Free Agent Review: Aramis Ramirez
If you haven’t heard by now, the first free agent signing of the winter was actually a re-signing. Aramis Ramirez stayed with the Chicago Cubs, collecting a cool $73 million over the next 5 years, or about $14.5 million per season.
So why am I posting this on the blog, since it doesn’t have much to do with the Mariners?
Because Aramis Ramirez is not even a better player than Adrian Beltre, and the reactions to this signing are going to point out how even the statistical analysts of the day incorrectly evaluate players. And they make the same mistake nearly every time. Ignoring – or undervaluing, really – defense.
Now I know some of you are rolling your eyes as you read this, thinking I’m an idiot for believing that the guy who posted a .291/.352/.561 line last season and has hit 105 home runs the last three seasons could actually be a worse player than Beltre, whose offensive production hasn’t been anything close to that since the 2004 season ended. So, let me do the math for you guys, just so you know I’m not completely crazy.
Aramis Ramirez, the last three years, has posted EqA’s of .306, .301, and .293. It’s fair to say that a .300 EqA level is his established record of performance, and that’s what he’s being paid for. He’s entering his age 29 season, so the decline in offensive production shouldn’t be severe for a few years, at least. What’s a .300 EqA worth over a full season? About 100 runs, give or take a few. That’s about 40 runs more than a replacement level third baseman would contribute offensively.
However, Aramis Ramirez is a butcher with the glove. He’s a bad defensive player. By bad, I mean terrible. Every defensive metric we can find agrees with this. UZR, the best of the defensive metrics, has had him at -15 runs per 150 games played in the past. It’s unlikely that he’s consistently that bad, but if you average together the consensus of all the metrics, the scouting reports, and what Cubs fans believe about his work with the leather, marking Ramirez down as about 10 runs below average is pretty fair.
So, offense and defense, Aramis Ramirez is worth something like 30 runs a season over a replacement level player. That’s a good player, certainly, and one contributing to a team.
But let’s look at the oft-maligned Beltre, shall we?
He posted a .268/.328/.465 line this season, giving up 130 points of OPS to Ramirez. But he did it in Safeco Field in the American League (even though the difference is overstated, the AL was clearly more talented than the NL this year), so after making context adjustments, Beltre’s offensive performance was worth something like 80 runs over a full season. That’s 20 runs less than Ramirez, but 20 runs more than replacement.
Defensively, Adrian Beltre’s one of the best third baseman in the game. He’s not the best – that’s probably Brandon Inge right now, or maybe a healthy Scott Rolen – but he’s in the discussion of guys in that next tier. Scouts love his defense, fans love his defense, and the stats love his defense. He’s basically the anti-Ramirez with the leather, and his defensive value is something about 10 runs above average over the course of a full season.
Guess what? 40-10 and 20+10 are equal. They were both something like +30 runs for 2006.
However, you’ll never see anyone write that Adrian Beltre and Aramis Ramirez just had similarly valuable seasons, and if Beltre was the one with an opt-out clause, there’s no way people would be talking about him as a premier talent lining up for a monster deal.
For whatever reason, mainstream analysts simply fail to grasp the importance of defense, especially at the extremes. Beltre’s one of the best gloveman at third base around and Ramirez is probably the worst. That fact wipes away the whole of the difference between their 2006 offensive performances.
When you factor in that Adrian Beltre is a year younger, has a pristine health record compared to Ramirez, and that it’s not at all clear that a .267 EqA isn’t a low-end projection for Beltre’s offensive performance the next few years, and you have the reasons that I believe Adrian Beltre is a slightly better bet for the next few years than Aramis Ramirez.
Even if you hate Beltre and love Ramirez, the best you can do is call it a pick-em. It’s close enough that trying to decipher the margins here is basically a waste of time. The question of which one is better isn’t important – the key is that they’re basically equal.
And you’ll never, ever see a mainstream sports writer come to that conclusion. Instead, the Beltre contract will continue to be lampooned, while Ramirez is talked about as an all-star who the Cubs can build around.
Comments
35 Responses to “Free Agent Review: Aramis Ramirez”
Leave a Reply
You must be logged in to post a comment.
Interesting, thanks Dave.
Absolutely awesome post. As someone who as little as a year and a half ago argued defensive metrics blow, I couldn’t run fast enough now to get in line to be a member of the choir….
I’ve been having similar conversations on Reds blogs with people who think Dunn’s production would be almost impossible to replace.
As it turns out, good defense IS important-even if it isn’t on the back of a baseball card or rewarded in a fantasy league….
Adam Dunn, when he hits like he did in 2006, is a marginal player. He’s legitimately a -20 defender. If it wasn’t for Jorge Cantu, he’d be the worst defensive player in the game.
“Beltre’s offensive performance was worth something like 80 runs over a full season. That’s 20 runs less than Ramirez, but 20 runs more than replacement.”
I’m always surprised at how small those numbers are. I was so surprised this time that I started to write an ode to getting pitchers over hitters, but a difference of 2 runs of ERA is only a 44 run difference if they can pitch 200 inning.
Yea. People underestimate replacement level players. The generally accepted guideline is that, at most positions, it’s something like 15% below average.
Very good point about Beltre’s defense Dave.
I think that with Beltre’s contract plus his defensive and offensive value he actually has a pretty good contract from here on out. I totally disagree with the people who are so set on unloading him this offseason. I don’t think you are going to be able to find a better use for the money Beltre is gonna be getting paid.
It’s just weird how we have so many people who can perform to a certain level and with the possible exception of Barry Bonds, no one can get much beyond it.
I was over at NetShrine arguing with Lee Sinins over the induction-worthiness of Ozzie Smith.
Lee loves hitters, because Runs Created measures hitting. He’ll even make positional adjustments by ranking to top RCAA guys by position. But that sort of ranking identifies fielders only by what they do, and not by how good they are at it.
Ozzie Smith looks like crap according to RCAA, so Lee Sinins doesn’t like him. But Sinins loves Derek Jeter, giving anyone a solid opening to argue with him about the value of defence. If you include defence (and I apologise for using WARP here, Dave, but I don’t have another source for historical defensive value), Ozzie’s annualised overall value is 96% that of Jeter by WARP3.
Defense matters.
As a great Beltre fan, I appreciate your taking the time and effort to quantify his contributions. Nothing like having my deep-seated and preexisting prejudices confirmed by an expert!
and I apologise for using WARP here, Dave, but I don’t have another source for historical defensive value
WARP would appear to be giving Ozzie about 4 wins/year on defense and Jeter about 1 win/year on defense. So if you agree with that assessment of their defense, then WARP happens across the right answer here. (Personally, I think the difference could be bigger than that, but I wouldn’t stake my life on it.)
From where I sit, Jeter is a prime example of why FRAR is a crazy-bad system for evaluating defenders. In Jeter’s age 25-29 seasons, it had him as a replacement-level defender. Then, the last three years–Jeter’s age 30-32 seasons–it thinks Jeter drastically improved, to the point where he’s valued at about 3 wins/year above replacement on defense. Old dogs might be able to learn new tricks, but that’s insane.
UZR from 2000 to 2003 puts the difference between them at 33 runs per 162 games. Not sure how relevant those numbers are in 2006.
The 2006 Fans Scouting Report had Beltre as a 77 and Ramirez as a 47. Does anyone know if Tango still computes Fan Runs? I’d like to know what that spread is worth.
what’s currently thought to be the best way to convert defensive metrics data to runs?
I know most people probably think UZR is the best but Dewan’s is the most accessible data (at least for me-I have to admit a bias for his system because its sooooooo easy to rough justice things) and closest to current (i.e. judging AROD based upon ’03 data really isn’t that appealing to me)…
Basically I’m wondering people’s thoughts on the best way to evaluate the defensive side of the runs equation in order to evaluate a trade with a guy like Hank Blalock for instance (where ’03 data might suggest something dramatically different than ’05).
Great post. Thoroughly thought out and indesputable. The kind of article I wish I could read in the local paper. I agree that there’s not enough proof to say that Beltre won’t have better offensive years in the next few and that he doesn’t haveto in order to be aproximately equal with Ramirez.
This is just an insane deal.
Its not like they can move him across the diamond to first once his defense becomes pathetically bad– and he’s not an MVP caliber bat either.
Interestingly, the mainstream sports writer is (on average) the same person who idolizes little ball and defense, yet ignores defense (and little ball) when rating contracts.
Which is another factor in understanding how sports writers and baseball coaches evaluate and vote for gold glovers: they don’t have a coherent defensive metric other than gut feelings and fielding percentage, meaning that even those who are trying to vote for the best fielders have little basis for a rational vote.
For example, I was astounded to find out that Betancourt was toward the bottom of several defensive metrics. Given his grace, agility, and what looked to be excellent range, I would have put him somewhere up toward the top.
Small sample selection, granted, but who would have guessed that any metric would portray him as mediocre?
And you’ll never, ever see a mainstream sports writer come to that conclusion.
Bleh. I’m not holding my breath for the mainstream media to ever learn to truly judge the players and the game they’re paid to report on. Mo Vaughn’s AL MVP is not too distant in the past, fer chrissakes. So many of these commentators out there are practically foaming at the mouth to jump on any player who looks to be underperforming for their 8-digit contract. Hardly anyone out there has the balls to says that Adrian Beltre, while not hitting close to his 2004 form, provides enough defense to largly mitigate the issue.
I try not to take extreme positions, and I’m open to correction, but with all due respect to whoever deserves respect, if you don’t think Ozzie Smith is a hall of famer, I can’t take your baseball opinions seriously.
WARP would appear to be giving Ozzie about 4 wins/year on defense and Jeter about 1 win/year on defense. So if you agree with that assessment of their defense, then WARP happens across the right answer here. (Personally, I think the difference could be bigger than that, but I wouldn’t stake my life on it.)
The best defensive shortstop of the last 5 to 10 years is clearly Adam Everett, who is putting up seasons of something like +25 to +30. Jeter’s generally a -5 to -10 or so. So, if we assume that Ozzie was close to Everett, a three win difference probably isn’t large enough, I’d agree. But if Ozzie was overrated defensively, even a little bit, a three win difference with the glove could be about right.
The 2006 Fans Scouting Report had Beltre as a 77 and Ramirez as a 47. Does anyone know if Tango still computes Fan Runs? I’d like to know what that spread is worth.
The last time Tango published his fanruns data, it had Ramirez as a -15 and Beltre as a +12. If anything, my suggestion that Beltre is 20 runs better than Ramirez defensively is conservative. It might be 30 runs.
what’s currently thought to be the best way to convert defensive metrics data to runs?
Well, the absolute best way would be to have the raw play-by-play data and calculate an expected run value based on each type of ball and where it was hit, and then calculate the actual run value, and then add that up over every opportunity a player has. I’m pretty sure this is how MGL calculates UZR. But since raw play-by-play data isn’t publically available, we don’t get to review the information, which is too bad.
Basically I’m wondering people’s thoughts on the best way to evaluate the defensive side of the runs equation in order to evaluate a trade with a guy like Hank Blalock for instance (where ‘03 data might suggest something dramatically different than ‘05).
Take a prism view – find as much UZR data as you can (and MGL’s posted a lot of recent numbers in threads at Baseball Primer and his blog), look at Chris Dial’s ZR (along with the other guys at Primer who are working on similar projects), look at Dewan’s +/-, look at Pinto’s PMR, look at Gassko’s RANGE (maybe, I’m still not convinced about it, actually), look at the fan scouting report data, and then just create a range of probability out of that. If every system has a guy as +10 to +15 over multiple seasons, and he’s won 10 gold gloves and he looks great when you watch him, you can be pretty sure that he’s probably +10 to +15.
Interestingly, the mainstream sports writer is (on average) the same person who idolizes little ball and defense, yet ignores defense (and little ball) when rating contracts.
Even the sabermetric writers are going to get this wrong. Keith Law had Aramis Ramirez as his #1 free agent in the list he did for ESPN. The Cubs blogs love this move. I’m getting ridiculed at Baseball Primer for this post. It’s not just the media – everyone underrates defense. It’s amazing.
I keep waiting for him to turn back into the Aramis I remember from the Pirates.
FWIW, in the Rosenthal thread (the only mention of you I found on BBTF), there are also folks agreeing with you.
Personally, given the differences in park and league, the scars I bear from Russ Davis’ butchery, and the fact that Adrian has more upside with the bat than does Ramirez with the glove, there’s absolutely zero question which one I’d rather have for the money.
poking around, the belief seems to be that he ‘left money on the table’ by re-signing with the Cubs. man.
$30 million, if you believe Jim Hendry. (And really, why should you?)
poking around, the belief seems to be that he ‘left money on the table’ by re-signing with the Cubs. man.
I think it’s certainly possible he could’ve gotten more money elsewhere. It seems to happen all too often that a GM will say “Player X has this attribute that I really, really like, so I’m willing to ignore the other aspects of his game.” For instance, on a smaller scale, Terry Ryan looked at Tony Batista during spring training last year and stuck to his guns that Batista would actually be a solid fielder at third base. Ryan was so enamored with Batista’s supposed soft hands that he didn’t care that he couldn’t range more than a step to either side of him–he would “make the routine plays” and that was good enough.
So I could imagine a situation where a GM looks around, sees that Ramirez has a good bat, can play a position where you want a good hitter, and decides that Ramirez is worth more than $73M/5yr. It only takes one crazy fool.
As long as you brought up Inge, can you (or anybody) explain how a catcher can get moved to 3B and become a top-tier defensive player?
He played C all through the minors, although I see he was the Colonial Athletic Conference All-Star SS in 1997 (but, for what it’s worth, he was also a Colonial Athletic Conference All-Star RP that year).
Yeah, C is to the left (or is it right) of 3B on the defensive spectrum, but to be able to play both at a top-MLB level is amazing.
Law: Ramirez a Bargain
I don’t think that most who are serious about evaluating performance in relation to payroll would disagree with the Dave’s assertion that the affect of defense is underrated. It’s just that given the fickleness of defensive metrics, its been easy for some to ignore the issue altogether-especially when offensive contribution is so well characterized these days.
I’m at least a more recent convert and certainly am still on the linear portion of the learning curve. But I can’t help but think that if Dewan would include a rigorous runs conversion metric with his plus/minus system and updated data was in the market place every season in a readily accessible form like his fielding bible, I think Dave’s sentiment would rule the day. Actually, I’m not sure what is stopping Dewan and Co. The run expectancy data is out there, Chris Dial appears to have done all of the hard brainwork (at least enough to form a foundation)….
One thing not mentioned here… like offense, defense can decline pretty rapidly as a player ages, often even earlier than offense declines. I’m not saying that Beltre’s defense is necessarily going to decline, but it probably will – it’s only natural. It only takes a one hamstrung pull to convert a + defender into a liability. That same hamstring pull might not affect that player’s offense at all.
Hey Dave,
I’m just wondering why you wrote this post defending Beltre’s contract. I’m hoping you are not doing this because you know something the public at large doesn’t know — like a decision by the Ms to shop Beltre rather than Sexson.
Pardon my paranoia if it is unwarranted, but shopping Beltre would be completely in character for an organization that focuses on what players can’t do rather than what they can.
#24– Inge was the classic ss/pitcher all through school; he was very surprised to hear he’d been drafted as a catcher.
#28 – I think you’re reading more into this than is probably there. The best comparable on the M’s roster to Aramis Ramirez is Beltre, regardless of whether he is being shopped or not.
There was a really good discussion on this topic at SOSH: [url=http://sonsofsamhorn.net/index.php?showtopic=12346&view=findpost&p=499740]link[/url]
Some people agreed with Dave’s methods while others did not. I think the consensus was that the fielding statistics are less reliable than the offensive statistics. I would agree with that sentiment. I don’t think you can just add/subtract runs saved in this manner and say with certainty that Beltre = Aramis. We had a similar discussion regarding Mike Lowell VS ARod VS Aramis FWIW.
I agree that defense is undervalued by the mainstream media, not to mention how many guys are misidentified as good fielders. However, it’s worth noting that you while you state the difference in their glovework wipes away the difference in their 2006 offensive season, that Ramirez looks like he may have improved a bit on defense this season.
Dial has Ramirez at -4 RS per 150 games. While everyone swears by UZRs, Lichtman posted how much in line Dial’s ratings were with his UZRs. So, its quite conceivable that Ramirez improved this year to the point of merely being below average. If so, I’m not so sure that the defensive gap is enough to make up for the offensive gap in 2006.
Keith Law doesn’t understand defense?
Great post, Dave.
While there’s something to be said in favor of seeing this kind of discussion in the mainstream media — specifically, that it would be nice to see players whose sum contribution on offense and defense is equal get the same kind of treatment, particularly when something like the HoF is at stake — I’d like to see absolutely *none* of that, and for the Mariners to exploit that inefficiency in the market.
I’ve often wondered how much of the “defense doesn’t matter” mantra that was attributed to Oakland in the years prior to and during the “Moneyball” era was purposeful. Perhaps Beane was simply lowering the price on his next chosen commodity.
Great post, but I have a minor quibble (and one that will probably go unnoticed, since it is 5 days late).
You use their numbers in relation to replacement level regarding their offensive contributions, but use their numbers in relation to average regarding their defense. The final outcome would still show them as equal, but it would be 30 – 10 = 10 + 10, and they would each be worth 20 runs above average, instead of 30 runs above some conglomoration of average and replacement.