Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Monday, January 28, 2013

Schoenfield: Big Hurt should be first-ballot Hall of Famer

Damn Secret Society He-Man-Haters Club!

tc

All that said, three reasons he’ll have trouble getting in on the first ballot:

1. He only spent five seasons as a full-time first baseman. He did play a higher percentage of his games in the field than Edgar Martinez—42 percent to 28 percent—but the DH factor could work against him.

2. The Edgar Martinez factor. How much better than Martinez was Thomas? Thomas does lead slightly in career Baseball-Reference WAR, 69.7 to 64.4, and certainly has big edges in home runs (521 to 309) and RBIs (1704 to 1261), but if you add up their 10 best seasons, Martinez edges out Thomas 56.1 WAR to 56.0. Their final slash lines are pretty similar: .301/.419/.555 for Thomas, .312/.418/.515 for Martinez (Thomas played 232 more games). The point here is that Martinez actually compares quite favorably with Thomas, but has languished below 40 percent of the vote.

3. The first-year bias. Simply put: Many voters won’t vote for Thomas on the first ballot because they won’t perceive him as a first-ballot guy.

Real numbers or not.

Repoz Posted: January 28, 2013 at 01:09 PM | 40 comment(s) Login to Bookmark
  Tags: hof

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Who Swished In Your Cornflakes? Posted: January 28, 2013 at 02:00 PM (#4356630)
I had no idea Edgar and Frank were so close, offensively. That's pretty awesome.
   2. JJ1986 Posted: January 28, 2013 at 02:13 PM (#4356638)
but if you add up their 10 best seasons, Martinez edges out Thomas 56.1 WAR to 56.0.


I think this means that the DH positional adjustment isn't high enough. Thomas was a much better hitter in these years.
   3. ThisElevatorIsDrivingMeUpTheWall Posted: January 28, 2013 at 02:22 PM (#4356649)
Or it could mean Thomas really was an awful fielder, which Martinez wasn't.
   4. Never Give an Inge (Dave) Posted: January 28, 2013 at 02:36 PM (#4356662)

I don't think the comparison with Edgar will hurt Frank. Even though the former made more All-Star teams, I don't think there's any question as to which one writers thought was more of a "future Hall of Famer" while they were playing. They are both going to be evaluated almost entirely as hitters, and Frank was the more impressive offensive player, with better rate stats, a longer career, and having reached the 500 HR milestone.

I actually think Edgar's early polling on the HOF ballot should be encouraging to Frank. If Martinez, who everyone knew would be an uphill battle for the Hall, can poll in the mid-30s, then Thomas should be a no-brainer.
   5. jacjacatk Posted: January 28, 2013 at 02:42 PM (#4356670)
Thomas' dWAR at BBRef for his 2006/7 seasons when he was a DH and played full-time are -1.2 and -1.3, which I assume is assigned just as the positional adjustment, since obviously you can't cost your team runs defensively while DHing. Given that he had seasons where he was primarily a 1B with dWAR numbers substantially worse than that it does seem odd that the positional adjustment wouldn't be higher, since presumably they were DHing someone else (George Bell in this case in '92/'93) who they thought would be even worse if put in the field.
   6. spike Posted: January 28, 2013 at 02:46 PM (#4356673)
Is their anything more dishonest than the "not a first ballot hof" voter? The idea that players they believe are HoFers would fall off the ballot unless other voters make up for their cork-sniffing must not occur to them.
   7. Never Give an Inge (Dave) Posted: January 28, 2013 at 02:46 PM (#4356674)

Given that he had seasons where he was primarily a 1B with dWAR numbers substantially worse than that it does seem odd that the positional adjustment wouldn't be higher, since presumably they were DHing someone else (George Bell in this case in '92/'93) who they thought would be even worse if put in the field.

Or Frank did not like to DH and claimed it was harder to hit on days when he was DHing, so they played him at 1B even if he was a worse option in the field than the alternatives.
   8. TJ Posted: January 28, 2013 at 02:57 PM (#4356688)
From 1990 through 1997 (1997 chosen since it was in 1998 that the steroid era was rfeally in full swing with the McGwire/Sosa home run chase...)

(Yearly average)
Thomas OPS+: 182
Barry Bonds: 181

Thomas also led Bonds in OPS, Slg, and Batting Average. Bonds led in HR, RBI, and Runs (Thomas was second in RBI and Runs). Those two led Griffey, Edgar, and all the PED guys in all those cats except HR (Thomas was 4th). I'm not saying Thomas was better than Bonds as a player when you take defense and baserunning into account, and maybe not even Griffey over that period. But with the stick, Thomas was a nudge better than the best player of the decade and was the best hitter before the video game numbers of the steroid era pushed him out of the spotlight.

Think that could be a reason why Thomas is so adamant that PEDs were cheating?
   9. David Concepcion de la Desviacion Estandar (Dan R) Posted: January 28, 2013 at 02:57 PM (#4356691)
They weren't so close offensively. If I combine Thomas's 1990 and 2004 half-seasons to make a full year, then here's how they stack up season-by-season by OPS+:

Thomas   212 181 180 179 178 177 174 165 163 146 140 126 125 125
Martínez 186 167 165 164 160 158 158 152 141 139 138 133 122 92 


Edgar was a great hitter no doubt, but he doesn't hold a candle to The Hurt. Then again, few do. Baseball-reference gives Martínez more WAR over 10 years because Thomas's best year was the strike year, he had two fantastic half-seasons which get dropped off by this analysis, Thomas was an awful fielder, and Edgar had three good years playing third base giving him more positional value. But with the stick alone, it's no comparison.
   10. JJ1986 Posted: January 28, 2013 at 03:00 PM (#4356694)
the best hitter before the video game numbers of the steroid era pushed him out of the spotlight


Now the steroid era starts in the 2000s?
   11. just plain joe Posted: January 28, 2013 at 03:12 PM (#4356708)
Now the steroid era starts in the 2000s?


Movable goalposts, duh.
   12. Walt Davis Posted: January 28, 2013 at 03:17 PM (#4356715)
They aren't really that close:

career oWAR

Thomas: 76
Edgar: 63

career dWAR

Thomas: -23
Edgar: -10

Edgar's career Rbat is 532, Thomas's is 690. Thomas's total is 15th all-time. Now there's a big era effect in that statement -- 11 of the top 26 played after 2000 -- but the only modern player ahead of Thomas is Bonds (Pujols will pass him this year or next).

Again, WAR measures value, not greatness. Sure, maybe the Sox would have been better off with Thomas a DH from day one. But in comparison with Edgar, the only greatness either one brings is being a great hitter. And Thomas was clearly a greater hitter than Edgar and he had the longer career.

But then this is really a non-existent issue as Thomas will do much better among voters than Edgar.

As to the DH penalty in WAR -- it certainly does not accurately reflect decisions made by actual teams. From ages 23-32, Thomas was dinged for -177 runs between Rfield and Rpos or nearly 18 runs per season. Edgar never received more than -15 runs in a season via his DH penalty. We saw similar team decision-making with guys like Manny and Sheffield. Teams regularly seem more willing to stick somebody out there costing them 20 or more runs a year rather than shift them to DH where they would "cost" them just 15. It wouldn't be the first time that teams (in general) have made sub-optimal decisions but actual behavior suggests to me that the DH penalty should be more in the range of -20 to -25.
   13. snapper (history's 42nd greatest monster) Posted: January 28, 2013 at 03:21 PM (#4356722)
As to the DH penalty in WAR -- it certainly does not accurately reflect decisions made by actual teams. From ages 23-32, Thomas was dinged for -177 runs between Rfield and Rpos or nearly 18 runs per season. Edgar never received more than -15 runs in a season via his DH penalty. We saw similar team decision-making with guys like Manny and Sheffield. Teams regularly seem more willing to stick somebody out there costing them 20 or more runs a year rather than shift them to DH where they would "cost" them just 15. It wouldn't be the first time that teams (in general) have made sub-optimal decisions but actual behavior suggests to me that the DH penalty should be more in the range of -20 to -25.

Concur 100%.
   14. John Northey Posted: January 28, 2013 at 03:24 PM (#4356730)
In the early 90's there was no question Frank Thomas would be a HOF'er and was the best hitter in the game. But boy did he fall off at 30 (1998).
Pre-30: 330/452/600 182 OPS+
30 and on: 276/389/515 134 OPS+

From age 30 on his best in the slash figures is 328/436/625 (590 his next best slg%) 163 OPS+. Only his Slg from age 30 on reached his average from before 30. That is why he dropped off from HOF voters POV - a big decline which wasn't shocking in some respects (his size and amazing figures before) but still cut down his rep drastically. No all-star games, no black ink, somehow got top 10 in MVP twice (top 5 including a second place finish in 2000 when he had the 163 OPS+, over A-Rod at SS with a 163 OPS+, 2 others with higher OPS also below him in voting - voters were pretty dumb that year as 2 guys with 10+ WAR were below Thomas and his 5.8...Thomas was 8th best among those who got MVP votes in WAR). His other top 10 was his last really good year in Oakland with a 140 OPS+, again voters not really thinking through how much less a slow DH is worth vs a catcher who had a better OPS (Mauer)...he was 29th in WAR vs the others who got MVP votes that year.

That amazingly quick decline, from 180 OPS+ guy (181 at 29) to 130 OPS+ (126 at 30 & 31) in one year really shot down his automatic HOF ticket. He still should get in, but isn't the 'wow' he once was. Weird how that happens.
   15. John Northey Posted: January 28, 2013 at 03:28 PM (#4356736)
FYI: his second best figures from 30 on are 305/434 (less than 1/2 a season)/590. He outdid all 3 of those stats in each of his age 26-29 seasons. And this is trying to cherry pick the 2nd best possible post-30 results even using partial seasons. He was effective, just not HOF effective from 30 to the end.
   16. thetailor Posted: January 28, 2013 at 04:02 PM (#4356775)
re: 14 and 15, it's amazing how much can be done with statistics. I had a completely different impression of Frank Thomas' career trajectory than you did.

Yes, Thomas was an unstoppable force from ages 22-29, but to me his blip at 30 and 31 was very clearly temporary when he bounced back at age 32 to post a .328 average with 43 HR and 143 RBI. At that point you're looking at a guy who already had 344 career HR and a career OPS of 169, and you're thinking -- "okay, he's got plenty of time to get himself into the Hall."

Then, from ages 34 to 39, he posted a very respectable .265/.381/.523 with 165 home runs, which I don't think anyone views as a disappointing second act to a career, even with his injuries. He's hurt by having a power-outage in 1999 and missing 2001 altogether, but you could add his second act to a lot of players' careers and end up with a Hall worthy player.
   17. BDC Posted: January 28, 2013 at 04:09 PM (#4356793)
30 and on: 276/389/515 134 OPS+

Among players with >3000 PAs at age 30 and over, though, that OPS+ ranks 58th all time. There are quite a few HOVG players around that rank, but also in that vicinity are HOFers like Al Kaline, Billy Williams, George Brett, and Reggie Jackson. IOW, it's hardly unusual for a HOF hitter to be "only" that good post-age-30. (I realize you were mostly just remarking on how much better he was before 30!)
   18. John Northey Posted: January 28, 2013 at 04:12 PM (#4356799)
re 16: I saw it as disappointing. Here was an inner circle HOF'er it seemed, where he never had been below 174 for OPS+ - the most dangerous hitter year in/year out. Nothing seemed impossible when he had that bat in his hands. Then suddenly ... poof. He was a solid hitter still, but 4 years of 140+ OPS+, 5 in the 119-135 range plus 2 partials sub 100 was not what anyone expected from 30 on. That is production one expects from solid DH's, solid 1B. That is Fred McGriff territory (134 lifetime OPS+, peak of 165). Solid, near HOF level for a DH who can barely play 1B but not what we grew used to.

Was he still valuable? As I said, of course he was. Was he a MVP candidate in any season? One could argue it in 2000, maybe 2003 and 2008 (if you forget he was really a DH only by then) but pre 30 he was top 8 MVP every single season outside his 60 game rookie campaign (would've been then too if called up earlier). His stats were seen as video game level before the steroid mess hit. I think most forget just how amazing he seemed back then. It was shocking when he shifted from Superman level to Aquaman level - still great but not #1.
   19. Baldrick Posted: January 28, 2013 at 04:31 PM (#4356829)
As to the DH penalty in WAR -- it certainly does not accurately reflect decisions made by actual teams. From ages 23-32, Thomas was dinged for -177 runs between Rfield and Rpos or nearly 18 runs per season. Edgar never received more than -15 runs in a season via his DH penalty. We saw similar team decision-making with guys like Manny and Sheffield. Teams regularly seem more willing to stick somebody out there costing them 20 or more runs a year rather than shift them to DH where they would "cost" them just 15. It wouldn't be the first time that teams (in general) have made sub-optimal decisions but actual behavior suggests to me that the DH penalty should be more in the range of -20 to -25.

It's a positional slot that has to be filled if you are an AL team. It's a positional slot with the peculiar quirk that everyone who plays it just happens to be a league-average fielder. Therefore: you can slot anyone in it that you want and you'll get league-average fielding. So the only thing you need to consider is how much offense you can get from guys who aren't going to give you more value playing the field.

By that standard, the positional value of a below-average fielding first baseman isn't really much higher than a DH. They're very likely the same guy, in fact.

If a guy has a good enough bat, and you're living in the pre-DH world, you'll try and find a place for him in the field. But you have to be willing to account for the fact that his defense is COSTING you runs compared to other guys. His bat might be good enough to justify it, but it is a real, actual cost.

If a guy (like Thomas, perhaps) insists on playing the field when you could accumulate more value DHing him, you need to account for that. It might be worth it to keep him happy. Or it might be worth it because he actually hits better while playing the field (as seemed to be the case with Thomas). But you're making a real tradeoff.

My general feeling is that in the post-DH world, you should zero out all negative defensive contributions below a certain number. But there should be SOME room in between the standard DH and the actively-hurting-you-in-the-field guy who doesn't play DH.

To me, it's kind of the flip-side of the 'Edgar wasn't called up early enough' debate. Thomas may have been hurting his teams by playing poorly in the field. On one hand, you don't want to punish HIM for the mistakes of management. But on the other hand, he really DID produce that (negative) defensive value.

Now, Thomas was clearly the (much) better player. That goes without saying.

But I really find it hard to believe that given the available talent it's really meaningfully harder to find a replacement-level defensive first baseman than it is to find a DH who can hit as well.
   20. vivaelpujols Posted: January 28, 2013 at 04:40 PM (#4356848)
Yes he absolutely should. Add him to him to list of Piazza, Schilling, Bagwell, Bonds, Clemens. Woohoo!
   21. cercopithecus aethiops Posted: January 28, 2013 at 05:02 PM (#4356869)
But there should be SOME room in between the standard DH and the actively-hurting-you-in-the-field guy who doesn't play DH.


This assumes facts not in evidence. You simply cannot know exactly how much anyone who isn't actually playing the field would be hurting you if he was, and you also cannot know if a player who should DH but "insists" on playing the field really would hit substantially less well if you forced him to DH full time.

But for purposes such as comparing Frank Thomas to Edgar Martinez, there is a much simpler solution as put forth in post #12: oWAR. And like Walt says, it really isn't that close.
   22. TJ Posted: January 28, 2013 at 05:14 PM (#4356890)
"Now the steroid era starts in the 2000s?"

Nope- it actually started a lot earlier. I chose 1997 as the end point in this comp because 1998 saw McGwire and Sosa both top 60 HR, with McGwire hitting 70. To me, those are video game numbers...

I can understand people looking at Frank Thomas as they look at Dale Murphy- great peak, drastic decline, and therefor not HOF material. My view is that Thomas had a longer and higher peak, but his record became obscured by the PED-fueled video game numbers that came afterwards, and people forget that Thomas was truly a devestating offensive force for about 8 seasons. Would I vote for him? Yes- by my yardstick (a combiantion of career WAR, seasonal WARs of 8+/5+/3+, and dominance scores based on times leading the AL in certain categories and times placing in the top ten), Thomas would rank as a middle tier first baseman compared with current HOF first basemen, and thus merits enshrinement. Murphy would rank with the bottom tier of HOF center fielders, so I wouldn't vote for him.
   23. Misirlou was a Buddhist prodigy Posted: January 28, 2013 at 05:28 PM (#4356904)
I can understand people looking at Frank Thomas as they look at Dale Murphy- great peak, drastic decline, and therefor not HOF material. My view is that Thomas had a longer and higher peak,


And a much better decline phase. From 1998 (his first sub-par season, by his standards) through the end of his career, he had a 134 OPS+. That's only a little worse than what Murphy did in his prime.
   24. Mefisto Posted: January 28, 2013 at 05:45 PM (#4356918)
IMO, we don't penalize DHs enough when it comes to measuring career value. A DH is a very poor fielding 1B. Frank Thomas is the paradigm example of that. He should be about -25 every year; -18 is doing him a big favor.
   25. Baldrick Posted: January 28, 2013 at 06:38 PM (#4356947)
This assumes facts not in evidence. You simply cannot know exactly how much anyone who isn't actually playing the field would be hurting you if he was, and you also cannot know if a player who should DH but "insists" on playing the field really would hit substantially less well if you forced him to DH full time.

I am not saying 'punish' a guy because he insisted on playing or whatever. I'm just saying that he DID play and that play DID hurt his team more than if they had slotted him at DH. I'm hesitant to assign all of the 'blame' to the player, so I'm willing to set a minimum, but I am absolutely not willing to accept that a DH is incapable of providing more defensive value than a terrible-fielding first baseman. In fact, it is perfectly reasonable for that circumstance to happen fairly regularly.

If a guy is basically replacement level defensively, over his career that should balance out to being worth basically the same as a DH. But in any given year, it's just as likely to make him worth 'less' than the DH. In the case of someone like Thomas, who consistently was worth less than a DH, it's worth considering whether that indicates a real and meaningful drag on his value.

As I said, it clearly doesn't make him and Edgar equivalent - and if that's what the BB-Ref WAR says, then it might well need a little bit of tweaking. But the very best you can say in Thomas' favor is that he team simply played him incorrectly and he shouldn't be blamed. To me, that's a somewhat valid, but weak argument. So even in that case, there should be some room for giving a first baseman worse defensive ratings than a DH. The less favorable case is that he simply couldn't hit as well as a full-time DH. In which case, the team played him correctly but his defensive failings are absolutely part of his actual value - and deserve to be counted fully. Or, perhaps, he insisted on playing out there when the team would have preferred otherwise. In which case it seems like he 'deserves' the blame again.

The point being: while assuming those facts would give you CERTAINTY that the gap is deserved, you don't need to assume anything at all in order to believe that a first baseman can and should sometimes be worth less defensively than a DH. It may be the team's fault, but it really did happen.

If you are the sort of person who prefers to zero out negative-WAR years on the front or back end of a career, you will be more willing to edit out these problems, too. And I'm certainly sympathetic to that case. But that's also why I'm sympathetic to the argument that Edgar's short career and lack of defensive innings are compounded by a mistake made by the M's management. Which is to say: I am not at all opposed to using a more thick reading of a player's case - but that can cut a number of different ways.

And again, none of this makes Edgar as good as Thomas. I'm just affirming the basic premise of TFA that they're actually a bit closer than you might expect - in part because Edgar's additional time at DH simply isn't much of a detriment, when compared to first basemen like Thomas.

Thomas was the better hitter, no doubt. And by a pretty sizable margin. But I think you really can make the case that Edgar was a more valuable defensive player over his career, given his time as a solid third baseman and the fact that Thomas was pretty terrible at first base. Maybe not the 10 wins claimed by BB-Ref WAR. But I don't think it's at all unreasonable to think there would be a real difference there.
   26. cercopithecus aethiops Posted: January 28, 2013 at 06:54 PM (#4356956)
I can understand people looking at Frank Thomas as they look at Dale Murphy- great peak, drastic decline, and therefor not HOF material.


Really? I can't understand that at all.

If a guy is basically replacement level defensively, over his career that should balance out to being worth basically the same as a DH. But in any given year, it's just as likely to make him worth 'less' than the DH. In the case of someone like Thomas, who consistently was worth less than a DH, it's worth considering whether that indicates a real and meaningful drag on his value.


My take on this is slightly different -- if the numbers say that guys like Frank Thomas are "consistently worth less than a DH" then we need to stop assuming that guys who don't play the field are "basically replacement level defensively." Defense is only "a real and meaningful drag on his value" because we're giving some other players defensive credit that they aren't actually earning.

I think you really can make the case that Edgar was a more valuable defensive player over his career, given his time as a solid third baseman and the fact that Thomas was pretty terrible at first base. Maybe not the 10 wins claimed by BB-Ref WAR. But I don't think it's at all unreasonable to think there would be a real difference there.


I have no doubt that Edgar Martinez was a more skilled defender than Frank Thomas. I am reluctant to assign value to his defense when he was sitting on the bench while his team played defense. If Edgar was DHing while a weaker defender played the field because the team decided that Edgar would stay healthier that way, then how is Edgar not hurting his team on defense in exactly the same way that Frank hurt his teams by playing first base poorly while a better defender DHed?
   27. Baldrick Posted: January 28, 2013 at 07:15 PM (#4356975)
I have no doubt that Edgar Martinez was a more skilled defender than Frank Thomas. I am reluctant to assign value to his defense when he was sitting on the bench while his team played defense. If Edgar was DHing while a weaker defender played the field because the team decided that Edgar would stay healthier that way, then how is Edgar not hurting his team on defense in exactly the same way that Frank hurt his teams by playing first base poorly while a better defender DHed?

Again, DH is a defensive position. It is. You may not like that it is, but the team *actually* has to fill the position. You can't earn bonus defensive value at the position, like you can at others, but neither can you lose defensive value. That doesn't mean you have NO defensive value - it means your defensive value is simply about filling the position. That is: you provide the team the value of getting your bat into the lineup by filling one of the 9 available slots.

Meaning: you have to play a guy at Bench. And sometimes a guy will be more valuable playing Bench than he will be playing first base. Just like Frank Thomas was more valuable playing first base than he would have been playing shortstop.

And, just like I wouldn't give Frank Thomas credit for playing a TERRIBLE shortstop simply because his manager happened to put him there, I don't think we have to give him automatic credit for playing a bad first base simply because he was positioned there. You aren't supposed to play guys beyond their defensive capabilities, but if you do, they actually hurt your team.

Edgar didn't hurt his team on defense as DH because DHs all play league-average defense. But they ARE playing defense. It's just automatically league average.

Now, the whole point of a stat like WAR is to get away from the particular details of who is on a team and ask the more general question of how a guy compares to replacement level at his position. But if you want to drill down and look at the details, you'll find that the M's were playing John Olerud (a world-class defensive first baseman) and Tino Martinez (very good) and David Segui (solid). Okay, and Paul Sorrento (yikes).

But anyways, that's not really the point. To some extent, you have to give people credit (and blame) for what they actually did. And playing terrible first base is, when you come down to it, no better (and probably worse) than playing Bench.
   28. epoc Posted: January 28, 2013 at 08:26 PM (#4357018)
Again, DH is a defensive position. It is. You may not like that it is, but the team *actually* has to fill the position. You can't earn bonus defensive value at the position, like you can at others, but neither can you lose defensive value. That doesn't mean you have NO defensive value - it means your defensive value is simply about filling the position. That is: you provide the team the value of getting your bat into the lineup by filling one of the 9 available slots.


That last sentence doesn't fit. If DH is a defensive position that must be filled but at which whoever fills it cannot provide any defensive value either positive or negative, then it's not the player providing value by filling the position, it's the position itself that provides the value simply by existing. As such, Frank Thomas offered just as much DH value when he was playing 1B as Edgar did when he was playing DH. The value is simply that he was on the roster and thus an option to fulfill the inherent value of the role. Considering that, it seems clear to me that the positional adjustment for a DH should be the inverse of the replacement credit, since the replacement for DH is just the average player.

I understand that WAR (all incarnations) give DHs extra credit because it's supposedly harder to hit as a DH than as a position player. That makes sense, if you're willing to buy into the theoretical concept of positional adjustments at all. But I'd prefer not to bother myself. In my opinion, A DH's value is his offensive production above or below average, period. As such, a DH should be compared to a position player only with respect to their respective offensive performance.
   29. Baldrick Posted: January 28, 2013 at 08:48 PM (#4357032)
I understand that WAR (all incarnations) give DHs extra credit because it's supposedly harder to hit as a DH than as a position player. That makes sense, if you're willing to buy into the theoretical concept of positional adjustments at all. But I'd prefer not to bother myself. In my opinion, A DH's value is his offensive production above or below average, period. As such, a DH should be compared to a position player only with respect to their respective offensive performance.

No. This is not what WAR is doing.

Let me try and explain this a different way. There are 9 batting slots. The only way you can get into the game is by occupying one of those batting slots. For 8 of them, you stand on the field during your opponent's half-inning. For one of them, you sit in the dugout. But all 9 of them are real actual slots. And if the team wants your offense, they have to find a place to put you.

The DH is not simply a null spot. It's a real, actual position. If you can't find a guy who can hit well, you still have to play Jose Vidro or whatever. This is exactly the same as what happens at any of the other 8 slots.

For some of the slots, you have to measure not only offensive contributions compared to replacement level - you also have to measure defensive contributions. The way that WAR arranges all of this is to simply assess a replacement line for ALL offensive contributions and then to measure your defense compared to other guys who play your position - and give you a positional adjustment. The DH does the first and last of these - but is automatically average on the middle one.

This doesn't make it NOT a position. It just makes it a position which won't allow you to accumulate additional (positive or negative) value. But it doesn't change the basic structure of the calculation. That is: compared to all the other teams in the league who are getting X production from their DH slot in the lineup, what kind of value are YOU getting?

The DH 'advantage' compared to terrible defensive first basemen is NOT a product of it being harder to hit as a DH. It's because (given optimal roster construction and a perfectly efficient market) you shouldn't HAVE to play a terrible defensive first baseman. You are actively hurting yourself if you do so. Now, because it ISN'T a perfectly efficient market for all kinds of reasons, you will often find cases where the particular details justify it. Maybe your guy is terrible in the field, but hits better when he's playing so it's worth it. Maybe you've got Jack Cust and Edgar and Ortiz and Hafner all on your team and you think it's worth it to play the less-terrible of them in the field because their offense makes up for the defensive hit.

But you really ARE taking a defensive hit by playing the guy. Because it just ISN'T VERY HARD to find bad defensive first baseman who can hit well. In fact, it's just about as easy to find them as it is to find DHs who can hit. Because they're usually the same guys.
   30. cercopithecus aethiops Posted: January 28, 2013 at 08:54 PM (#4357035)
The only way you can get into the game is by occupying one of those batting slots.


Unless, you know, you pitch.

Because it just ISN'T VERY HARD to find bad defensive first baseman who can hit well. In fact, it's just about as easy to find them as it is to find DHs who can hit. Because they're usually the same guys.


And if you have two of them, the one who plays first base is not any less valuable to your team than the one who doesn't play the field. See? One side's whole point is also the other side's whole point.

EDIT: also, the fact that it isn't hard to find bad defensive first baseman who can hit well is all well and good, but the issue is why your DH should get the credit if you happen to find a good defensive first basemen who can hit well.
   31. Walt Davis Posted: January 28, 2013 at 09:02 PM (#4357041)
So the only thing you need to consider is how much offense you can get from guys who aren't going to give you more value playing the field.

This may be correct but isn't really how teams handle the position. There are almost no full-time DHs. Primarily DH is used as a slot where a part-timer or a recovering regular bats.

I've done this before but no reason not to again:

Full-time (min 502 PA) DH seasons, 90% of games at DH, median OPS+:

1992-2002: 128 (32 seasons)
2003-2012: 139 (22 seasons)

That's the sort of bat a team needs to see to carry a full-time DH. Those 22 seasons (in 10 years!) have 6 by Ortiz, 3 by Hafner and some of the end of Edgar, Thomas and Thome.

I worried about the 90% cutoff. I mean if you're playing upwards of 20 games a year in the field, your team thinks you still have some defensive capability even beyond interleague games. But then PH appearances count and maybe those are throwing it off. If you drop it to 80% you get 32 seasons. The median is odd -- #16 is Ortiz at 137, #17 is Matsui at 126. But the only repeat names you see below 137 are Matsui, the last 2 years of Vlad and "bad" seasons from Thomas, Edgar, Thome and Ortiz. Those were the last full-time seasons for Thomas and Edgar. You also see the near bitter end of Sheff and Damon's careers.

I think it's clear teams are not willing to carry an offense-only player unless he can absolutely rake. Otherwise they would rather carry an extra position player with a pretty mediocre bat and rotate players through the DH slot. This suggests that teams value sub-stud DHs about the same as a 4th OF.

As to Edgar, repeating the above to match his DH career (essentially 95-04), you get 39 seasons (80% DH) with the median season at 132 OPS+. For those years, Edgar's OPS+ was 153. If you do the same for AL 1B, you get 80 seasons with the median OPS+ at 123. From 95-04, Carlos Delgado had a 143 OPS+, 20 points higher than the "median full-time AL 1B." He gave up 112 runs to defense and position (5900 PA) while Edgar gave up 135 in 6000 PA. What does that mean? I dunno, I'll have to think about it more but Edgar as Delgado seems a better comp to me than Edgar as Thomas.

Anyway, to bring it back to Thomas and Martinez. Both HoF cases are based purely on hitting. Thomas provided a lot more value there. Both were poor defenders, lousy baserunners, etc. Thomas even had a substantially longer career. Yet, WAR erases most of that advantage through Rfield + Rpos, knocking the gap down to 5 wins. That does not seem "right" to me, especially not for HoF comparison purposes. By the way, by Rbat, I get Edgar's best 10 as 488 runs and 55 oWAR (includes 3 seasons at 3B); Thomas's as 562 and 61 oWAR ... yet again, WAR calls those 10 years equal. It's not impossible obviously but it just doesn't smell right.
   32. Srul Itza Posted: January 28, 2013 at 09:05 PM (#4357042)
I can understand people looking at Frank Thomas as they look at Dale Murphy- great peak, drastic decline, and therefor not HOF material.


Murphy didn't "decline" -- he fell off a cliff. Here are the year by year WAR totals for Murphy Thomas, with the "All Star" caliber years bolded (and split years between teams combined)

7.4, 6.8, 6.2, 5.8, 5.3, 4.7, 2.8, 2.3, 1.5, 1.4, 0.7, 0.5, 0.1, -0.1, -0.4, -0.7, -0.8, -0.9

7.0, 6.8, 6.7, 6.2, 5.9, 5.8, 5.3, 5.0, 3.9, 3.4, 2.8, 2.7, 2.3, 2.0, 1.8, 1.6, 0.4, 0.1, 0.0

Thomas retained value a lot longer, and the 0.0 was the year he only played 20 games, due to injury.
   33. Baldrick Posted: January 28, 2013 at 09:38 PM (#4357054)
Unless, you know, you pitch.

There are a lot of American League teams who bat the pitcher, are there?

And if you have two of them, the one who plays first base is not any less valuable to your team than the one who doesn't play the field. See? One side's whole point is also the other side's whole point.

Again, it depends what you are looking at.

If I play one of them at SS, then wouldn't you say that guy is less valuable? I mean, he certainly costs the team a bunch of runs.

If the only thing you are interested in is true talent, and you want to assume perfect allocation of resources, then of course it doesn't make sense to 'punish' a guy for playing in the field. But if you want to look at what actually happened, then it certainly DOES matter that one guy hurt the team while another guy did not. It may not be his 'fault' but it doesn't change the reality of what happened.

This is a debate with multiple reasonable sides. Do you assign war credit? Do you reward peak and disregard accumulation of career value? And so on.

All that I'm saying is that unless you are an absolutist in favor of war credit, in favor of zeroing out negative-WAR years, etc. then you have to acknowledge that negative-defensive value is indeed possible and ought to be considered in DH-valuation.
   34. Walt Davis Posted: January 28, 2013 at 10:21 PM (#4357076)
By the way, we may have seen a fundamental shift in DH usage. Number of 502+ PA, 90%+ DH seasons then 80% DH seasons and the median OPS+ for the 80%+ seasons by 5-year chunks

73-77: 15 16 109
78-82: 14 18 111
83-87: 17 22 116
88-92: 15 21 112
93-97: 17 22 122
98-02: 13 19 137
03-07: 15 18 143
08-12: 7 14 124

There were 0 90% qualifying seasons in 2012 thanks to Ortiz's injury and the Royals giving Butler 20 starts at 1B. Three of those seven were by Ortiz, the only repeater over that 5 year span. There is also a big offense jump from 98-07 but the last 5 years they've been back to 93-97 levels so maybe mostly just a sillyball effect.

Taking those endpoint years, here's the overall OPS+ for the DH split (i.e. the average DH, full-time or not):

1977: 105
1982: 108
1987: 103
1992: 107
1997: 107
2002: 108
2007: 112
2012: 107

So overall DH production has been pretty constant but the use of full-time DHs has been declining and the offensive threshold for full-time DHs has been going up. Also, over the last 5 years, there are 254 players (about 3.5 per team) with at least 10 games at DH.* In comparison, there were 32 qualifying 90% at 1B AL seasons and 215 10+ game seasons.

Anyway, DH has always been a mostly part-time position with only about 3-4 teams per year putting a full-timer there. Given this mish-mash provides more defensive flexibility and an OPS+ around 107 pretty consistently, you're gonna have to rake pretty good to be a full-timer with virtually no defensive value -- apparently around a 125-130 OPS+ for a true full-timer (90%+).

* Some of those guys might be pinch-runners but the only one I thought was a candidate was Ciriaco and it turns out he got 9 starts at DH. That tells you something right there.
   35. epoc Posted: January 28, 2013 at 11:40 PM (#4357114)
No. This is not what WAR is doing.


Yes. It is.

WAR creates a positional adjustment based on the difference in fielding performance at different positions by those who play multiple positions. DH is obviously exempt from that construction, because it doesn't involve playing defense. So instead, the DH adjustment is based on the fact that a replacement DH is just any player at all, i.e., average. But because research suggested that it's tougher to hit as a DH than as a position player, Tango decided to give extra credit for DHing. Instead of making the DH penalty the full inverse of the replacement credit, he set the negative value of DHing as less than the positive value of replacement credit. And all implementations of WAR that I'm aware of have followed his lead. That's exactly what WAR is doing. And it makes sense, insofar as you buy into all that theoretical hooha in the first place.

I'm not arguing that a poor defensive 1B costs the team runs. I'm aware that he is. What I'm saying is that the guy who DHs doesn't provide more value by DHing than the guy who plays 1B does by playing 1B poorly. Both players are capable of DHing, so it's the manager's discretion and the inherent value of the DH option that provides the discrepency in value, not the player himself.

Forget Thomas and Martinez for a second and just think about Thomas. Compare his 1996 with his 2000. According to rWAR, he created an identical amount of offensive runs, but in the former he played 1B poorly while in the latter he mostly DHed. So his overall RAR (and thus WAR) is greater in 2000 than in 1996. Now, did Thomas himself really provide more value in 2000, or was it just that his manager used him in a way which maximized value for the team? Should Thomas get credit for the difference in value?
   36. Walt Davis Posted: January 28, 2013 at 11:59 PM (#4357120)
And, again, what do we mean by "no defensive value"?

Thome, 26-33, -118 Rfield + Rpos (he managed to pull a +6 Rfield one year)
Thomas, 24-29, -108
Sheff, 94-01, -134
Sheff, 02-09, -97 (mix of OF and DH)
Manny, 96-03, -99
Manny, 04-11, -126 (3900 PA)
Griff, 99-08, -84 (not as bad as I'd have thought, his Rfield seems better than I remember)
Dunn, 25-28, -112 (and I skipped his epic year at 1B)
Bonilla, 28-35, -106
Giambi, 25-34, -132
Grieve, 22-26, -98
Belle, 27-33, -110
Cabrera, 08-11, -58
Baylor, 27-31, -97

Edgar, 32-41 (10 seasons), -134

That's an average of 13 runs a year, a better average than Thome, Thomas, Sheff I, Manny II, Dunn, Grieve, Belle, Cabrera, Baylor and Giambi; roughly equal to Sheff II, Manny I and Bonilla. Note, these aren't players having a couple of terrible years and their team moving them to DH in response, this is guys stinking year after year after year and still their teams chose to play them in the field (or not trade them to the AL where they would have been more "valuable" as DHs). Heck, the Tigers response to this was to move Cabrera to 3B so they could play Fielder (-57 from 09-11) in the field.

Here's an interesting question -- how often has an AL team traded for a crappy-fielding NL slugger to put him at DH. Thome springs to mind but only after he injured his back. Dunn was an FA signing but qualifies I guess (certainly NL teams showed little interest). I've already mentioned Prince. Cuddyer put up a -65 the previous three years and the Rox signed him to play RF. Meanwhile Kubel was putting up a -37 and the DBacks grabbed him to put up a -11. The O's continued to trot out Mark Reynolds (-41 the last 2 years) who they seemed to consider an equal or better fielder than Chris Davis (and Jim Thome).

How did the O's make the playoffs?????

Anyway, the "tipping point" seems to be somewhere around 15 up to maybe 20 runs before a team will shift the guy to DH, yet the DH penalty is set around -13 to -15 runs. Again, it's possible that is correct in value terms but it does not seem to reflect the actual decisions that teams make with regard to that position.

Finally ...

Edgar is just -122 for his career -- fine, he apparently was a decent-good 3B for a while which is much more than Thomas ever was defensively. But Thomas is at -257! Thomas had that 158 run advantage in Rbat and, when all is said and done, that advantage is completely wiped out and he has only 8 more RAA than Edgar (in 1200 more PA). Thomas's 7-year WAR peak -- y'know, back when he was the second coming of Jimmie Foxx -- is only 2 wins better than Edgar's. In his first 7 full years, Thomas never had an OPS+ below 174; Edgar had one season above 174. By WAA, these two guys are equal and Thomas's WAR advantage is purely due to 5 more wins of playing time.

Is there anybody here who believes that? There doesn't seem to be. So either we're all nuts or there's something a bit off about WAR and DHs (or possibly WAR over-penalizing atrocious fielders).

Then there's Sheffield. He's about 4-5 wins ahead on counting offensive WAR stats (Rbat + Rbase + Rdp) and another 4 wins ahead on replacement value. He is 17 wins behind on defensive value. 17 wins. That's 2012 Trout + Cabrera. Nick Swisher's career WAR. The difference between the 2012 Cubs and a 500 team. No f'ing way.

Then there's money. Edgar made $50 M in his career, never more than $7 in a season, usually much less. Over roughly those same years, Sheff never made less than $10 and he's made $170 for his career. Thomas made $7 in his final arb year and never really fell below that. Bonilla and Klesko had higher career earnings. McGriff, Galarraga and Burks all retired in 04 too and their career earnings are more than Edgar's (and Galarraga and Burks had similarly discontinuous careers). Burks is actually a very good salary comp and Galarraga out-earned Edgar from ages 37-40 and it looks like McGriff was paid about the same.

Now I'm willing to admit it's crazy that somehow Edgar was paid in line with McGriff, Galarraga, Burks, Bonilla and Klesko. Maybe he could have gone elsewhere for more money at the end of his career but wanted to stay in Seattle. But there's no evidence he was anywhere near as highly valued as Sheffield or Thomas, even the diminished Thomas or the journeyman Sheffield.
   37. epoc Posted: January 29, 2013 at 08:28 AM (#4357208)
Is there anybody here who believes that? There doesn't seem to be. So either we're all nuts or there's something a bit off about WAR and DHs (or possibly WAR over-penalizing atrocious fielders).


There is certainly something off about WAR and DHs. Look at Edgar's 1995-2004. Adding up his replacement runs and positional runs yields 6-7 runs per year. He's getting credit for 3/4 of a win per year because of a decision his manager made. Since literally anyone on his team could have been assigned that role and done exactly as well at it on "defense," the positional adjustment and the replacement credit should cancel each other out. (Unless of course you believe it's really 7 full runs harder to hit as a DH than as a position player, which is unlikely. The research I'm aware of suggests the difference is more like 2-3 runs, if you want to give it any credit at all, which I don't.)
   38. BDC Posted: January 29, 2013 at 09:16 AM (#4357243)
He's getting credit for 3/4 of a win per year because of a decision his manager made

That statement points to a key cross-purposes element of the debate, I think. Players continually get credit – in terms of post-facto value evaluations – for decisions their managers make. Move Cal Ripken Jr from 3B to SS, and his WAR goes up magically, even beyond his defensive contributions, just positionally. Move Darin Erstad from CF to 1B, and his WAR goes way down. They're still the same players with the same abilities, and they still have a present-moment trade value identical to the one they had before the decision was made. Yet they're contributing a much different amount of value before and after the decision. And so with the DH, it seems. Maybe Guy 1 can play the field well and Guy 2 could never do so; but at the moment, they're both DHs and not playing it at all. Their defensive abilities are different, but their defensive contribution is now identical.
   39. Ivan Grushenko of Hong Kong Posted: January 29, 2013 at 01:31 PM (#4357506)
WAR creates a positional adjustment based on the difference in fielding performance at different positions by those who play multiple positions. DH is obviously exempt from that construction, because it doesn't involve playing defense. So instead, the DH adjustment is based on the fact that a replacement DH is just any player at all, i.e., average. But because research suggested that it's tougher to hit as a DH than as a position player, Tango decided to give extra credit for DHing. Instead of making the DH penalty the full inverse of the replacement credit, he set the negative value of DHing as less than the positive value of replacement credit. And all implementations of WAR that I'm aware of have followed his lead. That's exactly what WAR is doing. And it makes sense, insofar as you buy into all that theoretical hooha in the first place.

I'm not arguing that a poor defensive 1B costs the team runs. I'm aware that he is. What I'm saying is that the guy who DHs doesn't provide more value by DHing than the guy who plays 1B does by playing 1B poorly. Both players are capable of DHing, so it's the manager's discretion and the inherent value of the DH option that provides the discrepency in value, not the player himself.


It sounds like the issue is how much worse of a hitter a player becomes when DHing. If Frank the DH is a worse hitter than Edgar the DH, but Frank 1B is a better hitter than Edgar DH, then it makes sense to put Frank at 1B, Edgar at DH, and penalize Frank for being a terrible 1B defender, even to the point where his defensive value is lower than Edgar's.
   40. madvillain Posted: January 29, 2013 at 02:28 PM (#4357574)
Frank's one of the greatest "pure" hitters (hit for average, hit for power, was incredibly disciplined and patient) to ever play the game. Probably one of the top 25 best hitters of all time. That a bad foot curtailed his production at 30 is a footnote to his career IMO. When he got healthy in his mid 30's he produced again, even if his bat speed wasn't there anymore and instead of 310/420/550 he was more like 260/360/520.

I'm biased having grown up with Frank as the best player on my favorite team, but Frank Thomas was the ####### man for about 4-5 years. He was "the most feared hitter in baseball" during that time and although that's just a stupid cliche it's also true in this case -- he was the best hitter in baseball, he was feared for good reason.

Frank had the natural frame of an OT in the NFL (Check the photo from his Topps Rookie Card). That he saw pencil thin guys like Bonds balloon up and pass him just as injuries started to rob him of his greatness must have really, really burned Frank.

Ray can bloviate all he wants about Frank and steroids but Frank is as clean as they come in baseball and he's a no doubt 1st ballot HOFer.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Harveys Wallbangers
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOT: Politics - December 2014: Baseball & Politics Collide in New Thriller
(5190 - 5:19pm, Dec 21)
Last: bobm

NewsblogJUNICHI TAZAWA & CULTURE OF DENIAL
(3 - 5:19pm, Dec 21)
Last: vortex of dissipation

NewsblogOT: Monthly NBA Thread - December 2014
(768 - 5:19pm, Dec 21)
Last: PASTE Thinks This Trout Kid Might Be OK (Zeth)

NewsblogMarty Noble's HOF Ballot
(33 - 5:15pm, Dec 21)
Last: NattyBoh

NewsblogThe 2015 HOF Ballot Collecting Gizmo!
(170 - 5:14pm, Dec 21)
Last: bobm

NewsblogOT: NFL/NHL thread
(9201 - 5:14pm, Dec 21)
Last: PASTE Thinks This Trout Kid Might Be OK (Zeth)

NewsblogDetermining Hall vote is no easy task | New York Post
(18 - 5:04pm, Dec 21)
Last: Booey

NewsblogThe Yankees’ plan in case A-Rod can’t play at all
(20 - 5:01pm, Dec 21)
Last: Win Big Stein's Money

NewsblogGetting ready to gamble on Jung-Ho Kang | FOX Sports
(6 - 4:13pm, Dec 21)
Last: smileyy

NewsblogRuben Amaro Jr. says it would be best if Phillies move on from Ryan Howard
(44 - 4:02pm, Dec 21)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogOT: Soccer December 2014
(338 - 3:50pm, Dec 21)
Last: Mefisto

NewsblogThe Jeff Jacobs HOF Ballot: Keep The Voting Serious And Fair
(47 - 3:23pm, Dec 21)
Last: cmd600

NewsblogSunday Notes: Low Strikes and Winter Deals | FanGraphs Baseball
(1 - 2:59pm, Dec 21)
Last: Walt Davis

Hall of MeritMost Meritorious Player: 1960 Ballot
(11 - 2:37pm, Dec 21)
Last: bjhanke

NewsblogA Salute to Sy Berger, From a Card-Carrying Fan - NYTimes.com
(4 - 2:25pm, Dec 21)
Last: eddieot

Page rendered in 0.6251 seconds
52 querie(s) executed