The BBWAA is about to reveal whether they think a guy who spent a majority of his time DHing (Frank Thomas) and a guy who spent near all his time as a relief pitcher (Lee Smith) are worthy of induction. We will reveal our predictions about who will go in tomorrow, and in the meantime, these are two positions that arouse some debate about the nature of value, the quality of statistics, and the need for balance across positions. And since in our own deliberations we are or will soon be encountering strong candidates at each, it’s time we talked about them.
Eric: The DH is a position, and we should all accept that fact and move on. Paul Molitor was a DH by plurality, and the moment that The Big Hurt gets his plaque, he will be the first majority DH enshrined. Edgar Martinez, nearly as good a hitter as Thomas, merits induction as well. On the other hand, things get complicated when we start asking ourselves how to compare DHes to pre-DH guys who should have designatedly hit.
Miller: To establish that we’re starting from the same baseline, I want to agree up front the designated hitter is a position. It has been for forty-one years and we’re still debating it. I find that absolutely incredible. The continued debate speaks to how engrained some ideas are in the minds of some baseball fans. We’ve accepted interleague play. We’ve accepted the wild card and a second one. We’ve accepted that the All-Star Game means something even though it’s not managed like it means something. But some still act like being a DH isn’t the same as being a baseball player. It’s strange that the people who pooh-pooh the DH don’t similarly suggest that pitchers in the American League aren’t playing real baseball because they don’t hit. Okay, it’s not strange. It’s perfectly logical. Nearly everyone hates watching pitchers hit – even those who don’t like the DH.
You bring up a great question about comparing guys like Edgar Martinez to guys like Willie Stargell or Chuck Klein, wretched defenders who didn’t play in a league or at a time when they could benefit from designated hitting. While it’s an interesting philosophical discussion, I think it’s ground we’ve already covered at the HoME. We judge guys by the context in which they played. I’m pretty sure George Wright would struggle if plopped into the shortstop position at Dodger Stadium this April. Hell, I’m not even sure he’d understand how to play the game. We give Jimmy Sheckard extra credit defensively, and deservedly so, because left field when he was playing it was roughly like we view center field today. And we’ll never know if Charlie Bennett’s skill set would have been HoME-worthy had it been employed 100 years later. But we certainly know he’s worthy of extra credit for being incredibly durable at a brutal position at its most historically difficult time.
I could go on, but that would be like beating a dead horse, which you should never do.
I’m interested in modern players who didn’t get the benefit of DHing but could have. Or should have. Dick Allen played for five years after the DH was adopted, but he was allowed to stink it up in the field. The aforementioned Willie Stargell would have a much stronger case if he didn’t have such awful defensive numbers. Bernie Williams has his supporters, in large part because he’s compared to a group that’s relatively weak historically, center fielders. If you saw Bernie play those last half-dozen years or more, you’d support the idea that he was a center fielder in name only. Craig Biggio gets a lot of Hall support, and deservedly so, but he wasn’t a good catcher, he was bad in center, and he was a pretty miserable second baseman. You know who could have played a miserable second base? Frank Thomas, that’s who.
Eric: The list of guys who would have been DHes in the sixties and early seventies includes Stargell and Allen (the latter DH’ed late in his career) as well as Killebrew and McCovey. Boog Powell. Dick Stuart, of course. The question is whether we should adjust our expectations for a DH as we compare them to contemporaries and predecessors. For example, WAR thinks Dick Allen cost his teams 110 runs, about a win for every 650 plate appearances. Say that they had been able to make him a DH after his fifth full year, just like Edgar. That would have added back 60 runs to subsequent seasons. Or six wins. Give him six more wins, and his career value vaults into the realm of McCovey, Dawson, Reggie Smith, and Dave Winfield, guys he beats handily in peak value.
So let’s take a quick look at what WAR thinks the positional adjustment is for a DH versus a first baseman. Taking Edgar as our example, he played two games in the field from 2001 through 2004, so that’s close to a pure DH. The positional adjustment in those seasons was
2001: -12 runs in 581 PA, or .021 runs per PA
2002: -9 runs in 407 PA, or .022 runs per PA
2003: -13 runs in 603 PA, or .022 runs per PA
2004: -12 runs in 549 PA, or .022 runs per PA
So .022 runs per PA.
In those same seasons, another candidate, Jeff Bagwell, played a mere six games at DH and nearly every day at first base. Here’s his positional adjustments:
2001: -9 runs in 717 PA, or .012 runs per PA
2002: -9 runs in 691 PA, or .013 runs per PA
2003: -9 runs in 702 PA, or .013 runs per PA
2004: -9 runs in 679 PA, or .013 runs per PA
So .013 runs per PA, at least during those four seasons. [The positional adjustment floats based on the league’s average offensive performance at that position.] DHes got a 69% more difficult adjustment per plate appearance than a first baseman. Over the course of 650 PA, that’s about 6 runs, a little more than half a win’s difference.
Or to put it in plain English, WAR is telling us that the hitting expectation for a typical DH is six runs higher than for a first baseman. From which, we can infer that to make it worth their while to DH a player, a team would have to expect his glove to be six runs or more below average annually at first base. For Dick Allen or Harmon Killebrew, that might have been a reasonable tradeoff. Or for Derek Jeter.
Or in the cases of Edgar or Molitor, they’d have to expect lots of DL time without the DH option. Larry Walker should have played in the AL.
Miller: I can answer simply. No, we shouldn’t adjust for bad fielders before the DH. To do so would be inconsistent with our other HoME rules.
Also, not that it matters much, but I think Jeter would be an average or above average glove at 1B.
Eric: Then should we adjust DHes further downward than WAR does? Let’s take 1965-1968 as a sample. The worst first baseman in each league (minimum 300 PAs) were respectively worth -5 runs (Donn Clendenon), -12 (Deron Johnson), -6 (Ed Kranepool), and -4 (Rusty Staub). Other than Johnson, those figures are in line with the estimate for the difference between the DH adjustment and the first base adjustment that we arrived at above of -6 runs (for 2001-2004). The worst first basemen in baseball were about as hapless in the field as the DH adjustment we look at said they would be.
Let’s try another quick sample: 1955-1958. Our “winners” here are, respectively, -7 Mickey Vernon, -7 Roy Sievers, -6 Ray Boone, and -7 Ed Bouchee. Same thing. The DH penalty appears to align well with reality, so no we should not adjust DHes any more than they are.
Or to put it in simpler terms: just vote for ’em.
Miller: While this doesn’t make for interesting conversation – I agree.
WAR already makes corrections for the suspected lousy defense of the DH. It gives us a good estimate of how much the White Sox or the Mariners would have been hurt had they been foolish enough to use Frank Thomas or Edgar Martinez in the field more than they did.
Rather than punish the players, voters just acknowledge that DH is a position and can be understood with positional adjustments applied, just like any other.
Eric: And on to the relief pitchers. Lee Smith makes this conversation germane.
There’s a huge disconnect between the narrative around relievers and what’s observably true about them. For instance they rarely get long contracts. The best of them are paid about half of what star players make. They are replaced all the time. They are failed starters no matter how you slice it. Yet every blown save demonstrates how important shutdown relievers are. Especially now when the bullpen has gone from tactical weapon to strategic asset. What’s a Hall of Fame voter to think?
Miller: Your entire comment, Eric, it seems to me, points to the way teams view closers today and the way Hall voters will view closers in five or twenty years.
Teams had a learning curve. In 1977, the San Diego Padres signed Rollie Fingers, age 30, to a six-year contract. In 1985, the Atlanta Braves signed Bruce Sutter, age 32, to a six-year contract. Hell, in 2012, at age 31, Jon Papelbon got four years plus a vesting option.
My point is that closers used to get long contracts and get paid big money. They don’t so much today because teams have wised up – they had to wise up. If they didn’t, they wouldn’t be competitive. It makes sense that the BBWAA made mistakes with relief pitchers, just like teams did. However, it doesn’t make sense that they’d catch on as quickly because they have no financial interest in catching on.
What’s a Hall of Fame voter to think? It’s simple. Just wait. Wait until you’re sure. The relief position is still evolving 40+ years after the Save statistic was created. Writers need to be patient. They need not elect too quickly. Failing that – and we already know they’ve failed – they should do some math, maybe the math that I’m going to offer below.
There have been about 140 years of baseball history, give or take. And there are about 70 pitchers in the Hall. That means about half a pitcher per year. If we take the first 40 years of the save statistic, from 1969-2008, we’re looking at about 20 Hall of Fame pitchers on average – half a pitcher ever year. How many should be closers? Well, let’s keep the math going. Teams pitch 162 * 9 innings per season, or 1458 innings. If closers pitch an average of 75 innings per season, we’re looking at about 5% of the total. So perhaps closers should be 5% of the pitchers in the Hall. That would mean two from 1969-2008. I’ll take Mariano. Who ya got?
Eric: To be fair to the players, the BBWAA has been incredibly stingy with starting pitchers. You and I could probably think of half a raft of starters off the top of our heads who are Hall-level guys that the BBWAA has so far whiffed on because, anymore, they only elect 300-game winners (or damn close in Blyleven’s case). They’ve elected a grand total of 1.5 starters since 1999. One! Point!! Five!!! In thirteen years! In that time, they’ve only picked one-half of Eck’s career and Blyleven. And they only voted for Bly under threat of constant ridicule if they failed him. In fact, the last time they chose a starter who won fewer than Blyleven was Jim Palmer in 1990, and the last time they went for a guy with fewer than 250 career wins was the utterly ridiculous election of Catfish Hunter.
There is a certain delicious irony when the old-time writers complain about us young whooper-snappers all on their lawn with our fancy numbers, and they claim “It’s not the Hall of Stats.” Actually, for the BBWAA, it is. They only elect starters with huge piles of career wins because a chunk of their electorate doesn’t appear to understand that pitcher wins are no longer the barometer of greatness.
Anyway, in my eyes Dave Stieb, Rick Reuschel, Luis Tiant, David Cone, Kevin Brown, Curt Schilling, Bret Saberhagen, and, of course, Roger Clemens are clearly qualified for the Hall. And strong cases can be made for Kevin Appier, Chuck Finley, and Orel Hershiser. Figure Greg Maddux, Tom Glavine, and Mike Mussina are all overqualified, and you can add as many as a dozen starters to the Hall’s ranks. So we should really be thinking about 30 Hall pitchers during the Saves era. In which case, at 5%, we might have room for three relievers. I’d still say, however, that it’s Mo and one other guy, because when John Smoltz is elected, he and Eck can count for one reliever. Even if we wanted to say that relievers’ innings are worth double due to their leverage, that still gets us to only six. Tack on Hoyt Wilhelm and Goose Gossage, and maybe Trevor Hoffman or Billy Wagner someday. But for Pete’s sake, Fingers? Sutter? And pray no, Lee Smith….
Miller: I’d take Lee Smith if they let us kick out Sutter.
But back to the point, I don’t think you’re wrong with your math. I’m just inclined to be more conservative with closers since we haven’t had so many years to observe what’s still a changing position.
Eric: Here’s yet one more way to slice/dice the election of relief pitchers: There are 4.5 relief pitchers in the Hall (calling Eck one-half a reliever). All have been elected by the BBWAA since 1985. In that time, the writers have elected 10.5 starting pitchers. That’s 2.3 starters for every reliever—zany! If that rate held up, and the writers over the next several years elected Maddux, Glavine, Clemens, Schilling, Mussina, Smoltz, Randy Johnson, Pedro Martinez, and Roy Halladay by 2019, they’d not only be electing Mariano, but also Hoffman, Wagner, and Smith. The clamor for Smith is pretty tame (and mostly comes from Chicago sports writers), and no one talks about Wagner as a Hall reliever. And while Hoffman holds the saves record (as Smith did once) does anyone really think he belongs with Schilling and Smoltz, let alone Maddux and Pedro? Well, OK, some will, but I find it hard to take the premise seriously.
Look, the writers haven’t gone crazy with relievers. It’s still only two or three more than I’d like to see, but they’re not going Frankie Frisch on them. Still, with any luck, the electors will foreswear any closers until Mariano comes along, which will do much to improve that ratio. That is, provided they get a handle on starters in the meantime.
This whole thing does raise a question for me about the relative value of DHes and relief pitchers. Molitor is the only thing close to a DH elected. He was also the first truly worthwhile DHish candidate. Edgar was next, Thomas is now here. Even if they were both elected this year, it would still represent a player:DH ratio of about 10:1. On a seasonal level, full-time DHes and relievers are probably close in value. But DH is a much more difficult position to fill than relief pitching, in fact, Tangotiger says that a replacement level DH is a league average hitter. Relievers are failed starters with one or two good pitches. DHes are failed defenders who can hit. Yet, because teams are strongly incentivized to keep great hitters in the field as long as possible before making them a DH, Miguel Cabrera and Prince Fielder hold on to their gloves. Hell, so did Cecil Fielder. By the time they reach the DH stage in their career they are past-peak and winding down their thirties. (Exception: David Ortiz!) The opposite is usually true of relievers whose weaknesses as starters are exposed quickly by major league hitters, pushing them to the bullpen for their peak seasons. That’s how the best can hang around a while. Especially if the team didn’t give them much rope as a young starter (cough, cough, Mariano!, cough, cough) or simply put them into a place of immediate need (Aroldis Chapman and Jonathan Papelbon) and never returned them to the rotation after initial success.
This is why DH is seen as a position for broken players, and relief pitching a place for young guns with blazing fastballs. If you asked Edgar, he’d tell you that perception is everything.
Miller: Look how nice you’re being to the writers by calling Eck half a starter. I assure you that no part of Eck the starter was elected to the Hall. It was all reliever, which makes it a 2:1 starter reliever ratio elected since 1985. Welcome upstate John Franco!
On the other hand, I don’t want to just brush aside relievers either. Trevor Hoffman is #18 in Win Probability Added (WPA). WPA is a counting stat, which isn’t context neutral, unlike so many other new-fangled stats. It basically measures how a player’s individual contributions add to (or subtract from) a team’s chances of winning. Trevor Hoffman is #18, right between Curt Schilling and Juan Marichal. As for others, Fingers is #94, Sutter #77, Smith #57, Wagner, #31, Eck #28, Wilhelm #26, Gossage #24, and Mariano is #3. John Franco fans will be happy to know their hero is #70.
My point isn’t that we should follow WPA, though a Hall following reliever WPA would be better than our current collection. I’m suggesting that I don’t yet know what to follow, so I don’t want to close the book on guys like Hoffman, for example, just because the writers blew it on Sutter.
As for your premise that DH is a much more difficult position to fill than relief pitching, well, I guess I agree. However, we’re not talking about relievers in general – we’re talking about closers. I don’t think it’s meaningfully easier to find a league-average closer than a league-average hitter. If the Red Sox were to lose David Ortiz, they could easily turn to Dustin Pedroia or Mike Napoli or Shane Victorino. Voilà, brand spankin’ new DH. I suppose I agree with your point; I just don’t agree with the degree to which you went.
Further, relievers might be failed starters with one or two good pitches. But closers are failed starters with one or two great pitches. There’s a difference between good and great, right?
And one more thing, what’s the incentive to keep a great hitter in the field as long as possible? Is that incentive losing games? Getting fired? Miguel Cabrera and Prince Fielder weren’t in the Motown infield last season because of an incentive to keep them out there as long as possible. They were in the infield because the Tigers had Victor Martinez at DH, a player who they considered a weaker fielder and/or greater injury risk if asked to defend. It’s much easier to find a league-average glove than a league-average bat. Teams prefer the bats, and I think they should.
Eric: Teams stumble into closers all the time. They look around their bullpens or their farm system or the waiver wire and find one. Jason Grilli was out of the majors then suddenly awesome. Jim Johnson had a 3.86 minor league ERA with decent peripherals. Kevin Gregg? Jose Valverde? Todd Jones? They’re just guys, and they are average closers. Heck, John Franco had an awful minor league record, one of the worst imaginable for a closer, but he lasted twenty-one years! I don’t think it’s difficult at all to find an average closer. To find a great closer? Just look at your best a right-handed starting pitching prospect, and voila! Instant super closer.
Miller: So to sum up, elect designated hitters and closers. Just make sure they’re the right ones.
Eric: Good luck to Frank Thomas on Wednesday! Sadly, I think Edgar will need more than luck.
Miller: The good news is that it seems Lee Smith’s luck has run out.