one that elected Baines and Smith — will be there some day as well. Try to look surprised if it happens.

Source