Are We Undervaluing the Slugger Now?
The crowd says that a baseball player who doesn’t occupy either center field or shortstop — and who has never hit 30 homers or stolen 30 bases — is about to get a $184 million contract. Our fearless leader thinks that contract is going to be the best value on the market this year. There are plenty of reasons to agree, not the least of which is that past big contracts have been at their best when given to young, athletic players with defensive value. Of course Jason Heyward checks all those boxes.
On the other hand, it’ll be a departure. Carl Crawford is the only other $100 million man who’s played something else besides an up-the-middle defensive position while also recording an isolated slugging percentage under .200. The sport usually gives nine-figure deals to players who slug or play great defense at a premium spot.
And while Heyward might play center field on his new team, there’s still evidence that we don’t value sluggers as much these days. Look at the crowd’s projection for Chris Davis, coming off a 41-homer season — it’s more than $80 million less than the one for Heyward. Mike Trout, Kyle Seager, Evan Longoria, and Elvis Andrus represent more than half of the $100 million contracts that started in 2015. None is your typical Big Bat.
So. Have we gone too far? Are we undervaluing the slow-footed, no-defense slugger? There’s other evidence we are.
Jeff Zimmerman recently re-did the positional adjustments in WAR to adjust for the current state of the game. He found something interesting: not only was the spread smaller, but the penalties we are giving first basemen and designated hitters might be too harsh. Look at our current adjustments next to Zimmerman’s findings this year.
| Current | Zimmerman | |
| Catcher | +12.5 | +7.75 |
| Shortstop | +7.5 | +4.75 |
| Second Base | +2.5 | +1.75 |
| Third Base | +2.5 | +1.75 |
| Center Field | +2.5 | +1.75 |
| Left Field | -7.5 | -4.25 |
| Right Field | -7.5 | -4.25 |
| First Base | -12.5 | -9.25 |
| Designated Hitter | -17.5 | -9.25 |
Perhaps this chart has something to do with the change. There are fewer and fewer balls in play with every coming year.
With fewer balls in play, the spread between a great defender and a lesser one should be smaller. And the same should be true of defensive positional value.
Using these values, I re-ran this year’s Wins Above Replacement leaderboard for qualified batters only. It’s crude because it hasn’t been zeroed out or adjusted baseball-wide, but it can give us an idea of how the very best might look a little different if your defensive values were a little different.
The top of the leaderboard isn’t changed enough to affect the MVP races, or the top five players in baseball, really, but look at the most changed players. It’s obvious who they are, but it’s also instructive to see the amount of change.
| Name | Fielding | Positional | New Positional | Offense | Defense | WAR | newWAR | diffWAR |
| Prince Fielder | -3.7 | -16.3 | -8.9 | 12.4 | -20 | 1.6 | 2.4 | 0.80 |
| Evan Gattis | -0.9 | -15.1 | -8.0 | -3.8 | -16 | 0 | 0.8 | 0.77 |
| Billy Butler | 0.9 | -15.2 | -8.1 | -12.1 | -14.3 | -0.7 | 0.1 | 0.75 |
| David Ortiz | -0.2 | -15 | -8.0 | 21.6 | -15.2 | 2.8 | 3.5 | 0.75 |
| Kendrys Morales | -0.4 | -15.9 | -8.5 | 14.7 | -16.2 | 2.1 | 2.8 | 0.74 |
| Alex Rodriguez | -0.4 | -14.7 | -7.8 | 20 | -15.2 | 2.7 | 3.4 | 0.71 |
| Nelson Cruz | -6.2 | -11.4 | -6.2 | 41.5 | -17.6 | 4.8 | 5.4 | 0.57 |
| Edwin Encarnacion | 1.6 | -13.3 | -7.9 | 33.6 | -11.7 | 4.5 | 5.1 | 0.57 |
| Albert Pujols | 3.4 | -13.6 | -8.6 | 7.3 | -10.2 | 2 | 2.5 | 0.52 |
| Jose Abreu | -1.8 | -12.9 | -8.7 | 21.1 | -14.8 | 3 | 3.4 | 0.45 |
| Chris Davis | 6.3 | -11.8 | -8.0 | 36.3 | -5.5 | 5.6 | 6.0 | 0.40 |
| Avisail Garcia | -6.2 | -7.6 | -4.2 | -16.1 | -13.8 | -1.1 | -0.7 | 0.38 |
| Jose Bautista | -9.9 | -8.9 | -4.9 | 38.9 | -18.9 | 4.5 | 4.9 | 0.37 |
| Joe Mauer | 0.7 | -12.3 | -8.7 | -7.6 | -11.7 | 0.3 | 0.7 | 0.37 |
| J.D. Martinez | 7.7 | -7.8 | -4.4 | 25.7 | -0.1 | 5 | 5.4 | 0.37 |
| Kole Calhoun | 13.8 | -7.1 | -4.0 | 6.9 | 6.6 | 3.8 | 4.2 | 0.36 |
| Carlos Santana | 3.9 | -12.2 | -8.5 | 8.7 | -8.3 | 2.4 | 2.8 | 0.35 |
| Matt Kemp | -17.2 | -6.9 | -3.9 | 8.1 | -24.1 | 0.4 | 0.8 | 0.35 |
| Melky Cabrera | -7.4 | -7.7 | -4.3 | -10.4 | -15 | -0.3 | 0.0 | 0.34 |
| Mark Trumbo | -1.8 | -8 | -4.6 | 2.4 | -9.7 | 1.1 | 1.4 | 0.34 |
Prince Fielder gains almost a win. Evan Gattis goes from replacement level to worth something. Billy Butler finds replacement level.
And further on down, Chris Davis narrows the gap between himself (6.0) and Jason Heyward (6.3). Heyward’s still younger, and he’s still more valuable, but every bit of value is important here.
If position players generally lose about a half-win per year due to age-related decline, starting higher means ending higher. Looking past Davis’s Steamer projection, which he’s bested by triple in two of the last three years, that would mean potentially more than 22 wins over the next five years, worth even more than Dave Cameron’s $130 million estimate, most likely.
This sort of thing might help explain some of our recent findings that first basemen and sluggers have been overpaid. If the market continues to act differently, perhaps the market knows something we don’t. This could be part of catching up.
Aging is another part of the story, though. As Craig Edwards showed by looking at comps, and this aging curve for isolated slugging showed, and this aging curve for batted ball distance showed, power does not age well. If you’re still not convinced, here’s an aging curve for home runs per fly ball that Zimmerman just ran for us (a .02 change would take an 18% HR/FB number down to 16%).
Aging for first basemen and DHs is worse than for other positions. Between the ages of 29 and 34, first basemen lose 3.1 WAR according to research done by Jeff Zimmerman. That’s 25% worse than the population, and it takes a good deal back from any value they might get from changes in the defensive adjustment spectrum. Applied back to Chris Davis, for example, the first baseman aging curve, over a five-year contract, would cost him more than a win over the regular aging curve. He didn’t gain half of a win in value last year from the new spectrum.
So it’s a conundrum. Perhaps, looking backwards, we’ve stolen bits of value from sluggers at designated hitter and first base. You could add another ten wins or so to David Ortiz’s career value, for example, and it would help his Hall of Fame case, which looks good if you focus only on his bat, but traditionally suffers by WAR.
But going forward, which is what the market is most worried about, we probably haven’t been too far off when we call the Davises and Ortizes iffy bets. Power doesn’t age well, and those positions are all about the power.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
Are these new numbers going to be incorporated or just talked about?
Dunno yet. He did the work in-season and the offseason is traditionally when we change these types of things.
I hadn’t seen Zimmerman’s new positional adjustments. Glad to see that SOMEONE is at least going to back to look at Tango’s pull-out-of-his-ass approach. So many other Tango-created problems to correct, though.
Just because the positional adjustments have changed doesn’t mean Tango was wrong a decade ago. The game has changed considerably in the last 10 years.
Then again, the old adjustments could be garbage. The free agent market rates for 1st and DH would have me believe they were.
This seems completely unnecessary. His approach was certainly not “pull out of his ass”. Go read some of his research.
This is absurd, and I suppose I’m the idiot for taking the bait.
[1] When Tango publishes something he completely explains his methods, evidence, and rationale.
[2] Sometimes at his site, and on a less important issue, Tango will “pull numbers out of his ass” (his well educated and researched ass) or what could be called “back of the napkin calculations” in order to illustrate a point. In other words, the actual numbers are less important than the process using the numbers … such as how a math teacher might do when teaching functions in algebra.
Maybe it’s just me, but I haven’t observed situations where Tango just half-asses something with little to no regard for quality or accuracy.
Most of the discussions at his site seem to be about methodology, and I say that because I am unable to follow most of those discussions.
Phew! Much love Eno.
The idea that Heyward would get more than Davis is insane to me. Davis could break the home run record. His value isn’t based on some meaningless, made up stat like UZR, which is absolutely unreliable. Heyward is a WAR Supastar but not a true supastar. i feel like baseball is going backward in its analysis. What happened to Billy Beane wanting guys that could get on base and smack the ball over the fence?
Heyward probably will get more than Davis, because they’ve got similar overall offensive numbers over their careers, and Heyward’s three years younger and more athletic (which means he’ll likely age better).
Heyward will likely get a lot less than everyone on FG is expecting, though, because the market values dWAR less than it does oWAR, since it’s much more difficult to find a Kendrys Morales/Prince Fielder than an Austin Jackson/Alcides Escobar.
What homerun record?
Yeah that’s a bit much.
You can’t break the home run record striking out as much as he does.
You need every one of those cheap off the end of the bat jobs to get to 74 dude.
On topic, i do think the adjustments were off from the start and the contracts going to middle of the order bats demonstrates that.
But the warning signs on Davis are less around his defense than his offense.
His k rate is so dangerous. Any loss of bat speed makes him unplayable.
I think he’ll have several excellent years but the margin is small and his bad years will be disasters.
Heyward still had considerable offensive upside (every team bidding will be banking on getting the three year run when it all clicks and he hits 35 hrs) and his downside is so low because his speed and defense and walk rates mean at worst he’ll be a decent starter.
I think this will be one of the best long term contracts given to a free agent in ages.
Ok, UZR (or DRS) is hardly meaningless or made up. You may understandably disagree with some of the exact figures, but have you ever read about how these defensive stats are calculated? The process is quite meticulous and well contrived. There may be significant year to year variation, but these stats are hardly garbage, and I think the idea of valuing defense in a range of a couple wins a year is sound.
I’m not sure if you meant for your attack on defensive metrics to come off as so severe, but I’m sure the baseball fans who have worked their tails off to learn how to measure defense would have a bone to pick with you.
What would this do to Edgar’s WAR?
Well if we’re going to add another 10 wins to Big Papi’s total, Edgar would be at 75 or so wins. That would bolster his case a bit, although I’m sure people would still go “But he’s just a DH!”
Be careful applying positional adjustments based on the 2015 state of the game retroactively to the late 90’s and early 00’s. Was a totally different game then, so I imagine Tango’s original positional adjustments (which were developed closer to that period) would be much more accurate to apply to Edgar.
(And which should be PLENTY to get Edgar sailing into the Hall, but that’s a different tangent for another day.)
Very interesting – the long term trend in more strikeouts and lower offense should lead to a WAR adjustment every 5 years or so…
I do think the distribution of “power” in the post “roads” era is still not fully reflected – but that may take more time to sort out..
If I was the GM I would want Davis over Hayward for the same dollars – but that is because I love the long ball!!!!
If you were to adjust all players’ WAR using the new positional adjustments, would that make WAR line up better with actual wins? Should be easiest way to see if it’s accurate, no?
by this I mean team WAR and team wins
Positional adjustments don’t really change how they affect team wins.
Remember all teams have pretty much the same amount of innings at each position. Changing positional adjustments will just make every team’s 1B more valuable and every team’s middle infielder less value, etc.
It just changes the dynamic between a given team’s first basemen and shortstop, for example.
That may be partly the case, based on the table showing the new values. But the article began with a graph showing that BIP have steadily decreased. Eno commented that that should reduce the spread in defensive value between players, but in fact it should also reduce the defensive value of all the players.
BIP is reduced by an increase in (HR + K. Since K increases, this means that a decrease in BIP results in an increase in the value of pitchers; a greater proportion of the outs recorded are by them, rather than by a defensive player. This is where the decreased value of defensive players goes, not necessarily from one position like SS to another like 1B, but from all of them to the pitcher.
One might argue that since BIP is also reduced by more HR, that a decrease in BIP does not necessarily mean greater pitcher value. But if more HR are given up, runs per game increases, and runs per win increases. In other words, the greater number of runs given up doesn’t affect the pitcher’s value, because his own team is also scoring more runs.
If this is not clear, consider the extreme case where there are no BIP. All runs are scored by HR (some of them with men on base from walking), and all outs result from Ks. No defensive player has any value at all, in fact, in that extreme situation, defensive players wouldn’t even be needed. All the value would go to the pitcher, with the best pitchers giving up the fewest HR and walks.
This reminds me of the gambler/stock broker that only starts counting after the big losing streak. If you want accurate positional adjustment you have to look at the whole cycle, which goes back to at least 1993. The game hasn’t changed**. We are just at a down point in the run scoring cycle.
**Obviously there will be less BIP which will put less pressure on the defense and this needs to be accounted for. It just doesn’t work with only a 15 year sample.
re: Heyward, his offense is worth 20 runs, playing time another 22. So he’s at 42 runs plus whatever you think of his defense…which I don’t think anybody would argue isn’t above average. Even if you only think he’s +10 relative to the old -7 average RF, you’re still looking at a 4.5 WAR player. That’s going to cost about $36m in 2016. If I’m Heyward, I’m not giving up my age 26-29 for any less than that, and by age 30 price inflation will still probably be more than his talent has declined. He should get $175m just for the first five years of the deal. If his agent suggests he take 9/$195, his agent should be put in jail. IMO, there is no way he takes a 9 year deal though. It’s 10+ or 5 or less.
You don’t think the game is considerably, categorically different than what we saw from like 1995-2002? I don’t disagree that offense levels ebb and flow over time, but I do think the environment of the game is fundamentally different than what we saw then, not just a slight change.
Definitely, Ks are up and they aren’t going anywhere. But other than that? No, the game isn’t really different. The way you score runs is by getting people on and knocking them in. The way you prevent runs is by preventing people from getting on and knocking them in. It’s really that simple, even though the balance of power has shifted extreme in one direction.
Now, imagine having this conversation in 1999/2000. Was the game different back then? I’d argue no, other than it was at the peak of offense rather than the valley.
Interesting read. For what it’s worth, the magnitude of the negative adjustments have always bothered me at a general level. I’m just not familiar enough with the underlying math to have ever made a more specific quantitative observation about it. After all, it’s a lot easier to have a general feel that it’s too much than it is to arrive at a better figure.
Loved the article…reminds me how many circles we run in just to prove that at the end of the day, it’s all just baseball.
Pardon my nitpicking, but when you state “Mike Trout, Kyle Seager, Evan Longoria, and Elvis Andrus…None is your typical big bat”, I’m not sure what you mean by “big bat”. Obviously nothing about Trout is typical, but Kyle Seager is a corner infielder (known as a power position) with 20+ HR/season going back the last four years. Granted he isn’t hitting 35+/season, but this isn’t the early 2000s anymore, and getting that production out of EVERYONE is unrealistic. With that being said, I don’t know why Kyle Seager doesn’t qualify as a big bat in today’s game. Hell if you put him in Coors instead of SafeCo maybe he’s putting up Arenado-esque numbers (of course that’s purely speculation).
I think by “typical big bat” he means a player who derives their value almost exclusively from hitting, and hitting dingerz. Like Cabrera, Nelson Cruz, Encarnacion, Ortiz, etc.
Seager is a good power hitter, but he’s more than that.
So 1B/DH is basically implied
I’m curious how a DH that doesn’t play the field could have the same positional adjustment as a 1B who DOES play the field? According to this, you could have your 1Bmen sit in the dugout instead of stand at the bag and it makes no difference in his value?
Am I missing something?
No, that’s correct.
Wow. Ouch yeah that cannot be.
I’ve said for awhile that a huge portion of a good first baseman’s value isn’t counted.
Scoops can happen several times a game often in huge circumstances. And yet they don’t get credit for it.
The adjustment cannot be the same. It’s broken if that’s what you need to make it work.
And that can be easily proved. Hoping that’s a typo.
Except that there is an actual defensive WAR value added to the 1B that the DH doesn’t get.
I don’t get it either. The logic apparently is that the DH Penalty offsets the fact that the 1B plays the field, but I’m not seeing how a below average 1B is a worse defensive contributor than a DH.
I think the assumption is that if the DH could field better than the first basemen (especially if it was a below average fielding 1B), then the DH would be actually playing 1B.
And that hasn’t changed. Acknowledging the DH penalty is merely pointing out that there is more than playing DH than NOT fielding.
(Consider the following hypothetical: you have 2 1B/DH types who are roughly equal defenders at 1B. However, one of those players is a “skilled DH”–the Martinez/Ortiz type who thrives while hitting as a DH–and one of those players is not, suffering a severe penalty. The manager, knowing this, starts the skilled DH at DH and the poor DH at 1B. The current model of WAR, however, would severely penalize the DH for his assumed lack of defense, when in reality the position was being determined because the DH had a skill that the guy playing 1B lacked. Balancing the positional adjustments between 1B and DH merely acknowledges the fact that DHing is a skill, and that more goes into being a good DH than just being the worst glove among the top nine hitters on the team.)
Jason Heyward is actually not very good. OPS under .800 is not a star corner outfielder. Just let him go.
Ok
I think its ironic that you so heavily use David Ortiz as a comparison since he has just refused to stop hitting. Pick a guy that doesn’t seem to age as a point for how much sluggers age is not the best tactic for proving your point.
Is Ball In Play% the percent of PA that result in BIP? If so why should we want to include BB and HBP in the denominator?
Blue Jays have a couple tough decisions in 2016 with Encarnacion and Bautista.