BILL JAMES ONLINE

Cespedes' Twin

By Dave Fleming
February 9, 2016
 
Yoenis Cespedes has a twin.
 
Not an absolutely identical twin, mind you. Just a pretty close one. Cespedes and his twin were born fourteen years apart, which complicates things. 
 
Here are two sample seasons, to get at what we're talking about:
 
PA
H
2B
3B
HR
BB
BA
OBP
SLG
540
142
25
5
23
43
.292
.356
.505
536
153
23
6
26
33
.286
.328
.496
 
That first line is Cespedes’ first year in the majors. The second year is one in the career of our mystery player. Their ages, in these single seasons, don’t match up exactly, but they’re not too far off, either. Cespedes was twenty-six, while the other guy was twenty-four-and-a-half. 
 
We’ll come back to these two. 
 
*             *             *
 
I don’t know how many players have twins, whether it is relatively common or rare.
 
I do know that the idea of ‘twins’ or ‘parallels’ sometimes trickles into how we talk about baseball. Mike Trout is a good example: we talk about Mike Trout as being ‘the next Mantle.’ It’s a way of organizing the world, by parsing it into smaller sections. When we think of a player, we invariably think of whatever player seems like that player. I can’t think about Jose Fernandez without thinking about thinking about Pedro Martinez. It’s not a numbers things, just a sense: Pedro combined genius and competitiveness with a kind of joy, which Jose Fernandez shares.
 
Anyway, we can say that statistical identical twins are pretty hard to find. There are three factors that make them hard to find.
 
Factor One is the number of variables considered.
 
Take Nick Castellanos. Last year, the Tigers 3B posted an on-base percentage of .308, hitting 15 dingers. Can we find a twin for him?
 
Of course we can. We don’t have to look that far….we can just use other players from 2015.
 
Name
HR
OBP
Asdrubal Cabrera
15
.315
Marcus Semien
15
.310
Derek Norris
14
.305
Nick Castellanos
15
.303
Brett Lawrie
16
.299
Wilmer Flores
16
.295
Michael Taylor
14
.282
 
We can find seven players who were within 1 HR of Castellanos, while posting a comparable on-base percentage. Pretty easy.
 
But if we add another variable, we lose a few siblings. Let’s add stolen bases:  
 
Name
HR
OBP
SB
Asdrubal Cabrera
15
0.315
6
Marcus Semien
15
0.31
11
Derek Norris
14
0.305
4
Nick Castellanos
15
0.303
0
Brett Lawrie
16
0.299
5
Wilmer Flores
16
0.295
0
Michael Taylor
14
0.282
16
 
Our Man Nick stole zero bases in 2015. He is obviously not a twin to Michael Taylor or Marcus Semien.
 
If we add position to the list, it becomes even smaller. Asdrubal Cabrera is a middle-infielder. Norris and Flores are catchers. The only guy who plays the same position (3B) as Castellanos is Brett Lawrie.
 
And they don’t play third base the same way. Brett Lawrie is a brilliant defensive player: he is good enough to make starts at second base. Castellanos isn’t anything like Lawrie: Nick is a poor defensive player. If he is playing in the majors in five years, he will be almost certainly be a first-baseman or a DH.
 
So the amount of variables matters. That’s one factor.
 
Another factor is talent. It is very easy to find a match for Castellanos’ .308/15 line. It is much harder to find someone close to Bryce Harper’s 2015 batting line. Harper posted a .460 on-base percentage and hit 42 homers. Who is that like?
 
No one from 2015. Joey Votto comes close on on-base percentage. Mike Trout is close on homers. 
 
A few guys do show up as parallels, if you look though all seasons of baseball history: 
 
Year
Name
HR
OBP
1996
Gary Sheffield
42
.465
2000
Todd Helton
42
.463
1996
Barry Bonds
42
.461
2015
Bryce Harper
42
.460
1922
R. Hornsby
42
.459
1999
Jeff Bagwell
42
.454
1951
Ralph Kiner
42
.452
 
We found some twins to Harper, but we had to go back a few years. And, again, we’re just looking at two numbers. If we cross-checked this list with speed, and position, and attention-to-hair, and dugout fisticuffs, the list would shrink. The better the year, the tougher it is to find parallels.
 
The last factor is time: the longer the frame we’re looking at, the more differences start to creep in.
 
Castellanos and Semien and Asdrubal Cabrera and Brett Lawrie were pretty similar in their on-base averages and homers in 2015, but it’s unlikely that they were similar in 2014. It’s unlikely that they will perform similarly going forward. If you looked at three years of Castellanos’ career, you’d have a tougher time finding parallel players in his era.  
 
So we can say that ‘twining’ has an inverse relationship to a) the number of factors considered, b) the quality of performance, and c) duration of time.
 
It’s not difficult to take an average player having an average year and find other players who had similar years. It’s more difficult to find matches of good players, and it is even more difficult to find comparisons over multiple years. And the most factors you consider, the less the players are going to parallel, or 'twin.'
 
Which is what makes the existence of a Cespedes twin so weird.
 
Because Cespedes is a good player: he isn’t a mediocre player like Castellanos. And it’s weird because the parallel isn’t a one-year parallel: Cespedes matches his twin for the entirety of his career. And the parallels don’t exist in a few places: these two players match in everything...their style of play and skills and stats match across the board. They are identical twins. 
 
 
*             *             *
 
Yoenis Cespedes’ twin is Raul Mondesi.
 
We’ll get to the numbers eventually. But before we do, I want to note some of the non-numeric parallels between the two of them. First is handedness: both players are right/right….they bat and throw right-handed.
 
They are similarly sized: Cespedes is listed at 5’10", 210 pounds. Mondesi is listed at 5’11", 202. They have the same body-type, essentially: both men have thin core frames, with considerable muscle.
 
Neither one passes my spell-check. Mondesi and Cespedes both get underlined. Raul gets a pass, but not Yoenis. This probably doesn’t matter, but I’m mentioning it anyway.
 
They are both outfielders. Mondesi was a Gold-Glove winning outfielder whose strong throwing arm made up for a pretty high number of errors. He was a right-fielder who could be stretched to play center.
 
Cespedes has the exact same skill set, though he’s mostly played in left field, with some trials in center. Cespedes is famous for having a terrific arm, which makes up for his errors. Cespedes has a Gold Glove.
 
Both players are reasonably good baserunner, but neither one would ever win a footrace with Billy Hamilton. Mondesi stole more bases than Cespedes….that’s the biggest gap between them, by far. Let’s get some numbers in:
 
Player
SB
CS
BsR
Cespedes
37
18
9.7
Mondesi
106
40
3.6
 
While Mondesi stole more bases than Cespedes, the Mets outfielder rates as the slightly better baserunner, at least by Fangraph’s version of base-running runs.
 
Important note: those numbers are each player’s cumulative stats between ages 26 and 29. Cespedes had a late start, so we’re comparing his career against Mondesi’s numbers at the same age. That’s what all of the numbers reflect: each player’s Age-26 thru Age-29 seasons.
 
Before we get into numbers, one more surface parallel. Raul Mondesi won the Rookie-of-the-Year. Cespedes had an extremely strong rookie year, but actually came in second in the ROY vote. Some guy named Mike Trout won the award over Cespedes….I haven’t looked at the numbers, but I bet the voters got it wrong. Cespedes had a great rookie year. I’ve never heard of the other guy.
 
Both Mondesi and Cespedes are Latin American players, of course. Islanders. Cespedes was born in Cuba, of course, while Mondesi hails from the Dominican Republic. I don’t know that that matters, but it’s another parallel. It’s another thing they had in common: they were immigrants to the US. Spanish speakers.
 
One way it could matter is their walk rates. You ever hear that adage applied to Latin players: "You can’t walk off the island"? That holds for both of these players: they do not show as particularly disciplined batters:
 
Player
PA
BB
K
Cespedes
2435
148
508
Mondesi
2392
177
424
 
Cespedes walks a little less than his twin, and strikes out a little bit more. The gap is narrowed when you consider league contexts: Cespedes, fourteen years younger than Mondesi, is playing in an era with slightly higher strikeout rates, and slightly lower walk rates.
 
To understand their offensive contributions within the scoring conditions of their respective leagues, we should check on a few context adjusted metrics. We’ll try Weighted Runs Created (wRC+) and Adjusted On-base Plus Slugging (OPS+):
 
Player
wRC+
OPS+
Cespedes
121
122
Mondesi
118
119
 
Almost exactly the same. Cespedes shows as a hair better a hitter than Raul Mondesi.
 
Taking a step back, let’s look at their raw triple-slash lines:
 
Player
AVG
OBP
SLG
Cespedes
.271
.319
.486
Mondesi
.279
.335
.510
 
Again, we have two nearly identical players. Cespedes has a slightly better on-base percentage, but Mondesi has a bit more pop. I’ll remind you that we’re looking at about 2400 plate appearances of performance, which is a pretty sizeable data set.
 
Here’s their type-of-hit distribution. I initially calculated these as rate stats (the number of doubles, say, per 600 plate appearances), and then I realized that I didn’t have to list them as rate stats to show the similarity:
 
Player
AB
Singles
Doubles
Triples
HR
Cespedes
2249
358
124
21
106
Mondesi
2185
357
119
17
117
 
Just by counting them, they’re identical.
 
I’ve avoided using WAR thus far, but WAR has Cespedes as being a little more valuable than his twin. It helps that Cespedes posted the best season of his career in 2015:
 
Player
Games
WAR
WAR/162
Cespedes
575
15.4
4.3
Mondesi
562
12.4
3.6
 
They’re identical twins. I’m sort of astonished by this, though I don’t know if I should be.
 
*             *             *
 
We could say that this information tells us that Yoenis Cespedes isn’t a terrific long-term investment. Raul Mondesi was a pretty good player as a young man, but he declined pretty quickly in his thirties. That’s phrasing it nicely: Mondesi’s value fell off a cliff.
 
Which is what you’d expect of an aging player with low walk rates and an on-base percentage reliant on hard-hit balls, whose defensive and base running value depended on skills that almost always erode in a player’s thirties (arm strength, speed). Mondesi’s power spiked a little bit as he turned the corner, and his walk rate jumped, but the quality of his contact lessened, which caused his batting averages to plummet. His defense and speed declined, and his arm no longer made opposing baserunners nervous. By the time he was thirty-three, Mondesi was a part-time player, platooning for whatever contender was looking for a right-handed bat off the bench. He was out of baseball at thirty-four.
 
You can’t escape the fact that Cespedes has almost uncannily paralleled Mondesi’s career path. Although it is generally foolish to project one player on the basis of just one other player, it’s pretty rare to find a parallel so similar on all fronts.
 
But this is assumption ignores one factoid that I haven’t touched on. Cespedes played brilliantly last year, posting numbers that ranked him as one of the top-ten offensive players in baseball. At twenty-nine, Cespedes did not seem like a declining player: he seemed like a player whose abilities were finally reaching their apex.
 
This is not where Mondesi was in his Age-29 season. After signing a deal with the Blue Jays, Mondesi struggled through his first injury-limited year, playing just 96 games. While Mondesi posted the second-best slugging percentage of his career, his overall performance wasn’t close to the MVP-level performance that Cespedes turned in last year.  
 
Just on the basis of Cespedes’ better performance as a twenty-nine year old, I’d bet that the Mets outfielder is more productive going forward than Raul Mondesi was at the same age. That said, the comparison hints that there are some pretty big red flags about how Yoenis Cespedes’ career is going to unfold going forward, and teams were smart to avoid offering him the five- or six-year deal that most of us were expecting him to get this offseason.
 
Dave Fleming is a writer living in Wellington, New Zealand. He welcomes comments, questions, and suggestions here and at dfleming198@yahoo.com.
    
 
 

COMMENTS (26 Comments, most recent shown first)

MidnighttheCat
"So the amount of variables matters. That’s one factor."

Sorry to be a professor (I am one) but this website is among other things about rigor, so I will add some grammatical rigor.

It is "number" of variables, not "amount". Variables is plural and countable, amount is for non-countable nouns ("mail" "snow" "money") and number of is for countable nouns ("dollars" "hits" "runs" "variables").
2:32 AM Feb 24th
 
steve161
Oh, it has one. It's just no good, an acoustic wasteland.
5:32 PM Feb 14th
 
flyingfish
Steve161: London was my home for 3 years, and I never tire of it. Amsterdam is very fine. I didn't realize Munich, another place I enjoy, had no major music hall.
1:29 PM Feb 14th
 
steve161
Fish: Rijksmuseum, van Gogh, Stedelijk--Amsterdam is a short hop from Munich, where I live and which lacks a first class concert hall. It's my second visit this season and one of my very favorite cities, offering way more than legalized pot and prostitution, which get the most attention. Highly recommended, along with or perhaps even above the old standards like London, Paris and Rome.

Maris: the operative word was 'only'. Deafness commenced in the 'first period' and was well-nigh total well before the end of the 'second', which would be regarded as revolutionary even had there never been a 'third'. I don't doubt that it was a factor, I simply disagree with restricting the impact to the breathtaking and admittedly unforeseeable originality of the final years.
5:18 AM Feb 14th
 
flyingfish
In 1813, Beethoven was 42. He died at 56. Steve161: How lucky you are to have heard that concert in Amsterdam. I hope you got to the Rijksmuseum and other places.
5:33 PM Feb 13th
 
MarisFan61
Steve: It's not clearly dubious to say that. I don't know that I would say it (although I think I'd tend to); mainly just saying that what you point out doesn't at all rule it out:

Because: Degrees of it matter. Severe deafness -- being hardly being able to hear anything, and maybe nothing at all -- could well put someone into a distinctly different mindset than being just moderately impaired.

BTW, I love that we're on a Beethoven tangent.
But I hope it won't make anyone think we're necessarily done with the baseball part. :-)
2:42 PM Feb 13th
 
steve161
In the Heiligenstadt Testament, written in 1802 but not discovered until after his death, Beethoven writes that he had been dealing with increasing deafness for six years--from the age of 26. When he conducted the premier of the 7th symphony in 1813 (which I heard last night at the Amsterdam Concertgebouw), he was almost stone deaf. To suggest that only the third period, when he reinvented music no less than Einstein did the understanding of gravity, was impacted by his deafness is dubious.

I can't think of a comparable disability that would strike a ballplayer at the heart of his competence and still permit him to perform.
4:34 AM Feb 13th
 
MarisFan61
(typo: Near the end, thing, not 'think.' Sorry, folks!)
12:14 AM Feb 13th
 
MarisFan61
Fish: Nice elaboration!
BTW I didn't know Beethoven started having trouble with his hearing as early as age 30, and I doubted it -- but I looked it up and it seems to be true.

Related to what you're saying: I remember a comment Bill made, years ago, about Jim Eisenreich, something like, "What a player he would have been if not for the illness." I immediately wondered if maybe his illness was part of what drove him to become the athlete and ballplayer he was; in fact I'd guess this was so. If not for the illness, it's possible we never would have heard of him.

Ironically it may have been something else Bill said that primed me to wonder this thing that went somewhat against what he was saying there: He also wrote, roughly around the same time, something like that he thought every great ability is developed as a compensation for a weakness. I'm sure he didn't mean it literally or totally, but it made immediate sense to me as a think that may occur. Eisenreich's situation struck me as a possible example.
12:11 AM Feb 13th
 
flyingfish
Hey, MarisFan61: I love your Beethoven comment. It illustrates a point. Beethoven, of course, famously became stone deaf, roughly in the middle of his life, around age 30; this would seem to be a major handicap for a musician (and a way to divide his life into two periods). One famous music critic, I believe the Brit Sir Donald Tovey, said that Beethoven's later works, which Tovey didn't like, were that way because Beethoven was deaf.

Here's the point: Is the handicap--deafness in Beethoven's case--really something that affects the career of the person under consideration in the way that people assume? I don't believe for a second that Beethoven's late music sounds the way it does because he couldn't hear when he wrote it. It is possible that his deafness, by affecting his mental state, affected his music, but not because he couldn't hear.

So we can ask the same question about Mantle's knees, Jim Bouton's (and a zillion other pitchers) hurt arm, CC Sabathia's struggle with alcoholism, Darryl Straberry's stuggle with drugs, Barry Bonds's use of PEDs, and so on: Did those things affect the individuals' careers in the way that you would expect them to have? I think the answer is definitely yes in some cases, but probably no in others; and I think we need to keep the question in mind and look for evidence.
5:28 PM Feb 12th
 
OldBackstop
Hey Dave, Raul has been a favorite of mine, ever since he flew out of Boston early when Torres pinch hit for him. The statistical similarities you point out are remarkable, but I think Mondesi’s career ended less due to a player “type” but because he got caught up in 2004 in what was sort of the OJ trial of the Dominican Republic. The numbers maybe don’t tell the story.

You mention Mondesi’s injured Age 29 year in 2000, (even so, a .852 OPS in 426 PAs) but over his next three years (Age 30-32) Mondesi averaged 146 games, 26 HRs, 22 SBs and a .786 OPS. He finishes that streak in his Age 32 year in 2003 with 24 HRs, 22 bags, .827 .OPS in 586 PAs, and is fifth in the league in RF range factor.

But as he hit the free agent market, he was in the middle of a legal and media hellstorm, being sued by former MLBer Mario Guerrero based on alleged verbal agreements when they were teenagers. Guerrero was suing various Domincan MLB players, and the case is All We Can Talk About in baseball-mad DR. From what I can see, Mondesi had plenty of bucks and this was no threat to his financial well-being, but it was personal. He had made $66 million playing, which is equivalent to ten hundred thousand trillion Dominican bucks. In November, Mondesi has a judgement levied against him, and officials were trying to grab his house back in DR with his wife and four kids in it. When someone finally does sign him, the Pirates wind up in the middle, holding an enraged Mondesi's paycheck in an escrow fund.

Even so, Mondesi starts out his Age 33 year slashing .338/.398/.513/.910 in April. On May 6th, with a crucial court hearing coming up, he leaves the team to go back to the DR, saying to the press "My family comes first, baseball comes second." The Pirates, Sister Sledge echoing in their ears, release him on May 21, 2004.

Mondesi signs a few weeks later with the Angels, which Bud Selig thought was maybe notso-hotso, and everybody had to get deposed. Mondesi plays a week and tears a quad, maybe not so surprising given his likely workout mindset at that point, and the Angels cut him loose.

So that is his Age 33 season. In 2005, the Braves gave him a shot, but took his full-time job away in May and gave him the option of his release, which he took. You mention platooning, but as far as I can tell, platooning wasn’t part of his final few years, it was more like refusing to platoon. And his numbers are quite strikingly similar for his career:

vs: RHP: .274/.326/.486/.812
vs: LHP .270/.346/.482/.828

Now, was Mondesi really shot at that point, or could he have come back? I can't speak to where his head was at, but between the Pirates holding his checks, Selig investigating his next deal, and his family needing him in the DR, I can see him saying to hell with the Norte Americanos.

6:55 PM Feb 11th
 
MarisFan61
I agree it doesn't "invalidate" the method.
That's why I said "might"!

I appreciate your thoughtful replies and well-tempered reactions to these questions -- challenges, if you will. Not everyone manages that. :-)
5:28 PM Feb 11th
 
KaiserD2
Marisfan: I admit that "disastrous" might or might not be accurate. But I don't think an outlier like that invalidates the measurement. They do occur quite often, but I think they are simply luck. So are the outlier good seasons like 1952 (for Mantle.) It's very easy to have an unusual number of balls drop just within or just outside one's reach in a given season. In fact--you'll see this in my book--sometimes it happens to a whole team. The same thing happens for hitters, obviously, and probably accounts for some unusually good or bad years. It's what keeps us watching!
4:10 PM Feb 11th
 
OldBackstop
Hey Dave, great article.

Clavens:

-- Raoooolll has been mayor of San Cristobal in the DR, a big ass city with a quarter of a million people, since 2010.

-- the Fox News link below got the World Series story a little unfair and inaccurate. The son of Raul Mondesi that debuted in the World Series is Raul A. Mondesi. Mondesi has another son named "Raul Mondesi, Jr." age 23, who was a prospect in the Brewers system but went off the grid a few years ago. George Foreman was taken.

-- due to Tejada getting his leg broken, the Mets Matt Reynolds looked like he was going to make his debut in the playoffs, maybe the WS, but Juan Uribe recovered in time to snag the WS roster spot and get one very productive at bat in Game 3. This was on the radar of my daughters because Reynolds is "super cute."



1:39 PM Feb 11th
 
MarisFan61
(sorry about the typo redundancy)
11:02 AM Feb 11th
 
MarisFan61
Kaiser: I likewise!

About that metric: It's funny how two of us can look at the exact same data and come out with such extremely differing takes. You look at it and confidently state what it means about Mantle's fielding in those years; I look at it and what I see is that the method shoots itself in the foot.

I don't have much trouble buying the idea that Mantle had "2" fielding periods, not 3, first of all because what I said before only depended on his fielding history not being unitary -- you want to call it 2 rather than 3, it doesn't change what I said -- plus that things like this depend on whether we want to be lumpers or splitters. You want to say Beethoven had 2 periods rather than 3? Sure. To me, this is like that.

But where the method IMO shoots itself in the foot is the 1954 result. I completely do not understand how you simply take it take it at face value and state what you do about his fielding year ("disastrous"). To me, the fairly 'obvious' thing ('obvious' not necessarily meaning correct, but I think a fairly obvious suggested thing) is that it's just a wildly inaccurate result, and thus that it boldly displays the unreliability of the method -- and suggests that if we felt to begin with, as I did, that it's not much to hang anything on, we were right, and that if you didn't, you now might.

There are ways to look for confirmation or refutation for such an outlying result -- and what I see is refutation. Look at his hitting numbers, perhaps look especially at things that reflect speed, like triples, runs scored, stolen bases, stolen-base percentage, and the detailed baserunning stats readily available on baseball-ref.com (of course I realize that speed and running aren't the only relevant things for negative aberrations in fielding, but I think they're a strong indicator). What I see is that there's no reason to think he would really have had an outlyingly poor fielding year.
11:01 AM Feb 11th
 
KaiserD2
I'm glad to have started this discussion.

MarisFan61, if you accept DRA as the metric, there are really only two Mantles, not three, and even the line between them isn't that clear. What is rather odd is that his best year, ever when he saved 13 runs, was 1952, when he was coming off his big knee injury in the WS. After that, he was generally average except for a disastrous 1954, when he was -18 runs. Starting in 1958, he's below average more often than not, and he's double-digit below average five times. Incidentally, Mantle in his last year, 1968, was still a significant asset as a hitter (+24 runs above average), but he was giving back 2/3 of that at first base. Had the DH been in effect he could have continued for some time.

In Bill's Abstracts, one thing that always struck me as rather odd was that while Bill very properly had no respect for contemporary opinion (in the press especially) of players' defensive capabilities--Larry Bowa comes to mind--he thought that contemporary opinion had to be given some weight with respect to players from the past. DRA suggests that observers weren't any smarter than than they were in the 1980s, and that many supposed defensive standouts lived on their reputations for many years.

DK

1951 -
10:22 AM Feb 11th
 
steve161
The biggest difference might be in the unmeasurable area of personality. I remember Mondesi once saying he didn't want to play center field because he'd have to run so much, moving to back up plays to both left and right. I've seen no sign of that kind of attitude in Cespedes, quite the contrary.
5:55 AM Feb 11th
 
MarisFan61
About how the differing environments affect the meaningfulness of the "twin" thing: I was speaking just theoretically; I didn't necessarily mean that the Cespedes-Mondesi thing doesn't work, although I suspected that I wouldn't think it very much does.

So, to be more relevant to that example, I looked at it more specifically -- and I do think that in a sense they're similar players but that it doesn't much work to consider them "twins." I do think they're comparable in value and overall "goodness" as players. But: Their respective run environments differed so very greatly different that I don't see how we could consider the similarity of their raw offensive numbers as anything but a casual curiosity; it's little more than sort of a pun. I looked at their Win Shares for these seasons, which overall are similar -- similar enough to say that the value is comparable -- BUT, the division between Offensive and Defensive Win Shares is probably quite different (I can't be sure because the breakdowns for 1997-2000 aren't readily available), with Mondesi probably having fewer offensive Win Shares but more defensive Win Shares. Plus, as Dave indicates, their career trajectories may be importantly different. But however you slice it, anything like "twins" seems to be taking it way too far. I don't know that I'd even call them brothers; cousins seems more like it.
1:11 AM Feb 11th
 
sansho1
Regardless of how the raw numbers translate to value, the comparison is useful in and of itself. Cespedes and Mondesi have, to this point, used their time on the baseball field to do remarkably similar things with remarkably similar frequency. If you're attempting to spin Cespedes forward, the raw data are MORE useful than their contextual value, as they redound to skill set, approach, and career trajectory.
4:15 PM Feb 10th
 
MarisFan61
I agree with those who are saying that the "twin" thing isn't very meaningful unless the contexts of the respective seasons are highly similar. I think it's almost completely meaningless when the contexts are very different.


Kaiser: I was very interested to know what defensive metric you were using, when you said you used a different one than baseball-ref. I'm glad you elaborated.

But, a few things about what you said:

"....Mantle, like Joe DiMaggio, had the reputation of being a very good center fielder, but sophisticated measurement shows that those impressions were incorrect."

A couple of things about that: DiMaggio doesn't have the reputation of being a very good CF; he's most widely considered a historically great CF. I guess the metric you used doesn't show that, but others do -- IMO most notably, defensive Win Shares, which I consider the very most telling defensive metric. (I know that I'm in a minority here on that, which is a little ironic in view of what the site is.) I wouldn't say it's invalid to question DiMag's usual reputation because of what a metric shows, but I'd say it's wholly invalid to imagine that you can flat-out reject the correctness of the reputation on the basis of a metric, especially when another major metric seems to confirm the reputation.

And about Mantle: It depends heavily on which Mantle we're talking about. In terms of fielding, I think there were basically three: The young Mantle, who was the best fielder of them; the middle Mantle, who was less good; and the latter Mantle, who wasn't really a CF any more even though he still played there a lot. Even though the next part of what you said referred to individual seasons, the first comment looked like it was based on averaging his whole career; if it wasn't, you're on shaky ground. There's a strong argument that at least until the late '50's he was indeed a "very good" CF, although indeed not on the level of Mays (or DiMaggio).
3:16 PM Feb 10th
 
KaiserD2
For defensive rankings I'm relying on DRA, designed by Michael Humphreys and available for current players at baseball-reference.com. It measures how many runs a player at any given position allowed or saved, relative to average, at his position in a given year. Here are Mantle's scores starting 1951: -12, 13, -3, -18, 4, -2, 0, 016, 8, -9, 3 (that's through 1962.) He didn't get better after that.

Yes, Mantle, like Joe DiMaggio, had the reputation of being a very good center fielder, but sophisticated measurement shows that those impressions were incorrect. That is why, by the way, as I showed in a SABR presentation in 2014, Willie Mays's best seasons were almost indistinguishable from Mantle's best when fielding was taken into account: he was a great center fielder and remained so well into his 30s. Mantle (and Joe D) never were great. Another center fielder who cost his team quite a few runs was Duke Snider.

Trout saved 9 runs relative to average in his rookie year but his subsequent figures are -13, 2, and -1. Very few people substantially improve their outfield play as they get older. The Angels' other outfielders contributed very little last year, and if they could come up with even an average hitter who was a really good center fielder and move Trout, they would improve themselves. (It is by the way a dreadful commentary on Angels' management that even with the huge advantage of having Trout, they have missed the postseason in 3 of his 4 seasons.)

David Kaiser


12:45 PM Feb 10th
 
hotstatrat
You are a very critical person, KaiserD2. I don't think the point of this article was to say that these twins are exactly equal in abilities.

If I may be critical of your essay, it was well written, but you are being way too picky about Mantle and Trout not being similar. In fact, Mantle is the actual most similar player to Trout up to this point in his career and Trout was Mantle's most similar player at the same age. That's according to Baseball-Reference's formula, which, I believe, was created by Bill James many years ago. Not that that is the last word, but I think better evidence than counting how many seasons above 6.0 WAR and if it isn't an exact match, then phooey.

Your comments about Trout's defense also came off overly critical to my ears. I had to look up a couple ratings. Both Trout and Mantle received positive defensive WAR by both B-R and FanGraphs. Even if Trout is average for a centerfielder that doesn't mean he needs to be moved off. Even if he is below average and needs to move, that was a bold statement to throw into your essay as if we'd all agree without some backing evidence. Mantle wasn't regarded as a below average CF in younger days either. (I know "mediocre" means average, but in the context of your essay, it implied below average.)​
10:43 AM Feb 10th
 
Davidg32
You're absolutely right, Tigerlily:

www.foxsports.com/mlb/story/kansas-city-royals-raul-mondesi-jr-mlb-debut-world-series-appearance-postse​ason-103015
10:19 AM Feb 10th
 
KaiserD2
About 35 years ago, a guy named Bill James taught us that statistics had to be evaluated in context. It's amazing to me how easily people seem to forget that.

Let's take the table of players with 42 home runs and an OBP of about .460. In 1996, when Barry Bonds and Gary Sheffield did this, National League teams averaged 759 runs scored. In 1951, when Ralph Kiner did it, they averaged 694 runs scored. In 1999, when Bagwell did it, they averaged 810 runs scored. In 2015, when Bryce Harper did it, they averaged just 666 runs scored.

What this means is that last year, Bryce Harper had 7.9 Wins Above Average (WAA), which is a young Ted Williams-type figure. In 1996, Bonds had 7.1, and Sheffield had 4.2 (largely because of dreadful defense.) In 1951, Ralph Kiner had 6.8. In 1999, Bagwell had 5. "Twin" statistics conceal very significant differences in value. (These figures are a bit different from the ones you'll find on baseball-reference, largely because I use a different defensive metric.)

Mondesi had his "twin" season with Cespedes in 2001 when the AL averaged 787 runs per team. Now things get complicated. Cespedes last year played 102 games in the AL, which averaged 710 runs per team, and 57 in the NL, which averaged 666 runs. Pro-rating those figures, his run environment seems to be 681 runs per team per year. That makes Cespedes's season last year significantly better than Mondesi's in 2001.

Regarding Mike Trout--he has had seasons of 7.5, 6.6, 6.5, and 6 WAA in his first four seasons. He isn't "the new Mickey Mantle." Trout was only 23 at the end of last year, and when Mantle was 23 he had broken 6 WAA in only two seasons out of four. (He was on the verge of his two greatest seasons.) Trout's pattern is much more similar to Henry Aaron's, except that Aaron didn't have his first 6 WAA season until he was 23, and Trout at that age already has 4. (Having reached that level, Aaron essentially stayed there for 13 years in a row--something for Trout to shoot for.) Trout is similar to Mantle in that he's a mediocre center fielder and the Angels would probably be well advised to move him.

Similarity measures and scores become a trap when they are applied across different eras.

David Kaiser




8:39 AM Feb 10th
 
tigerlily
Mondesi has a 19-year old son who's a top prospect (as a SS) in the Royals system. Last year he made his major league debut in the World Series. He pinch hit in Game 3 and that is only major league experience. I think that's probably a first in major league history.
11:06 PM Feb 9th
 
 
© 2011 Be Jolly, Inc. All Rights Reserved.