Let’s Explain Eric Hosmer’s WAR

Eric Hosmer is still sitting on one home run as we approach the one-fourth point of the regular season. He hasn’t hit fewer than 14 dingers yet in a year, so it’s clear that Hosmer’s still looking for his power stroke. But don’t make the mistake of believing that Hosmer has yet to make a positive contribution — he’s got a 120 wRC+, which is basically right on what he did a year ago. Let’s keep doing that, comparing last year to this year. The offense has been identical, overall. Hosmer last year was nine runs below average defensively. Hosmer this year is on pace for about -7. So in a sense, Eric Hosmer has been just as good a player. But, last season, Hosmer was worth just over three wins. This season, he’s on pace to be worth just over one win. How do you explain that, when a guy’s been hitting the same and fielding the same? Is WAR losing the mind that it doesn’t have?

That’s one option. Or you could look at WAR’s other, oft-forgotten input. You think about baserunning value when it comes to burners like Billy Hamilton and Jacoby Ellsbury. It’s easy to kind of forget about it when you’re dealing with a first baseman or a DH. But, to this point, according to our leaderboards, Hosmer has been the worst baserunner in baseball, at almost five runs below average, already. That puts him on pace for -21, eclipsing Kendrys Morales‘ recent record of -14 in 2009. Hosmer, presumably, won’t keep up this impossible pace. Previously, for his career, he was actually above average. But how did things get to this point? How has Hosmer already cost his team that many runs in such a small sample of games?

There are .gifs, and, unfortunately, they are big.

The first thing that comes to mind: caught steals. People think of stolen bases as a huge part of baserunning value, yet so far, Hosmer is just 0-for-1 in steal attempts. Before the year, he was 38-for-48, so he’s a first baseman with some wheels. That makes it all the more surprising that he ranks where he does for the moment. But there are an awful lot of ways to do something good or bad or in between on the bases, and Hosmer’s wound up on too many bad ends through a month and a half.

Excluding steals, he’s already made seven outs on the bases, one shy of last year’s total and higher than his 2011 and 2012 totals. He’s run into four outs at home. This year, only Yasiel Puig has also made seven outs on the bases, and nobody else has made as many outs at home. Hosmer’s been thrown out leaving first on a single, he’s been thrown out leaving first on a double, and he’s been thrown out twice leaving second on singles. It hasn’t been Eric Hosmer’s year, and three times, he’s made outs on the bases in consecutive games.

So how does this happen? And perhaps more importantly, what does this tell us about Hosmer, versus what does this tell us about the nature of good and bad luck? How reflective is this of Hosmer’s true talent? We’re going to quickly go through the plays in chronological order. Following will be the eight Hosmer outs on the bases, caught-steal included.

#1: April 2

Hosmer1.gif.opt

Welp. Nurts. Absolutely nothing for Hosmer to do. The numbers don’t know how to interpret a line-drive double play, and we don’t have live game-to-game stringers.

#2: April 5

Hosmer2_1.gif.opt

Hosmer2_2.gif.opt

The problem here? Part Hosmer, part not-Hosmer. From the recap:

Gordon dropped a single into right field, just in front of diving right fielder Dayan Viciedo.

But Hosmer feared a catch, hesitated between second and third and then had what he called a miscommunication with the team’s new third-base coach, Dale Sveum. Rounding third base, Hosmer got hung up and was tagged out in a rundown.

“My first peek, I saw his hand and saw him point so I figured the ball got away from somebody or something like that and I just saw the hand and went,” Hosmer said. “It wasn’t the right thing to do.”

Hosmer gets penalized for being over-aggressive, but at least some of the responsibility falls on the third-base coach, who I guess wasn’t completely clear.

#3: April 6

Hosmer3.gif.opt

Runners last year were 19-for-21 against Chris Sale. Runners this year are 0-for-1. I’m not a huge fan of running in a 3-and-1 count against a good hitter, but perhaps Hosmer figured Billy Butler would get a pitch to drive, and then Hosmer could come all the way around to snap the deadlock given a head start.

#4: April 17

Hosmer4.gif.opt

A good relay from deep center and Hosmer was just barely out. That’s something to remember about plays like this: they’re frequently decided by a small fraction of a fraction of a second, and the decision to go or not has to be made with the ball still quite far away. The Royals were leading comfortably against a team they were playing comfortably, so in a sense Hosmer didn’t need to risk it, but in another sense he didn’t have a lot to lose since the Royals’ win expectancy was already so high. The outcome of this was a coin flip, and it required the Astros to do a lot right, after the first mistake by Dexter Fowler.

#5: April 18

Hosmer5.gif.opt

Too much aggressiveness, here. Hosmer might’ve misread the height of the throw in to home plate. Or he might’ve forgotten about the pitcher. Or he might’ve just plain old done something dumb. But that’s an out that doesn’t need to be made. That extra base is not worth a ton.

#6: April 30

Hosmer6.gif.opt

Can’t help the contact play. The contact play is out of the runner’s hands.

#7: May 1

Hosmer7.gif.opt

Hosmer was going full speed and he certainly didn’t misread his third-base coach. He just made an attempt and got thrown out after a strong throw that yielded a good hop and got the catcher in good position to apply an easy tag. Hosmer was out by literally an inch or two, and plays like this are like one-run wins: you figure they’ll balance out in the long run because so much just comes down to luck. If the outfielder grips the baseball a little differently, Hosmer might be safe. If the catcher does anything, almost genuinely anything differently, Hosmer might be safe. He was out, but not in a way that really reflected a mistake on his own part. Wasn’t about running, or speed — this was just about the fact that nobody is safe all of the time, no matter what. Defenses just make plays.

#8: May 7

Hosmer8.gif.opt

And here’s something stupid. This is flat-out bad baserunning.

—–

Eric Hosmer has done some things to earn his low baserunning value. He’s also been a victim of some bad luck and circumstances, and if all these plays were to repeat, Hosmer might not make outs on all of them. On a couple, there was nothing for him to do. In the eight games in which Hosmer has made an out on the bases, the Royals have gone 5-3, so it’s not like this has been particularly crippling, so we’re all left just considering what this means. Seems to me there’s a lot of noise in the baserunning values, but there’s a lot of noise in all the data, and it’s not like Hosmer didn’t make all the outs. But you end up in a situation where you’re deciding whether you want to look at a number descriptively or predictively. Descriptively, Hosmer has made all these outs, and they’ve cost runs. But predictively, Hosmer now has a career BsR of about exactly league-average. Which seems about right to me. Overall, Eric Hosmer is roughly a league-average baserunner, and over a month and a half, a league-average baserunner can, statistically, look like the worst runner in the majors.



Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.


Sort by:   newest | oldest | most voted
tz
Guest
tz
2 years 18 days ago

Dale Sveum is the KC 3rd base coach.!!!????!!!!

This Red Sox fan knows that’s about a 2-3 WAR loss on the basepaths.

Only in KC….

Simon
Guest
Simon
2 years 18 days ago

Does anyone keep track of 1st- and 3rd-base coach BsR numbers?

AC of DC
Guest
AC of DC
2 years 18 days ago

Perhaps if he could get that uniform in “Sexy Antifreeze” . . .

Johan Santa
Guest
2 years 18 days ago

I totally came down here to post something like that too. 2004-05 were brutal with guys getting gunned down at home.

juan pierres mustache
Member
Member
2 years 18 days ago

Dale Sveum’s baserunning decisions are those of a man who believes his team will never make contact with a baseball again.

Big Jgke
Member
Big Jgke
2 years 18 days ago

I feel like this could be the inspiration for a pretty hilarious Dale Sveum commencement speech.

libre
Member
Member
libre
2 years 18 days ago

If you have trouble with gifs that are too large, webms are a good alternative in that they are an alternative that is better than gifs for animated images since it was designed for that.

http://www.webmproject.org/

Ned Yost
Guest
Ned Yost
2 years 18 days ago

No time to practice running the bases when you have to focus on bunting practice.

ben
Guest
ben
2 years 17 days ago

HAAAAA!!! Ned, what’s it like to be the smartest guy in the room at all times?

Bip
Member
Member
Bip
2 years 18 days ago

It’s interesting that although Puig has run into as many outs on the bases, his baserunning value is barely negative. Whereas Hosmer appears to be unlucky, Puig is legitimately overaggressive. However, while there is no upside to being unlucky, the upside to being overaggressive is that sometimes you successfully take a base that no one would ever expect you to take. Apparently it’s been a more or less equal trade-off with Puig.

I also just noticed that Puig is slashing almost the exact same line he had last year.

Stan Gable
Guest
Stan Gable
2 years 17 days ago

I noticed that Yasiel Puig is pretty good at baseball.

Bob
Guest
Bob
2 years 18 days ago

A faster baserunner avoids a few of those outs. Hosmer’s getting older, he’s probably slower than he was last year, so he’ll need to be smarter on the base paths.

BTW, can you get -4 runs on one play? If you hit a grand slam with two outs, but pass a baserunner before the lead runner touches home, do you get credited with -4 baserunning runs?

Steven
Guest
Steven
2 years 18 days ago

Hosmer went from being 23 to 24, and he didn’t gain any weight. He’s not slower than he was last year.

Schuxu
Guest
Schuxu
2 years 18 days ago

Yeah, he is not in the getting older part where it correlates with getting worse.

Value arb
Guest
2 years 18 days ago

It is if he’s leaving his legs in Baseball Annie’s bedrooms.

Jason B
Guest
Jason B
2 years 18 days ago

I wonder at what age speed does peak? I’m legitimately curious. I would guess early 20’s, maybe very late teens? There’s probably a very slight, probably imperceptible change between average speed at 23 and average speed at 24 for most people, but I don’t know if it’s for the better or for the worse, or if you tend to just plateau for a little while?

GreyStreet
Guest
GreyStreet
2 years 18 days ago

Most athletes, including their speed, peak around age 26-29. Most sprinting records you’ll see occur with an athlete in this range.

Jason B
Guest
Jason B
2 years 18 days ago

Neat! Thanks for the response. Now that you mention it, Olympic sprinters do tend to peak a little older than I would have previously expected.

Matthew Murphy
Member
2 years 18 days ago

Just sticking with sprinters, Usain Bolt set the world records in the 100m/200m at the 2008 Olympics as a 21-year old (just a few weeks before he turned 22). He set new world records a year later at the world championships, and hasn’t matched those marks since (although he came close as a 25-year old in the 2012 olympics).

Asafa Powell, who held the 100m world record before Bolt, set it at the age of 25 in 2008 (and has yet to match it since). His best 200m time came as a 23-year old.

Just two examples, but I would guess that raw speed peaks in the 21-25 year range. Some baseball players may improve their baserunning/stealing even as their speed declines, however.

bob
Guest
bob
2 years 18 days ago

The Royals have a worse base running problem. Billy Butler was almost thrown out going from first to third on a double, in a situation where he was running on contact, and the batter got to second before he got to third. The Royals have recently moved him from cleanup to third in the batting order. Why? Because his lack of speed and lack of hitting hurt them less when he bats third? And what about Moustakas? Moose is valuable for defense but the Royals could let the pitcher bat for himself and use Butler as DH in the eighth spot for Moose. Half joking here, but really, what’s up with this team?

Jason
Guest
Jason
2 years 18 days ago

It was put together by a GM with a fundamental misunderstanding of the game.

Jason B
Guest
Jason B
2 years 18 days ago

To be fair, I think all of the dudes in question (Butler, Hosmer, Moose) were all pretty highly regarded and most any team would have been glad to get them back when they were drafted (which may speak to a player development issue more so than a talent identification issue).

Bill
Guest
Bill
2 years 18 days ago

Butler has had a nice career, this year notwithstanding, and Hosmer is George Brett, but he’s been decent. I don’t like GMDM, but it’s hard to fault how he’s handled Moose Tacos. He was supposed to be good and has been terrible. He hasn’t even shown signs of being good. I think Moore is right, however, to give him every opportunity. Patience paid off with Hosmer, maybe it will pay off here too. But, as he’s done in the past, Moore did not count on Moose Tacos getting it together. He made a very nice move to sign Valencia. I can’t find fault with how Moore has handled things. I suspect that the Royals player development people suck and this is Moore’s fault, but the player moves are good. The Royals had a stocked farm system with can’t miss prospects and the only one to live up to his potential is the one the trade away? This seems improbable.

Nate
Guest
Nate
2 years 18 days ago

Things like this is why the various run totals need to come with error bars.

diegosanchez
Guest
diegosanchez
2 years 18 days ago

I agree. When you’re getting docked points for overall value because of plays where no lack of baserunning skill is involved, there is error in calculation. For this reason I will never take fielding or baserunning value totals half as seriously as hitting values.

indyralph
Member
Member
indyralph
2 years 18 days ago

If only there were projections systems that could regress small sample data sets to the mean.

Anon21
Member
Anon21
2 years 18 days ago

When you’re getting docked points for overall value because of plays where no lack of baserunning skill is involved, there is error in calculation.

Sorry, but you’re just wrong: there is no error in calculation. As Jeff says, Hosmer really did make those outs, and BSR is not trying to answer the question “Would a different runner have also made this out?”, nor should it. If a guy puts 650 balls in play in a season and, grossly improbably, not a single one falls in for a hit, wOBA doesn’t say “Well really, he got 195 hits, because there was nothing he could do about fielders always being in the way of his lines drives.” Instead, it reports what actually happened: the guy got no hits. That’s not an error in calculation, and neither are Hosmer’s outs on the basepaths.

Jason B
Guest
Jason B
2 years 18 days ago

Anon21 FTW. Well said. It’s not that the metric is bad, it’s rather that it’s being asked to do things it was never intended to do.

DavidKB
Guest
DavidKB
2 years 18 days ago

There is error in the calculation: random error. And it is also an error not to take that into account. But that’s quite an undertaking, so I can understand why it hasn’t happened yet.

Anon21
Member
Anon21
2 years 18 days ago

What kind of “random error” would affect the calculation of Hosmer’s baserunning value? Presumably, you are not saying that the number of outs he has made on the basepaths this year has been miscounted.

You must be referring to the valuation of baserunning events, which, if I understand correctly, is derived mechanically from thousands of games’ worth of data about the relationship between runs, outs, and base states. Are you referring, then, to an error in tabulating that game data, an error in translating the data into run values, or the possibility that conditions that obtained in the past do not obtain in the present (e.g., that an out on the basepaths has a smaller effect on run scoring now than it did in the past)? I wouldn’t think of the latter as a “calculation error,” just an inherent difficulty of predicting the future.

DavidKB
Guest
DavidKB
2 years 17 days ago

Anon21: Random error is a technical statistical term (see Wikipedia) that describes the statistical noise inherent in any measurement. In the case of BsR, because we have so few outs on the basepaths the noise is going to be very large relative to the number of innings observed. For that reason it should probably be treated differently than other metrics that are less noisy when compiling WAR, particularly early in the season.

Elias
Member
Elias
2 years 17 days ago

You are both right/wrong, but Anon21 is more right than wrong.

Anon21 recognizes that before one can talk about confidence intervals or estimation error, one needs to know what is the question?

I’m not sure whether DavidKB wants to know:
1) How many runs has Hosmer’s base running outs cost his team compared to how an average base runner might have performed?
2) How many runs would we expect Hosmer’s base running skills to cost his team (relative to average) if he were faced with the identical situation again?
3) How many runs should we expect Hosmer’s base running skills to cost his team (relative to average) in X amount of plate appearances in the future?

For #1, if you think the formula used for BsR is accurate, then there is no error because we know precisely what happened in each situation. I won’t claim that the formula is accurate, but if it isn’t then that is a different kind of error from what DavidKB seems to be talking about.

For #2, there definitely is estimation error, and I think that this is what DavidKB is thinking. This kind of error is exactly what Jeff is talking about when he wonders whether the outcome of a play would have been different had a catcher been in a slightly different position.

For #3, there also is definitely estimation error, and more than in #2. All of the estimation error in #2 is relevant, PLUS the error associated with uncertainty regarding the future baserunning situations that Hosmer will face.

If my response sounds snarky, I apologize. I’m just trying to point out that the question matters when talking about this kind of thing, and recommending that folks be more precise in talking about “error”.

Elias
Member
Elias
2 years 17 days ago

I’ll also note that quantifying and distinguishing between these types of errors can be very difficult.

DavidKB
Guest
DavidKB
2 years 17 days ago

I’m using a precisely defined statistical term. You can look it up if you want.

DavidKB
Guest
DavidKB
2 years 18 days ago

I guess ideally WAR calculations would account for how quickly each contribution nears its mean on average and reduce the weight on noisy statistics early in the season. Or that error should be carried over into a confidence interval. It seems like BsR might turn out to be relatively noisy early on given how seldom outs are made on the bases.

jake the snake
Guest
jake the snake
2 years 18 days ago

too bad he dont hit hrs lol no way he top 5

srpst23
Member
srpst23
2 years 18 days ago

Is this English? I have no idea what you are trying to say.

Bill
Guest
Bill
2 years 18 days ago

I believe he is saying, “It’s a shame that he fails to hit the ball over the fence. I am now laughing loudly to myself. This failure prevents him from being one of the five most valuable first basemen in American Major League Baseball.”

Big Jgke
Member
Big Jgke
2 years 17 days ago

I think you just have to run that one through Google’s FOXsports.com user/comment translator. That exists right?

Iron
Guest
Iron
2 years 18 days ago

One point I would make regarding ‘luck’… how many close plays has Hosmer been safe on this year? If it’s zero, then point taken, he is unlucky. But merely looking at the plays he was called out on and determining some of them were 50/50 plays, does not make him unlucky if he also has a nearly equal number of 50/50 plays in which he was safe.

indyralph
Member
Member
indyralph
2 years 18 days ago

Good point. I wonder though if these are actually 50/50 plays. I don’t know how to measure it, but I would guess the odds are actually quite a bit more favorable to the runner. I mean, it’s 50/50 in the sense that the catcher receives the ball just as the runner arrives. But at the point that the runner needs to make the decision, everything needs to go right from the fielder’s perspective to even make it a 50/50 play. So at the time of the decision, the odds seem significantly in favor of the runner.

DavidKB
Guest
DavidKB
2 years 18 days ago

That’s not quite true. If the 3rd base coach sent him 100 times and he was safe 50 of them that shouldn’t reflect his value at all. I guess maybe he should stop listening to the base coach, but you get what I mean.

Jason B
Guest
Jason B
2 years 18 days ago

When his third base coach is Dale Sveum, that may be a good idea anyway.

Chris Traina
Guest
Chris Traina
2 years 17 days ago

Hosmer just needs to get the barrel on the ball. Barrels Bro. Nut sacks.

Chris Traina
Guest
Chris Traina
2 years 17 days ago

I’m referencing this article from eno of course:

http://www.hardballtimes.com/learning-the-language-of-the-clubhouse/

jim fetterolf
Guest
jim fetterolf
2 years 17 days ago

Part of the problem is that in real baseball there are coaches who sometimes take chances and send runners based on a game situation and sometimes they get thrown out. Fangraphs doesn’t have a metric for that, so it’s either blame the player or blame luck. Hosmer has been a little sloppy this year, but not a projected two wins worth.

Helladecimal
Guest
Helladecimal
2 years 17 days ago

Isn’t this the guy and bro-trolled Eno Sarris in the locker room for asking pertinent, intelligent questions?

Screw ’em.

Nick
Guest
Nick
2 years 17 days ago

So basically 2014 Eric Hosmer = 2013 Allen Craig (the subject of an identical post on this topic last year.)

Tom
Guest
Tom
2 years 17 days ago

Like Iron touched on above – if you’re going to absolve him of blame on some of the outs on the base paths, you have to look more deeply. Has he safely reached a base because a throw was off target, where an on-target throw would have resulted in an out? If so, by your logic, shouldn’t that reflect poorly on his true value?

Tom
Guest
2 years 17 days ago

Also, were there instances in past seasons that he made outs on the base paths that were not his fault? It seems there most likely were. So when comparing numbers of outs season to season, you can’t just ignore the ones where there was nothing he could do, unless you ignore them from previous seasons too. Any way you look at it, he is running more poorly this year than in the past.

Balthazar
Guest
Balthazar
2 years 17 days ago

More concerning to me than Hosmer’s deflated WAR is his shaky .375 MTBOOPPFAGS Factor (Made the best or only possible play for a given situation factor). Even anonymous replacement level guy is good for about a .480.

wpDiscuz