FanGraphs Baseball

Comments

RSS feed for comments on this post.

  1. Here’s a question: could handedness be part of the equation for GMs? Generally speaking, left handed hitter play first base or in the outfield and those are the four highest paid positions. I wonder if good hitting lefties are skewing because they are more scarce than good hitting righties.

    It’s a thought, at least.

    Comment by Jason461 — August 6, 2012 @ 2:21 pm

  2. Matt Swartz needs to write more articles on FanGraphs. This is superb.

    Comment by Matt Hunter — August 6, 2012 @ 2:33 pm

  3. If teams played some of the many replacement-level corner outfielders at 1B, which is generally considered to be an easier position on the defensive spectrum, would the pool of replacement-level 1B increase? It seems intuitively as though there are plenty of capable of playing replacement-level 1B, even if they aren’t “first basemen” by trade.

    Comment by Nathan Biemiller — August 6, 2012 @ 2:50 pm

  4. I would think a big part of why outfielders and 1st basemen (maybe) get paid so much is because they’re #3 and #4 hitters. The MLB GM’s and managers try to get 60 of these guys onto their teams every year. Other lineup slots they’ll fudge a bit with (Cubs batting Starlin Castro of the .300 OBP batting leadoff because he can run a bit), but you HAVE to have 25 to 40 HR guys batting third and fourth. This mentality artificially drives up their value.

    When A-Ro got $25 million a year, it was $22.5 of it because he could hit 50 HR’s, and $2.5 of it because he happened to play short at the time.

    Comment by The Real Neal — August 6, 2012 @ 2:50 pm

  5. I concur.

    Comment by zipperz — August 6, 2012 @ 2:53 pm

  6. I think this is a big part of the explanation for why it happens, but it’s an open question whether it should be happening this way. I also wonder how much of this might be because of perceived injury expectations, since 1B should theoretically project to be healthier than C or 2B.

    Comment by ralph — August 6, 2012 @ 3:04 pm

  7. First basemen and outfielders also have greater utility as pinch-hitters than similar-WAR players of other infield positions. This likely contributes to the scarcity of a true “replacement level” 1B, for example.

    Take the Cardinals. Organizationally, they choose to have considerable depth at first base. With Allen Craig, Lance Berkman, and Matt Carpenter at the MLB level, and Matt Adams who proved to be above MLB replacement level before being demoted back to AAA, they clearly choose to keep tons of organizational depth at 1B. Carpenter can also play other positions, but generally is a poor defender at 3B and 2B, so it is reasonable to consider him in the 1B category. No wonder so many teams have trouble finding a “replacement level” 1B as easily as they can find a replacement level SS. Replacement level players at tough defensive positions are much less likely to be stockpiled as pinch hitters at respectable salaries, their value derives directly from their presence on the field.

    Comment by Sweeny — August 6, 2012 @ 3:05 pm

  8. interesting argument – does this mean that a lot of those big contracts to first basemen and corner outfielders that look bad on the surface but may in fact be market price need to be re-evaluated?

    Comment by miffleball — August 6, 2012 @ 3:13 pm

  9. Great article. It is funny to see that AROD’s 2001-2007 deal was considered a bargain, given the enornomity of that contract. The fact that is considered a bargain is a good reflection of just amazing he was as a younger player. When AROD is as much of a bargain as Luis Castillo, it can change your thinking about contracts.

    Comment by Pinstripe Wizard — August 6, 2012 @ 3:14 pm

  10. Simple answer: They hit the most homeruns. Homeruns = what casual fans care about. If a banana costs $100 and your team needs a banana, guess what, you’re gonna buy a banana.

    Comment by Harry — August 6, 2012 @ 3:16 pm

  11. This is prompting me to think out loud a little more about offense and defense. Obviously, just looking at WAR does not separate out hitting and defense. Pitching can replace defense to some extent, but clearly cannot replace hitting. So perhaps all things being equal, teams are more willing to pay for hitting WAR than defense WAR (and I’d classify positional adjustment as contributing to defense WAR) simply because of those facts of life, and then hope they can make it up on the pitching side?

    It’d be interesting to see a breakdown of how the WAR of top teams’ hitters through the years have been broken down into hitting versus defense… I suspect the Rays would be high the defense side, perhaps showing that the two are more interchangeable than prevailing wisdom allows for.

    Comment by ralph — August 6, 2012 @ 3:16 pm

  12. I think it’s almost that simple, yeah. I think it ultimately comes down to the fact that each WAR is not created equal; it’s not linear. The dollar value from 0 to 1 WAR is not the same as the dollar value equivalent from 5 to 6 WAR. It’s pretty evident that teams are willing to pay more for that 6th win than they are for that first win, likely because the market for players that can put up a 6th win is terribly scarce compared to the market for players that can put up the 1st win.

    Players who are capable of 5 WAR seasons have a better likelihood of putting up a 6th or 7th win than players capable of 1 WAR have of putting up a 2nd or 3rd win…that’s the current mantra of front office execs these days not named Andrew Friedman.

    Comment by KMiB — August 6, 2012 @ 3:42 pm

  13. This is very interesting but I think the answer is probably much more simple than this. There is a large amount of debate about defensive statistics and the value of defense, while there is not much debate about the value of offense. Furthermore, we know that some of the organizations do not utilize sabermetrics much or at all. Most of the better hitters are your outfielders and first basemen, with some exceptions in other postions like Tulo, Longoria, etc. When you have an undetermined value (defense) influence a player’s worth, more weight will be given to the statistics that are a certainty. I would believe that is why you have those players receiving the most money since they put up a large amount of offense that is easily measured.

    Comment by Average_Casey — August 6, 2012 @ 3:51 pm

  14. All organizations now use some saber. Some just don’t talk about it at all in the press.

    Nor have much reason to, given that doing so doesn’t do much to put fannies into the seats.

    Comment by Richie — August 6, 2012 @ 3:55 pm

  15. Yep. Some, at least.

    Comment by Richie — August 6, 2012 @ 3:56 pm

  16. Interesting that Matt wrote this article and never mentioned arbitration salaries (he is the guru after all). The fact that 1B/corner OFers tend to accumulate the right stats for arb raises (Ryan Howard is the poster boy, along with Fielder) causes those 40-60-80 arb years to protend big FA contracts. GMs can also sleep better at night knowing they have the guy’s decline years at an easy defensive position (no worries about moving him down the spectrum later) and can often use him at DH too. This is different than a SS or 2B who is more likely to be beat up by the demands of their position, but if they are above average will be locked up prior to FA since it will keep their cost down and they won’t have to worry about locking up too many decline years (see Kinsler, Utley, possibly Cano, Pedroia, etc.) This is great analysis, but to overlook how they BECOME expensive FAs is not putting all the cards on the table.

    Comment by DD — August 6, 2012 @ 3:59 pm

  17. I dunno. Matt might be too good. Cameron writes something controversial or maybe even obnoxious, with a few logical or empirical holes just begging for rejoinders, and so generates dozens of response hits. Which figures into profitability, if I recall my internet economics accurately.

    Matt, on the other hand, covers each base you can think of. Often nothing to respond with other than ‘this is superb’.

    Comment by Richie — August 6, 2012 @ 4:03 pm

  18. Given that lefties can only play those positions, that ought to push supply there toward surplus then. If you designed a study to account for that, tho’, I do wonder what might come out of it.

    Comment by Richie — August 6, 2012 @ 4:05 pm

  19. Hitting is also a more scarce resource than defense. Any decent minor leaguer could probably play in the majors purely in terms of fielding & throwing, but it’s the ability to hit that is placed at a premium.

    Comment by bill — August 6, 2012 @ 4:06 pm

  20. Interesting thought here, thanks. Basically every team will want 2 1B to 1.33 2B, SS and 3B. Then 2 more in the OF. With the backup 1B and OF guys getting onto the field more than the other backups for special duty.

    Does WAR really capture that?

    Comment by Richie — August 6, 2012 @ 4:12 pm

  21. ” that ought to push supply there toward surplus”

    dont think that holds, because lefties are a minority of the population, only about 10-20 %

    def an interesting angle to consider here

    Comment by brendan — August 6, 2012 @ 4:27 pm

  22. excellent excellent point, especially with regard to Howard… (looking back, the Phillies big mistake was in NOT extending him either during or after his 2006 season… something like Utley’s deal would be over after next season, after which, assuming everything else remained the same [big assumption, I know], he simply wouldn’t command any kind of enormous deal, no matter how deluded the GM; I have no idea whether such an early extension was even discussed, or would have been accepted, but it was clearly the correct move in retrospect)

    Comment by Richard — August 6, 2012 @ 4:29 pm

  23. This needs a new title. I’ll be searching for it a few times over the next half dozen years and it needs to be easier to find.

    Comment by Rob — August 6, 2012 @ 4:34 pm

  24. My next question would be related to the other people’s players vs your own players point. How many OTP at premium positions who are of star caliber are reaching FA? If none of the select group of stars reach FA wouldn’t that depress the $/war since the remaining group probably bundles close together in quality?

    Don’t we think it’s possible that this is just a highlight of GM’s preferring an exponential $/war preference? If higher quality FA at certain positions (1b, OF) vs (2b, 3b) we would expect the 2b, 3b, SS group to be depressed if they preferred a sliding $/war scale based on more exponential growth?

    I mean lets take a look at the guys getting locked up at those positions: Tulo, Longo, Zimmerman, Hanley, Kinsler, Pedroia, ect. That would skew the data if it wasn’t a perfectly linear scale no?

    Comment by Colin — August 6, 2012 @ 4:39 pm

  25. Why wouldn’t GMs try to sign more right-handed players then? And why aren’t catchers overpaid by this logic? Are you suggesting that there is irrational GMs who feel like they need to pay more for the same production from a lefty slugger than a righty slugger or some need to balance? I’m skeptical that free agency significantly changes the lefty/righty splits of your lineup.

    Comment by Matt Swartz — August 6, 2012 @ 5:31 pm

  26. How much could this be biased by the ass-end of long contracts? Where the front end of the contract has a younger player capable of playing a tougher defensive position (ss/2b/3b and to a lesser degree cf) and then in the later years gets shifted to RF/LF/1B because he’s old now? When they’re 4/5/6, they’re underpaid, and then when they’re 3/7/9, they’re overpaid. And that’s just the way a fair long-term contract plays out and produces disparity in the direction mentioned with no misevaluation.

    Looking at 1 or 2 year FA deals exclusively might be more telling since it should mostly remove the position-shifting bias.

    Comment by Calvin — August 6, 2012 @ 5:38 pm

  27. Also, Matt leaves far fewer opportunities to shamelessly collect upvotes with witty rejoinders.

    Comment by Well-Beered Englishman — August 6, 2012 @ 5:53 pm

  28. It holds in that:

    As hitters, they’re definitely more than 10-20% of the MLB population, especially when you figure in that half the fielding positions are denied to them.

    Even were they only 10%, that amount still plops down exclusively at half the positions. Creating that much of a surplus there in comparison to the other half of positions.

    Comment by Richie — August 6, 2012 @ 6:21 pm

  29. Ralph, I don’t think the defensive explanation hits it, because all it’s saying is that is that different positions are worth different amounts even if the fielders are average for their position. In your example, you would still want to pay more for good hitting at glove-first positions if the other options at the position were worse.

    Comment by Matt Swartz — August 6, 2012 @ 9:27 pm

  30. Actually WAR is linearly paid for– check this out:
    http://www.fangraphs.com/blogs/index.php/a-retrospective-look-at-the-price-of-a-win-part-two/

    Comment by Matt Swartz — August 6, 2012 @ 9:30 pm

  31. Again, an important point– it is not that I’m saying teams are underpaying for defense. They are underpaying for infielders in general. If there are very few SS that can hit compared to 1B, then you should pay more for the same batting stats regardless of whether this is an average fielding shortstop.

    Comment by Matt Swartz — August 6, 2012 @ 9:50 pm

  32. Well, they seem to do it with pitchers. For example, Oliver Perez. Why not hitters?

    And at this point we’re not arguing about whether GM’s are irrational, we’re only talking about the how.

    Comment by David — August 6, 2012 @ 9:55 pm

  33. Thanks for recognizing my arb work!

    I actually thought about this concept, which is why I put in the last table at the bottom of this article, and mentioned the players like ARod, Beltre, etc. after. They are rare, but so are good outfielders (the upcoming free agent class not withstanding)

    Comment by Matt Swartz — August 6, 2012 @ 9:56 pm

  34. It’s a very good point, but it’s not actually the issue. Check out the second table in this article: http://www.fangraphs.com/blogs/index.php/positional-differences-in-the-price-of-war-2/

    Comment by Matt Swartz — August 6, 2012 @ 9:58 pm

  35. See above comments for more detail, but the $/WAR preference is actually linear and there were quiet a few OTP who were stars in the infield.

    Comment by Matt Swartz — August 6, 2012 @ 10:00 pm

  36. I don’t see this, Matt.

    “(M)inor-league catchers are not all that much worse than minor-league first basemen at hitting “, your 4th paragraph under ‘Method 1″. If that holds for minor-league shortstops too, then who cares how few shortstops hit like Mark Texeira?

    Your tables 1 + 3 suggest different things regarding the replacement level gap vis-a-vis 1st base and short. Now if teams do restrict FA supply by signing up their promising you middle infielders in a way that they don’t for 1st base or outfield, then that again goes quite a ways toward wiping out any payment gap.

    Comment by Richie — August 6, 2012 @ 10:44 pm

  37. “promising YOUNG middle infielders”, of course.

    Comment by Richie — August 6, 2012 @ 10:45 pm

  38. Maybe teams measure offense different than we do. Try RBI totals per position, or change the way you measure offensive WAR using OPS, although I believe RBIs are the main culript.

    Awesome series of articles by the way, I appreciate how you stay discussing in the comment section.

    Comment by chel — August 6, 2012 @ 10:50 pm

  39. For 2007-11, there were an 7.6 WAR per year at 2B from other team’s players, 12.3 WAR per year from SS from other team’s players, and 13.1 WAR per year from 3B. These players are out there. There are more resigned infielders than outfielders, but the difference isn’t that huge. In fact, 1B have had more of their WAR from extensions of any position:

    C- 53%
    1B- 78%
    2B- 66%
    SS- 56%
    3B- 65%
    LF- 44%
    CF- 21%
    RF- 45%
    DH- 36%
    SP- 58%
    RP- 40%

    Comment by Matt Swartz — August 6, 2012 @ 10:56 pm

  40. Yeah, I do plan on throwing RBI into a model at some point and seeing what happens. I know it drives arb salaries. I wouldn’t be surprised if reputation played a role too and RBI can certainly proxy for that too.

    Comment by Matt Swartz — August 6, 2012 @ 10:57 pm

  41. Interesting. Although averaging the numbers above shows a $3m/WAR difference between the C/2b/ss/3b and 1b/lf/rf averages while the chart in your link has it at 2mm/war. So it looks like that effect could be about a third of the issue (starting from pure salary).

    Also, aren’t injury rates/aging curves far the worst for C and 2b? I don’t have time ATM to look for how you selected players here, but is this controlled for?

    Comment by Calvin — August 6, 2012 @ 11:11 pm

  42. And since no player will sign a contract that pays him less than what he is making already, it drives extensions and free agent contracts.

    Like other commenter said, it’s important how they get expensive as well.

    Comment by chel — August 6, 2012 @ 11:13 pm

  43. The link in the chart is a $2.5MM difference for all years but $2.0MM for the first year, so at most, it explains 20% of the effect. That’s a good point, but it’s not really driving this. I think the control for different aging curves is pretty much accounted for by taking the two charts together. Actually, I think it’s somewhat of a myth that second basemen age poorly. Catchers do a little from what I’ve read, but I remember articles saying 2Bs and other players are comparable.

    Comment by Matt Swartz — August 6, 2012 @ 11:38 pm

  44. Players sign contracts that are less expensive than their previous salary all the time in their mid 30s. Nowadays, many former stars sign $2MM deals at the end of their career or even minor league contracts.

    I think that players will take the best offer, for the most part, with a few exceptions here and there. The big drivers of salary are going to be willingness to pay, not willingness to accept. No one has the option of making ten million dollars per year doing something else.

    Comment by Matt Swartz — August 6, 2012 @ 11:40 pm

  45. Am I getting this right? The difference between a good 1b and a replacement 1b is far greater than the difference between a good C and a replacement C?

    Comment by adohaj — August 7, 2012 @ 12:24 am

  46. I don’t know about the rest of you guys, but here are some things that I just can’t get past:

    1) Tampa Bay’s success during the Andrew Friedman era relative to their peers (surely the Rays are top five in MLB in wins/year over the past 4-5 years) despite not allocating many dollars to 1B;

    2) The contracts of Ryan Howard, Mark Teixeira, Albert Pujols, etc.;

    3) The run differential of St. Louis (#1 in MLB this season) after letting some other idiot grossly overpay for Pujols;

    4) The fact that 1B is surely the easiest position to play in the field;

    5) The fact that so many of the teams legitimately competing for playoff spots (Baltimore, Tampa Bay, Texas, Oakland, Washington, Atlanta, Pittsburgh, St. Louis, San Francisco, Los Angeles, Arizona) are committing minimal payroll to the 1B position.

    I appreciate the analysis by the author here but it just never seems like a good idea to commit $100+ mil to any 1B, period. I mean, my god, Albert Pujols may wind up challenging Lou Gehrig for the title of best 1B in the history of MLB, and he’s got some extremely good years left in him yet, but there is close to a 0% chance that he will live up to that contract.

    Comment by Robbie G. — August 7, 2012 @ 3:08 am

  47. I really have no idea. It was just something I noticed. You do hear an awful lot of talk about wanting a balanced lineup. Lefties also have a natural advantage since the majority of pitchers are righties.

    Comment by Jason461 — August 7, 2012 @ 6:25 am

  48. I meant players in their arb years or just becoming free agents signing their first contract.

    Could you find the average number of contracts that each position signs in their entire career? That might be an interesting correlation

    Comment by chel — August 7, 2012 @ 9:03 am

  49. I was thinking the same thing. How quickly will a SS decline v a 1B? And the defensive replacement value at SS might be huge. The RS traded Scutaro with no decline at SS. They can replace Aviles with Ciriaco this year, or Iglesias next year, and the defensive values might be enough to cover the lack of hitting. The difference in range is really that meaningful.

    The difference in range at 1st for guys like Tex and Gonzo, 5 years from now, will hardly be noticeable.

    Comment by Joebrady — August 7, 2012 @ 10:58 am

  50. That’s paper value. In real life, the contract was so bad that TX paid the NYY $27M (?) to take the contract.

    Comment by Joebrady — August 7, 2012 @ 11:00 am

  51. One of things I haven’t seen mentioned is name recognition. I don’t think that even LA thinks that Pujols was a smart BB move. But you have to consider offsetting cash flows. Signing a slick-fielding 2B with a good OBP is usually a nice move, but will doesn’t register much with the casual fan.

    Comment by Joebrady — August 7, 2012 @ 11:16 am

  52. Could they measure offensive production based partially on WPA? Over the past 3 calendar years, 6 first baseman are in the top 10 for WPA (Votto, Cabrera, Pujols, Fielder, Gonzalez and Howard).

    Comment by Roy S — August 7, 2012 @ 12:27 pm

  53. Actually if you use historical data (WAR) to match one teams best case scenario projection for a player, you’re always going to have incorrect conclusions.

    A basic principle of good business decsion modelling is to never use agglomerations of data to make a specific forecast (or to due the reverse, which is what the linked articles and dozesn of others over the past few years have done).

    If your FA value projections don’t match up to what the market is doing… your model is wrong.

    Comment by The Real Neal — August 7, 2012 @ 12:54 pm

  54. You hit my pet peeve: the myth that 1B’s are less prone to injury than players at other positions.
    Whenever I bring this up, I get a lot of replies of the “it’s logical so it must be true” sort. I don’t even agree with the “logic” even if “logic” were a good guide to what is true.
    The argument that 1B’s are older doesn’t prove anything, as players are moved to 1B when they get older because of their limited range–a clear case of selection bias.
    I have never gotten a reply referring to an actual study nor any statistical work done by the replier himself.

    Comment by Baltar — August 7, 2012 @ 12:57 pm

  55. I think this injury difference at 1B is a myth. The “logic” (what you call “theoretical”) does not hold up, except perhaps in the case of 2B’s and C’s, who probably are more likely to get injured due to collisions than other positions.
    I have challenged this theory many times and never have received a reply in the form of referral to an actual study nor a statistical case by the replier.

    Comment by Baltar — August 7, 2012 @ 1:03 pm

  56. Sorry for the double post. I got an error screen when posting the first, so thought it didn’t get in.

    Comment by Baltar — August 7, 2012 @ 1:04 pm

  57. I definitely think that GMs will pay for lefthandedness in hitting. It probably is more pronounced for the secondary tier of free agent hitters, since they may be picked up to fill a lefthanded role in the lineup or on the bench. LHBs generally have bigger platoon splits, and are more likely to be picked up purely for platoon roles on the team. It may well be rational, given that platoon hitters may be more cost effective or that teams may be vulnerable to tough RHPs based on their existing lineup. I think you would have to look at each signing to judge whether it is rational or irrational.

    Comment by CJ in Austin, TX — August 7, 2012 @ 1:31 pm

  58. I was looking at league stats and I noticed something odd- 1b and corner OF have huge negative defensive value every year. How is this possible, since it should be reasonably normalized to position averages? SS and CF have huge positive values (C and 3b somewhat -, 2b neutralish from eye test), which also shouldn’t be possible.

    Comment by Calvin — August 7, 2012 @ 5:16 pm

  59. I’m not really good at all these numbers, but I thought I would propose something: Could the massive disparity in team salary’s be to blame for this?

    Really superb SS/2Bs, especially hitting-wise, are almost always signed by big name teams for huge bucks, for example Jeter in his prime years, A-Rod when he played SS, you get the idea. Because of this, they are almost always set at these positions, with the backup usually being a prospect or a replacement level guy, with exceptions. What I would think would logically follow, then, is that the price for people who put up good WARs, but not enough to match with the elites the Yankee/Red Sox/etc-level teams sign, would logically get paid less due to the decreasing salary of a team: Although a guy who puts up great defensive WAR and decent offensive WAR might be worth as much or only slightly less than bigger named players or otherwise be more useful, because they will not get any looks from the much higher priced teams, teams with smaller salaries will get them for a cheaper price than their expected market value might be.

    How this related to 1B/OF payment is a bit simple and that is that there are not very many great 1B, but there are many OF who sit just above replacement level, but not to the undervalued extent of 2B/SS, possibly due to their nature: Most OF of this type, to my knowledge, hit well but with some issue(High strikeout rate/low walk rate, for example), with poor defense, but which is slightly more forgivable due to there being three outfield positions and the fact that defense is less needed in RF/LF than 2B/SS. Because of the need to sign OFs and, perhaps importantly, the ability for them to be pinch hitters and DHs, means that more higher level and mid tier teams will give them looks, which causes competition and thus drives up the price. Because of this, outfielders who may not be worth the value are overpaid, due to the view on offense and perhaps the percieved value they bring in other aspects. In addition, while I may be wrong as I do not have the numbers, I imagined more backup outfielders are on a team then backup infielders, as a good utility man can often play multiple roles in the infield, while few good RF/LFs can also play CF while still being a backup.

    Bringing it into 1B, the big gap is likely simply there being so few good 1B, as the defense from the position is often not seen as quite as critical and, perhaps importantly, is also much harder to measure than other positions, leading to less hard numbers to use to get undervalued talent. Thus, market value on 1B is very competitive and driven up, while the talent level for lower 1B is not as good and thus not as much of a deal for their value, dropping their usefulness in WAR terms compared to an undervalued 2B/SS. This is probably pretty shaky, I lack numbers.

    Something I think might be interesting is to run the numbers on the gulf between replacement level and the top half or so of players at each position for their price: Are the highest end OFs still overpriced or are they more of a fair market value and are highest tier 2B/SS paid appropriately? I’m thinking OFs may be less overpaid when looking solely at the top end, but I could be wrong.

    In short, apologies for a long rambling thought process from a guy who doesn’t know his stuff, but he figured he’d put it out there: The worst it can do is not be useful, right?

    Comment by Ruki Motomiya — August 7, 2012 @ 6:57 pm

  60. yes- well the difference between an average 1b vs. replacement 1b and average c vs. replacement c.

    Comment by Matt Swartz — August 7, 2012 @ 8:28 pm

  61. This is a silly argument that mostly amounts to “you may have data but I have an anecdote!” Point by point:

    1) The Rays didn’t spend much at a lot of positions. Would you argue their unwillingness to pay for 3B is the reason for succeeding too? Both Carlos Pena and Evan Longoria have been underpaid and important in recent years.

    2) There are some bad contracts to 1B and good contracts to 1B, just like every position.

    3) St. Louis’ run differential has been great because other areas of the team have improved. Are you arguing that their drop in RA/G from 4.3 to 3.9 is somehow because they let Pujols go? They’ve scored more runs too, but because of improvements from Molina, Freese, Holliday, etc. Again, anecdotes don’t prove a thing especially if there’s a hole in your anecdote.

    4) Yes, that’s why it gets a positional adjustment. But I’m arguing that perhaps it should be smaller.

    5) The 6 large contracts for 1B right now are:
    Yankees (Teixeira)
    Tigers (Fielder)
    Red Sox (Gonzalez)
    Phillies (Howard)
    Reds (Votto)
    Angels (Pujols)
    That is six teams that have had a hell of a lot of success in recent years. That list of six teams contains half of the division winners, two likely wild cards, and a team about to miss the playoffs for the first time in five years. It’s not causal per se, but giving me a cute little cherry-picked list of teams not spending money on 1B isn’t proving anything.

    Also, you can’t say 0% chance of Pujols contract not working because you neither know baseball salary inflation nor his projected aging rate. There’s nothing that’s 0% in baseball contracts.

    Comment by Matt Swartz — August 7, 2012 @ 8:40 pm

  62. chel, I only have five years of data so I don’t have that. I did check in one of these articles’ comment sections that the average age at signing was really similar, though, so I do not believe that will be a big factor. Nate Silver had a piece on aging by position at some point too. I don’t remember what his methodology was, or if it was later improved upon, but I don’t think anyone’s found anything like that about 2B aging quickly. Catchers a little, but otherwise no.

    Comment by Matt Swartz — August 7, 2012 @ 8:42 pm

  63. As long as they got Alfonso Soriano, who had plenty of value. Not only that, but ARod averaged 7.5 WAR per year after that in the next four years with the Yankees, which was definitely above his projection (or anybody’s) at the time. The other factor was that Texas failed to build around ARod– it was paper value because it was a transferable contract for which they could get a player of considerable value if they paid a little money down.

    Comment by Matt Swartz — August 7, 2012 @ 8:46 pm

  64. But I like anecdotes!

    Comment by Robbie G. — August 7, 2012 @ 8:47 pm

  65. Did you see Dave Cameron’s article the other? The Angels have had a drop in attendance this year! In truth, most analyses about a star recognition have showed up as minor. Also on average, some schmuck looked at this at THT recently and found big long deals cost about as much as other deals per win: http://www.hardballtimes.com/main/article/big-long-deals/

    Comment by Matt Swartz — August 7, 2012 @ 8:48 pm

  66. There’s some evidence of that but the effect of possible clutchness is probably < $1MM per season: http://www.baseballprospectus.com/article.php?articleid=11861

    Comment by Matt Swartz — August 7, 2012 @ 8:50 pm

  67. My FA values are matched pretty well with the data, and they aren’t projections, so I’m not sure what you mean.

    Comment by Matt Swartz — August 7, 2012 @ 8:55 pm

  68. I’m pretty sure I thought I discovered this before, but it’s fine– it’s because players’ defensive values are their defensive values across all positions, while they are listed with their most common position. So SS & CF are above average at other positions, but when a guy who is an average SS plays 120 games at SS & 30 games at 2B, he’s above average defensively combined. C don’t play other positions much, and 2B & 3B tend to mostly play each others’ positions so they’re neutral too.

    Comment by Matt Swartz — August 7, 2012 @ 8:57 pm

  69. You put a lot of thought into this model, so I don’t want to bust your bubble too bad, but there are a couple things that probably hold it up as written.

    The first is that different teams pay similar prices per WAR for free agents. The smaller market teams spend on fewer WAR at the same price (e.g. they buy 8 WAR on the FA market for $48MM, while bigger teams buy 25 WAR on the FA market for $150MM).

    The second is that WAR is paid for linearly anyway, so weaker players going to some teams wouldn’t change the price per WAR:
    http://www.fangraphs.com/blogs/index.php/a-retrospective-look-at-the-price-of-a-win-part-two/

    Comment by Matt Swartz — August 7, 2012 @ 9:01 pm

  70. Oh, I didn’t know the second. Doesn’t the first only apply if the team wins less than the bigger team, though, unless they manage to win above what their WAR suggests they should?

    I got the impression that the better lower-salary teams did that alongside bringing up prospects which are usually cheaper than normal players and, if the team is doing good, generally do good. It’s irrelevant given the second point, but how far off am I? Pretty casual fan, so I’m not too smart on these things. :P

    Comment by Ruki Motomiya — August 7, 2012 @ 10:16 pm

  71. Well, it is just free agent WAR I’m talking about. Non-free agent WAR on average is maybe 25% of salary.

    Comment by Matt Swartz — August 7, 2012 @ 11:26 pm

  72. Ahh, makes sense. Didn’t realize FG added by normal position instead of where they started each game.

    Comment by Calvin — August 8, 2012 @ 12:19 pm

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Close this window.

0.220 Powered by WordPress