Effectively Wild: A FanGraphs Baseball Podcast - Effectively Wild Episode 517: The MVP Conversation Conversation

Episode Date: August 19, 2014

Ben and Sam discuss Alex Gordon, Mike Trout, and the relationship between defensive stats and the MVP award.

Good morning and welcome to episode 517 of Effectively Wild, the daily podcast from Baseball Perspectives, presented by The Play Index at BaseballReference.com. I'm Sam Miller with Ben Lindberg of Grantland. Hi, Ben. How are you? Hi. We're trying something different today. Yeah. A little different. Not that different. Not really any different for anyone else's purposes, but for our own. We're recording in the late afternoon, early evening Eastern time instead of our usual late night talk. And this is a change that's calculated to make us happier and healthier people and podcasters.

calculated to make us happier and healthier people and podcasters. Just because as much as we love doing the podcast every day, it does kind of weigh on you throughout the evening hours that you have a podcast to do and maybe you need a topic to talk about. And maybe you're frightened by public speaking just in general. Maybe. So I guess the only change here is that if something happens during the night's games, we won't have our hot takes ready to go until the day after that happens, which probably won't change all that much because even though the original idea

for this podcast was that we would talk about things that had happened the night before or, or loosely, loosely related to things that had happened the night before, it rarely ends up being that. Um, so probably no one will, would even notice if I hadn't said something. So probably no one would even notice if I hadn't said something. Yeah, and we also might only do this like twice. You know, like this is not anything to be admitted to. True, yeah. All right.

Anyway, anything to banter about? Not really. All right. So Ben, it's MVP season. It is. Have you noticed how many MVP mentions have been popping up in your timeline lately? Not really. I haven't looked at my timeline lately.

Yeah, I have been. And it seems like there's a sort of a, I guess a three-headed monster maybe that has emerged or I guess maybe a two-headed monster of debate that is emerging. One of the monsters is, of course, Mike Trout, who, if nothing else, is due for a Lifetime Achievement Award at this point. And I think a lot of people would like to see him win one. And then you've got Felix Hernandez, who has been so dominant as a starting pitcher that he is leading all of baseball in some models of war. And then you have the kind of unexpected pick, the sort of, if it weren't for the presence of Trout, what might develop

into the Sabre Darling pick, which is Alex Gordon, who is currently leading the Fangraph's position player war and playing for a team that has surprised. And in a way, it feels to me like Gordon is the type of player who, you know, sort of in the Zobrist, Josh Donaldson kind of mold stat heads would probably be championing a great deal, although we might not see it develop because of Trout. Anyway, the conversation has begun, and I wanted to talk to you about, I don't know, not so much who you care to vote for.

Did you, by the way, did the EBWwa sent out their ballots yeah i did not i did not even i don't think i ever will because i'm in the new york chapter along with half of the bbwa so it's a rotating thing and it will take a long time to rotate my way yeah uh all right so uh anyway here's what made me want to talk about this. Earlier today, Jeff Passan tweeted, love Alex Gordon as a player, a legitimate star. The idea that he's the best player in baseball this year is absurd. And then he linked to the Fangraphs leaderboard, which showed Gordon atop. And I'm just going to do a little bit of a radio play here. Neil Weinberg, who is Fangraphs, what is he like, their statistics ombudsman

or something? Like he is their site educator. If you have a question about their stats or about stats, he's a good person to ask, replied, why? And somebody else jumped in and said, I'll take a shot because we know defensive stats still have much more noise than offensive stats, yet we treat equally here. Neil says, but they could be underrating him as well. I object to the use of absurd predominantly. Neil adds, look at his defensive runs saved, too. Pretty consistent. Jeff Passon writes, I understand great D is of great value, but to think Gordon is a top player.

Hayward is top 10-ish. Sorry, don't buy it. Neil, but why don't you buy it? Why do you think defense is being overweighted? Jeff, because I don't think we have anywhere near enough data to understand what really happens. And absent that, I don't know what he said after that because of the way that Twitter stacks dialogue. I'm sure he said something smart. Paul Sporer jumps in.

What's enough? Is there anything, numbers or scouting, suggesting Gordon isn't an exemplary outfielder? Jeff Passan, no, but to put a run value on it without the sort of objective data we have on offense makes it unequal. And I'll stop there. I would have enjoyed that more if you'd done different voices for everyone. If we were doing this at the regular 9 p.m. midnight Eastern time, I think I would have. But we pulled this show together about 12 seconds ago. So I wanted to ask you sort of a larger question, which is this. Alex Gordon is currently, let me see, I'll see if this is true. If he's not leading the league in defensive runs, or I guess in UZR runs is what we're talking about here.

Yeah, he is. He is leading baseball with 19.5 runs above average, according to UZR. And that's where his case comes from. If he were simply a plus defender, like a plus seven or plus eight, which would be a very good defender, he would be a few spots down the rung, and people would talk about him being in the conversation, but no more than they would be talking about Michael Brantley being in the conversation, or Kyle Seager being in the conversation, or Anthony Rendon being in the conversation. He would be in that kind of conversation. But he's not. He's leading all of baseball in defensive runs. And so here's my hot take kind of a question. If a player is leading the league in a defensive metric and therefore is atop a war leaderboard,

should that automatically just disqualify him? Like, I guess what I'm saying is, is the leader in a defensive metric almost by definition not really that good? Can we rule out the top, can we just chop 10 runs off of the top defensive rating, no matter what the metric, every year, just knowing that there's going to be noise, there's going to be inexact measurements, there are going to be mistakes in the system or in the assessments or whatever the case may be, and just chop 10 off, like right off the top of every liter? Yeah, sort of.

I was really worried you were going to say no, and then I was going to have no argument. Well, I mean, it's much more likely that it's overrating him than that it's underrating him. If you're rated as the best in the league, technically it's possible that it's underselling how good he's been. But it's more likely that the opposite is the case, that he hasn't actually been that great or that his true talent level isn't actually that great. Although those are two different debates. But Gordon is maybe as close as you can come to an exception to that rule. And this is the same debate that we have every now and then when someone who doesn't like war or wants to point out

that there's still a lot of uncertainty in war picks out some guy who doesn't have as good a reputation but is on top of a war leaderboard because of defense and uses it as kind of a cudgel to bash war. And this was the inspiration for Colin Weyer's article at BP last May, when John Heyman did the same thing. At the time, he was comparing Mark Reynolds and Elliot Johnson, neither of whom seems like all that distinguished a player now. But Mark Reynolds at the time had a 988 OPS, Elliot Johnson had a 641 OPS, and Elliot Johnson had the higher war. And so Heyman said that this was his war mystery of the week. And Colin sort of sided with Heyman and pointed out

that yes, there is a lot more uncertainty about the defensive components of war than the offensive components of war. And yet in the formula, in the equation, they are weighted as if we are equally confident in the runs saved on defense as we are on the runs added on offense. And that's not really the case. And as Colin said, you kind of have to regress the offense more or the defense more, should be more confident in the offense so so going by that you would think that if if trout and gordon have the same war and and right now they do at fangrass or you know within a tenth of a win then you would go with trout because trout has been by far the better offensive player uh and gordon is making all of that up on defense but yeah i

mean neil has a point in that gordon has been one of the better defensive players in baseball for a while now both statistically and and just eye test and and scouting wise and traditional defensive award wise. Um, Fangraphs war, I guess it's still, still based on UZR, right? And, and Gordon,

Gordon's UZR has not historically been as high as it is right now. Uh, this is a career high for him already, but yeah, by a lot, I already more than double his career high by a lot. And as Neil points out, his DRS, his defensive runs saved, have been pretty consistently in this range, close to this range.

He is a good defensive player. He's a very good defensive player. So, yeah, I mean, it's probably, I would guess that it's not overselling him by as much as I would think if it were some player who didn't have a history of doing this. But even so, all else being equal or not equal as the case may be, I would go with the offensive guy. I don't really see a reason to pick Gordon over Trout at this point.

Gordon over Trout at this point. You know, if there were more than a tenth of a win separating them, fine. But based on what we know about Trout, based on what we know about war, there doesn't seem to be a whole lot of reason to choose Gordon over Trout at this point. Trout, according to Baseball Prospectus Warp, he is leading the league. According to Baseball Reference War, he is second in the league behind Josh Donaldson. And Gordon is, what, three, four, five, sixth in the league. So it depends.

I mean, if you look at the other stats, if you want to say that the guy who's winning in two of the three win value variants is most likely better, then that would be Trout. So if you want to put Gordon in the conversation, there have been much less deserving players who've been in that conversation before. So fine. Yeah, this is getting very close, Ben, to a sincere MVP discussion. And I was trying very hard to make it not actually about the MVP

and more about defensive outlier seasons. Because I will say that I'm not saying that you need to treat every defensive rating as totally suspect and just throw it out. I mean, especially if it adds up, if it makes sense, if there's some logic to it, if it makes sense with the eye test and with what you know about the guy, that's fine. So if Gordon were at plus seven and Trout were at minus 3 or whatever, I wouldn't say, oh, well, just make them equal or anything like that. I'm specifically talking about the leader in any given year, not the guys who are plus 9, not the guys who are plus 11,

not the guys who rate well, but specifically the guys who get to plus 20 and plus 30 and sometimes even plus 40. So with Gordon, I'm trying to figure out how this works. Okay, so on the one hand, we know he's a good defensive player because everybody tells us he is. And so that is used as a defense for his 19.5 rating this year. Well, of course we know he's good. And yet we have five years of that same metric telling us he's not that good, just okay, pretty good. So you would think that it would show up in the numbers you would think that it would show up in the numbers before that if it was real. Otherwise, the number itself isn't worth a whole lot.

And then using defensive runs saved, a different competing metric, as a proof that he is that good is also tricky because it's consistent. It's like he is not worth significantly more this year by defensive runs saved than in previous years. It's saying he's pretty much as good as he's always been. And so if he's as good as he's always been, then why don't we just take UZR's previous years and say, well, he's as good as UZR always thought he was. You know what I'm saying?

Does that make sense? Yeah. Yeah. Like, there is this kind of way that you can use any of these facts in any way you want to. Like, given the uncertainty around everything, you can just, you can take any fact about his defense, about what people think of it, about what the metrics have shown in the past, about what some other metric has shown in the past, and just sort of selectively wield that. And I guess that just goes to the point that a lot of times these defensive metrics and a lot of times what the competing wars do is rather than give you multiple perspectives that you can then weight appropriately in your mind

and come to a conclusion, they become narrative machines that you can use to try to win an argument. Mm-hmm. And so maybe that's where the... I'm sort of conflicted about whether putting Alex Gordon in the conversation does him a service by making sure that he gets recognition for being a legitimately awesome ball player, a legitimate superstar, a player who deserves more attention than he probably has gotten in Kansas City, or whether it does him a disservice by just turning him into, you know, another person that we can disagree with, disagree over, and use sort of half-formed data in order to

make the case for or against him. So I'm not sure about that. What I do know is this, Ben. I went back and I looked at the leading, the top two players by fielding runs above average, which is baseball prospectus' defensive metric. I looked at the top two players for every year going back to 2005. So nine seasons, not counting 2014, and looked to see what they did the following year. And, of course, they're going to regress. We know that. We know that if you look at the home run leaders every year

and then look at what they did the following year, they will probably regress. Most of them, on average, will. That's how you get to be number one out of a group of 800 is you have the best year of your life or you have circumstances that all come together to make you slightly better than you really are. However, with defensive metrics, it's not just that. It's not just player fluctuation that changes. There are three factors that change. One is that you yourself might have the best hot streak of your life defensively. One is that as a fielder, you might get, for random reasons that have nothing to do with

your skill, more opportunities to stand out in some years than others. The distribution of opportunities is not the same. In that sense, it becomes much more like RBIs than like OPS, I guess you would say. And thirdly, you have kind of metric error. These metrics, the ones that we have, aren't necessarily precise enough to actually pick up true ability. You have subjectivity in some of them. You have limitations in some of them. And so there's three factors, and one of those is the same as in the home runs example,

and the other two are unique to defensive metrics. And so you would kind of test and you would probably hypothesize that you're going to see more fluctuation downward from the top guys. So anyway, I looked at the top guys from the last nine years and usually it takes about 20 to 30 runs to lead baseball in fielding runs above average. And so all these guys are basically between 20 and 30. So that gives you a sense of where they started. And here are their next season defensive ratings. If I say minus six, I'm not saying that they're a minus six defender, but that they lost six runs from the previous year. That make sense? runs from the previous year. That make sense?

Okay. So, minus 23, minus 18, minus 24, minus 34, minus 16, minus 5. So, that's the first guy we've got that really did anything close to the same thing. Minus 25, minus 13, minus 12, minus 16, minus 32, minus 13, minus 28,

minus 26, minus 23, minus 5, minus 11, and then finally, Brandon Inge from 2005 to 2006 actually improved, went from plus 17 to plus 24, so he's a plus 7. So basically, you have more of the league leaders in defensive rating in this advanced defensive metric. More of them, five of them, were actually below average defenders by the same metric the following year. Five of them were below average defenders and only three of them managed to come within

even ten runs of their seasons total in the following year. And so I was in the middle of doing the same thing for offense and I didn't finish it before we started and I didn't want to give you an out because I knew that you didn't want to start this experiment today. However, I will just say that the names, it's like Cabrera's been top two the last four years, so you can sort of see where it's going to go. It's much more consistent for obvious reasons. That's no shock to anybody. And yet, it just kind of goes to my point, which is that if you're an MVP candidate because you lead the league in a defensive metric, you are not, in my opinion're an MVP candidate because you lead the league in a defensive metric, you are not, in my opinion, an MVP candidate.

I'm sorry to say I am simply going to cut 10 to 12, maybe even 14 runs off of your defense, maybe even 20 runs in some cases, and you are not going to like me. So let me ask you this along the same lines. Alex Gordon is essentially a defensive-driven MVP candidate. Mike Trout is essentially an offensive-driven MVP candidate. And then you've got the third kind. You've got Felix Hernandez, pitcher, not offense, not defense, totally different thing, totally different position.

Where does he fall in this as far as the unreliability of his stats, as far as the ability to sort of give him total credit for his performance or to be skeptical and want to regress? Trout? No, Felix Fernandez. Oh, Fernandez. Hmm. for Nunes. I mean, I don't really have an objection to a pitcher winning the award, except that I kind of wish, much as I wish that they would just clarify what valuable means so that we wouldn't have to have these unproductive debates every year about whether value is is context specific or not i sort of wish that they

would just make mvp you know most valuable position player and so that we could have a most valuable position player award and a most valuable pitcher award and not not have to put them both on the same scale or get rid of the the cy young if you want just so that you just so that you don't have a special award for pitchers also not my question no not your question but um but anyway as far as as long as that uh as long as that situation persists i'm kind of kind of against the pitchers winning mvps even if they are, just because they have their own award already. But as far as him regressing or it being unsustainable or anything like that, I think... Or maybe a better way to think of it is if you had access to the other million universes that are exactly like ours and could see Felix Hernandez pitching in all these other exactly like our scenarios, would you expect him to be ahead of Trout and Gordon as he is here?

Or would you expect, like I would expect in this, if I could see all the realities, I would expect that there would be a lot more situations where Gordon is a two and a half win player than the Trout is a two andand-a-half win player. That's sort of what we're talking about here. But Hernandez, I don't know. I mean, Hernandez is clearly super good this year, so how much noise is there in his performance, do you think, or in his statistics? I guess a lot less than Gordon's, but maybe more than Trout's, probably.

I mean, I don't know. It's not like the things that he's been really good at, like strikeouts and walks and all those things, which he's been better at those things than he has ever been. And those things don't seem all that random or subject to random fluctuation but on the other hand he he has a good catcher this year and that might be part of why he's been better um and so maybe he shouldn't get credit for that when people when people ask why war you know why framing ratings haven't been added to to any the win value stats yet. It's because it's

kind of complicated. And if you add value, give value to catchers, then you have to subtract it from pitchers. So if you think that Mike Zunino is good at getting strikes called for his pitchers, which he seems to be, then that means you have to take that value away from Felix Hernandez and other Mariners pitchers. So say that we take some of that away from him and say that, you know, according to the metrics that care about things that are somewhat influenced by luck, like, you know, he has a career 300 BABIP basically, just average BABIP.

He hasn't shown that he is especially good at limiting hits on balls in play, and yet this year he has. He has the lowest BABIP of his career for a full season. So there's that. He also has the lowest home run per fly ball rate of his career this season. So maybe he's doing something differently that has caused that to happen, home run per fly ball rate of his career this season. So, you know, maybe those are, maybe he's doing something differently that has caused that to happen.

But I would guess that if the same talent level Felix Hernandez pitched this season over and over, that he would not in general do as well in those categories as he has done. So probably more variable than than trout i would think um so one kind of luck i guess it's maybe it's like luck that we don't ever really talk about is uh that when a hitter is facing a pitcher, he is also partly benefiting from whether the pitcher does something good. And we might look at strength of competition, but if Felix Hernandez, obviously,

he makes mistakes as a pitcher, or since we're talking about Felix Hernandez, let's take him out of the picture. Chris Sale obviously makes mistakes as a pitcher. And if he happens to make 48 mistakes in the course of a to make you know 48 mistakes in the course of a year like really bad mistakes in the middle of the plate and uh three of them happen to be to mike trout and he hits a home run in two doubles off chris sale um he somewhat benefited from the fact that the mistakes came to him instead of to alex gordon or josh donaldson or miguel cabrera or anybody else right and similarly And similarly, Felix Hernandez throws a pitch, and then his job is done. He throws the pitch, and then what happens after that is partly up to the batter.

And if the batter is on his game and slept really well that night and gets lucky or whatever and hits a good pitch, there's nothing that Felix can really do. And if he sucks because he's been drinking NyQuil and misses a pitch right in the middle, there's really nothing Felix did there to sort of earn that. You know what I'm saying. And so who do you think, in that kind of unacknowledged luck, who do you think has the bigger swings? Does the pitcher, is the pitcher more reliant on what hitters give him,

basically, or is the hitter more reliant on what pitchers give him? That's a deep, profound, thought-provoking question you've asked me. Well, I think Russell did some research on batters having more to do controlling the outcome of the plate appearance more than the pitcher does. And I forget what the context of that research was. I mean, intuitively, I would have thought, I guess the mistake thing sounds like it could vary more. That a batter might get just more meatballs than another batter. Just based on the luck of the draw and the day he happened to face the pitcher. And batter, I don't know, maybe position players have their routines. Maybe a lot of them get the same amount of sleep every night

and they're equally prepared to go. Then again, maybe if you're a pitcher, you face batters who are nursing some injury more often than another pitcher did. It seems impossible to say. Doing a little construction? No. Guy next to me is a that oh see that's the problem with recording while other people are awake and as i should have had i should have invited him to uh play paul spore in my reenactment of the jeff passen account so i don't know i don't have a good answer. Do you have a good answer? I don't.

Okay. I feel better then. I would think that the pitcher would be more at the mercy of the hitter than the hitter would be at the mercy of the pitcher. That's what I would feel. Okay. Which would mean that Felix Hernandez, you might regress slightly more just on that. But of course, you wouldn't do that because why would you listen to me? Why would you go through all the trouble of regressing Felix Hernandez based on something I said offhand conflictedly. For what it's worth... You'd be an idiot.

For what it's worth, Felix Hernandez has the 18th highest opponent OPS. I know, but that's not what I mean. Out of 117 pitchers with 100 innings pitched this year. I know that's not what I mean. But people might have interpreted what we were saying as saying that

he had faced weak competition or something, which is not the case. No, it's not. Although, in a sense it is because, as you know, Ben, power is at a premium in today's game. That's true. Hard to find good power hitters. Yes. All right. Anyway, can we just all not do this this year?

Can we not have this? Can we not manufacture a conversation if a if a conversation organically develops then we can have it but let's not manufacture an mvp conversation about i mean i guess the the nice thing is that the the people who would be pushing the Gordon for MVP narrative would be, if they exist, if they're really out there, they would have to be the people who buy into defensive metrics, which would suggest that they're the people who were in lot of support if only just because these are the same people who've wanted Trout to win for the last couple of years and probably would be happy to see him win this year. And this doesn't seem like it's going to be a case where we're going to be arguing about a player on a playoff team versus a player who's not on a playoff team. If the Royals do make the playoffs, maybe they'd be a better story than the Angels.

Felix, Felix, Felix though. Felix, yeah. But Trout is also likely to be on a playoff team. So hopefully that will just be taken out of the equation entirely. We won't have to argue about that. And we will only have to argue about the players' performance on the merits and not what their teammates did. That would be a refreshing change. You know what's interesting is that Gordon's defensive runs saved, which is baseball references metric, is actually higher than his UZR rating is.

And yet his war is significantly, significantly lower, like a win and a half, or not a win and a half. Sorry, a little less than a win, but he drops from first to, like, 10th. So I wonder if that's a park factor thing or what. But it doesn't matter. I'm voting for Josh Donaldson. In 2011, Alex Gordon had a great season. He was among the leaders in the various metrics,

all between six and, I don't know, six and a half and seven and a half wins. And he finished 21st in MVP voting that year. I'm guessing he'll do better this year, but I'm not buying the Gordon candidacy. I'm not taking it seriously. Okay. All right. Okay. So send us some emails for tomorrow at podcast at baseballprospectus.com.

And please support our sponsor, Baseball Reference, by going to baseballreference.com, subscribing to the Play Index using the coupon code BP to get the discounted price of $30 on a one-year subscription.