Bang for the Yanks’ buck

A Joba musing
The RAB Radio Show - December 30, 2008 - Episode 8

Fangraphs is great, and continues to get better. While B-Ref is still the undisputed king of … well … baseball reference sites, Fangraphs has established itself as the clear #2 thanks to its assortment of advanced statistics, including those pertaining to plate discipline, win probability and defense. Last week they added a new replacement level section for hitters, and included in it is a cool little feature that represents how much money the player is worth based on how many wins they provide above a replacement player.

Essentially, this value is determined by adding up how many runs the player adds over one of those mythical replacement level players in terms of hitting and defense (runs saved, in this case). The numbers are adjusted for park effects and position, so a first basemen in a small park needs to do more to provide positive value than, say, a shortstop in a big park. Here’s a quick example:

Ryan Howard
Batting: +22.3 runs
Fielding: +0.8 runs
Position Adjustment: -12.6 runs
Replacement Level: +23.3 runs
Total: +33.9 runs

Howard produced 22.3 offensive runs above replacement level in 2008, and his defense was ever so slighty above RL at 0.8 runs. His position works against him, taking away 12.6 runs. Add it all up (including a replacement level first basemen’s output to get his total contribution), and Ryan Howard was worth 33.9 runs last year. Ten runs approximately equals one win, so he was worth 3.39 wins. 2008 wins were worth $4.5M (according to the Fangraphs guys), so Ryan Howard provided the Phils’ $15.2M worth of value, $5.2M more than his actual salary. Now take a look at another player for a quick comparison:

JJ Hardy
Batting: +16.3 runs
Fielding: +11.6 runs
Position Adjustment: +6.8 runs
Replacement Level: +21.0 runs
Total: +55.6 runs
Value: $25.0M

So because he provided well above average offense and defense at an important (actually the most important) position, Hardy’s value to his team was nearly 65% more than Howard’s. That might be hard to grasp because of Howard’s gaudy HR and RBI totals, but the whole is greater than the sum of the parts in baseball. Check out Albert Pujols; dude was worth $40.5M last year, or 291.4% more than his actual salary. Insane.

Using this cool little analysis, we can determine which Yankees’ players provided the most bang for the buck last season. Data is available for hitters only, so we won’t be able to get an idea of how much value the Yanks got out of their arms in relation to their salary. Perhaps another time.

A really big table is after the jump.

Player Hitting Fielding Total Value Salary
Alex Rodriguez 40.5 -0.1 61.5 $27.6M $27M
Johnny Damon 21.6 2.7 38.9 $17.4M $13M
Derek Jeter 7.0 -0.4 35.5 $15.9M $20M
Jason Giambi 21.5 -1.8 26.5 $11.9M $21M
Xavier Nady 9.2 0.4 14.8 $6.7M $1.36M
Bobby Abreu 21.0 -25.2 11.2 $5.0M $16M
Hideki Matsui 5.5 -1.6 8.0 $3.6M $13M
Jorge Posada 1.5 -0.6 7.5 $3.4M $13.1M
Brett Gardner -5.7 7.2 5.7 $2.5M $0.12M
Robinson Cano -12.3 -7.3 4.0 $1.8M $3M
Cody Ransom 4.8 -1.6 3.9 $1.8M $0.097M
Ivan Rodriguez -1.6 ? 3.74 $1.7M $3.06M
Chad Moeller -3.7 ? 2.0 $0.9M $.3M
Juan Miranda 1.2 0.2 1.5 $0.7M $0.39M
Alberto Gonzalez -1.3 0.2 1.5 $0.7M $0.101M
Jose Molina -18.1 ? -0.7 -$0.3M $1.75M
Chris Stewart -0.8 ? -0.6 -$0.3M $0.002M
Frankie Cervelli -1.4 ? -1.0 -$0.4M $0.065M
Justin Christian -0.7 -0.9 -1.4 -$0.6M $0.083M
Melky Cabrera -17.3 -0.5 -1.5 -$0.7M $0.4612M
Wilson Betemit -3.8 -3.8 -3.2 -$1.4M $1.165M
Shelley Duncan -4.5 -0.7 -4.7 -$2.1M $0.3983M
Morgan Ensberg -6.8 -2.1 -6.5 -$2.9M $1.75M
Richie Sexson -0.3 -0.9 -0.8 -$3.8M $0.086M
Total 55.5 -36.8 205.8 $89.1M $137.2885M

All salary data came from (where else?) Cot’s, and anything that needs to be pro-rated for partial seasons has been (approximately). Note that the Mariners were still on the hook for Richie Sexson’s full salary despite cutting him loose, the Yanks were only responsible for a pro-rated portion of the league minimum.

As we all already knew, long term injuries to Hideki Matsui & Jorge Posada hurt the club’s output, and it’s evident here. Posada’s value from ’02-’07 suggest he would have earned just about all of his salary, although Matsui’s value would have been about two-thirds of his salary. A-Rod‘s value also would have been a bit gaudier if not for a three week DL stint.

It’s interesting to see that Bobby Abreu & Jason Giambi – widely considered the club’s second & third best hitters last season – were the two most overpaid players on the team. Nady & Damon more than earned their salary, and Brett Gardner’s defense made him roughly $3.2M more valuable than Melky. A typical Robbie Cano year would have done wonders; he was valued at $12.7M & $17M in ’06 & ’07, respectively. Morgan Ensberg & Richie Sexson .. yikes.

We all knew the Yanks were overpaying, but now we can get an idea of just how much. The $89.1M in value received is just 64.9% of their actual salary payout; compare that to Hanley Ramirez & Dan Uggla, whose combined value ($53.0M) was roughly 242.7% of the Marlins’ Opening Day payoll. Mark Teixeira will certainly help balance the value-salary ratio, he was valued at $30.5M last season.

Anytime you spend upwards of $200M on the 40-man ML roster, there’s bound to be a great deal of inefficiency, but a 64.9% return on investment is just ghastly. Healthy comebacks from Posada & Matsui combined with the departures of Giambi & Abreu will go a long way towards restoring some efficiency to the Yanks’ books.

A Joba musing
The RAB Radio Show - December 30, 2008 - Episode 8
  • Eric

    Wow, Abreu’s defense really subtracted from his value.

    • Chip

      Wow, and people complain about Jeter’s defense. I wonder if any teams looking to sign him ever look at these advanced defensive metrics to see just how bad he is with the glove.

  • Reggie C.

    I thought the knock on Hardy was his glove. Weird.

    • Mike Pop

      Hardy is awesome. There is just so many great SS’s in the NL he gets overlooked. But he can mash.

  • KW

    Interesting, losing Abreu and Giambi results in a net deficit of -$27mm. Matsui and Posada contributing like they can decreases the deficit to -$7mm and say everyone including Tex replicates their 2008, the yanks’ll actually be getting more value than they’re paying out.

  • Manimal

    Why is there no fielding stats for Molina?

    • Mike A.

      I don’t know. Fangraphs didn’t have any fielding stats for the backup catcher types – Molina, Moeller, Pudge, etc.

    • ????

      I was wondering that as well. Molina percentage of throw outs was the best in the MLB.

    • KW

      I think it’s because the catcher fielding stats don’t translate well over to replacement value (ie. runners caught, etc)

    • CB

      Fangraphs doesn’t provide fielding data for catchers. There’s simply no good way to measure catcher defense at this time. UZR is largely meaningless for catchers. So Fangraphs hasn’t included UZR data for catchers.

      This also has a major impact on the Win Value for catchers as listed on Fangraphs. It doesn’t include any defensive value in it other than positional adjustment. They basically concede that you have to just estimate how many wins a catcher provides for his behind the plate work through subjective estimates.That makes things confusing. Rob Neyer on his blog this morning commented on how Sizemore was first in the AL in win value at +7 and Pedroia second at 6.6. He said Mauer was 4th at 5.7.

      That’s not really true as Mauer’s 5.7 gives him no credit for his outstanding defense/handling of pitchers (which considered by most to be not only far above average but outstanding).

      To assess Mauer’s true net worth you’d have to add at least 1-2 wins for his defense as a subjective estimate (probably 2). Mauer was more than likely the top win value player in the AL last year and should have been MVP.

      • steve (different one)

        yeah, and he probably should have been MVP in 2006 too, Jeter’s awesome year withstanding.

        i have no idea what Mauer has to do to win an MVP…

        • Mike Pop

          Save 3 kids from a burning metrodome.

          • 27 this year

            i think he already did that. However, one was left behind and they hated him for that.

            • Mike Pop

              Ya the voters were pissed because he only batter 750 and not 1000.

  • CB

    Players who are above average both offensively and defensively often produce value that far surpasses their perceived value. This is especially true of guys that don’t have flashy offensive numbers but are above average offensively and very good defensively.

    JJ Hardy is a great example of this. People don’t realize how good he is. They’ll look at his BA/HR/RBI’s or at his OPS+ and not take into consideration his position and glove. He would be a terrific replacement for Jeter at SS after 2010 (believe Hardy is a free agent then).

    Another guy who falls into this category is Brian Giles. How many people think Giles is one of the best players in the game? Last year he was one of the top 20 players in the whole game.

    One of the main reasons not to trade Cano is that he also has the ability to be one of these kinds of two way players that produce tremendous value by being both above average offensively and defensively at a difficult to play position. At the plate he could obviously be well above average if things click. On top of that he could be a +10 run defender at 2b. With him it’s consistency.

    I could also see Austin Jackson becoming this kind of player and being very underrated because of it.

    • Matt

      “Another guy who falls into this category is Brian Giles. How many people think Giles is one of the best players in the game? Last year he was one of the top 20 players in the whole game.”

      I fuckin’ love Giles. Great player, massively underrated.

      • steve (different one)

        Giles is a great player. unfortunately he’s a scumbag.

        • E-ROC

          It was only a love tap………

        • Mike Pop

          You think he did roids ? Didnt he just lose all his power or is that because of Petco

          • steve (different one)


            It’s impossible to watch the video without getting chills. Thanks to a security camera, we see Giles walk into a bar and approach Olvera. In a stunning few seconds, he appears to pull her hair and slap or shove her.

            • CB

              That’s just horrendous. I just wanted to clarify my above post – I was in no way at all advocating for Giles as a player that I’d want on the yankees – he was just an example of how the numbers work for some players on the field.

              I hadn’t heard anything about his violent streak until his ex-girlfriend filed that lawsuit. It’s just disgusting. I didn’t know there was any videotape like this out there. If his ex-girlfriend’s allegations are true, and this videotape is pretty damning, he should go to jail in addition to losing $10M. It shouldn’t just be a civil case. How does that guy live with himself?

              • steve (different one)

                That’s just horrendous. I just wanted to clarify my above post – I was in no way at all advocating for Giles as a player that I’d want on the yankees

                oh, i know. sorry if i implied that you did.

                this was just fresh on my mind as i just read the article.

            • E-ROC

              Hopefully, Olvera punishes him to the fullest extent of the law. Lets see what Bud Selig does once the lawsuit is done or before.

  • KW

    One Last thing, Ellsbury was worth $15 mil or so according to FanGraphs almost completely due to his fielding (17.5 runs). Goes to show we might be ok with Gardner manning center, even if he can’t hit a lick (+7.5 runs in very limited duty).

    • radnom

      No, it goes to show that these equations weigh defense waaay too heavily.

      • KW

        Or it shows that preventing runs from scoring is just as important as scoring them…

        • Stephen

          It makes sense logically and I believe it, but it’s hard to straight up look at Adam Everett and Carlos Lee’s 2006 seasons and saying they were almost equally valuable. What I mean to say is that no hit-great fielding Shortstops can be nearly as valuable as Poor Fielding-Great hitting leftfielders would actually be a revolution in thinking about baseball players.

          • KW

            I hear ya, and I didn’t see that coming. I looked it up and, yup, they were close to equivalent in value. To be fair though, Lee was pretty terrible in the field, while everett had a terrific season in the field. I guess it just puts an emphasis on positional value and adjustments.

            Using that lens, Carlos Lee was a better hitter than the average LF, but an atrocious fielder compared to the average LF. Everett was atrocious compared to average SS’s, but had a spectacular glove in comparison. I can see how that can make sense. I just have to look at Bobby Abreu – average stats for a RF, but truly horrible at fielding his position. I think there, it’s tough to argue that that isn’t true!

            • Stephen

              I think traditionalist-minded people, in the end, are going to have the biggest problem with adjusting a player’s value for his position.
              I love sabermetrics, but as I said before, I even have a problem saying Adam Everett (with a .276 wOBA) and Carlos Lee (with a .379 wOBA) were nearly equally in 2006 because of the disparity in defense.

        • radnom

          Or it shows that preventing runs from scoring is just as important as scoring them…

          No. You can replicate Elsbury’s value for less than $15 million. All other strong defensive players are also overvalued.
          You could argue that ML teams don’t value defense enough, but regardless this is a deviation from reality by the equation.

        • RollingWave

          unfortunately, if you even look at a few players defensive numbers, you’ll realize that defensive value’s reliability is far lower than offesnive onces. so yes, while in one season you might be able to save up enough runs to make up for your bad bat, in the longer run, your going to have (plenty) of seasons where you don’t save those runs… and your bat is still crap. which makes you a utter blackhole.

          so yes, in a single season, a no hit all field guy could be equal to a all hit no field type, but in the longer run (assuming similar age / health of course) your probably better off with the all hit no field guy.

  • Matt

    I just finished writing a similar blog on this, but talking about the acquisitions in the off season:

    /tooting my own horn.

  • Tom Zig

    What are the yanks gonna do about the leadoff spot in 2010?

    We don’t have anyone in the farm system (at least at this moment, it is how it appears) that would be successful as a lead off guy.

    Scenario A:
    Sign Damon to a 2 year extension.

    Scenario B:
    Let Damon walk, take the draft pick(s)
    Push Jeter into lead off spot.
    Who hits 2nd?

    Scenario C:
    Let Damon walk
    Leave Jeter in 2nd spot
    Who hits lead off?

    Brian Roberts would be the ideal candidate for B and C. But where would he play?

    Well maybe you could move Cano to the outfield or move Roberts to the outfield.
    or move Jeter off short, and have Roberts play there. who knows.

    Your thoughts?

    • Mike Pop

      Brett the Jet Gardner

      • Should be working

        Agree. He’s got a good eye, great speed. If the bat works, its a good spot. Or Jeter 1 and Swisher 2. I think i’d rather hope that Gardner can hit by then but Jeter-Swish isnt the end of the world.

      • D.B.H.O.F. p.k.a The Last Don

        I really hope this kid gets the at bats to prove who or what he really is. With the power and overall hitting coming from all over on the Yanks I could take a guy with very little power in CF if he is top rate in the field and has speed and knows how to run the bases (and gets on them enough for all that to matter)

    • Reggie C.

      I wonder if its possible for Damon to play himself to a 2 year contract. He seems to really love playing here and his bat hasn’t slowed. However, I just don’t think its a smart idea to view him as a full time OF beyond this season. His range and arm aren’t going to get better this season.

      Should Damon match his ’08, then its not a heinous idea to extend him a couple more years, but he’s likely going to split time b/w DH and LF.

      • steve (different one)

        i have nothing to back this up, but i’d be shocked if Damon were on the 2010 Yankees.

        they seem committed to getting younger.

    • steve (different one)

      just let Jeter lead off.

      • D.B.H.O.F. p.k.a The Last Don

        Big Jeter fan, and I know he could lead off, but I think his best days at that spot are behind him.

        With that said no way I give Damon a two year deal unless its a discount, a deep discount.

    • D.B.H.O.F. p.k.a The Last Don

      Brian Roberts. Yuk. If I want a pint size guy on my team I prefer one who can actually bunt, and does not use the whole “I did steroids once” gag. That gag is as weak or possibly weaker than Roidger claiming no use at all.

  • monkeypants

    Interesting how not overpaid (relatively speaking) Jeter was, despite his stinky year at the plate. It will be further interesting to see how the last couple years of his big contract play out (and pay out), depending on the degree to which his offense and/or defense slides.

    • Mike A.

      It’s funny, Fangraphs data goes back to 2002, and based on the $15.9M value, 2008 was his second best season in that time. It’s all because he improved his defense from atrocious to merely below average.

      Check it out:

      • monkeypants

        Yep. But even then, when you look at his overall value since 2002 (Value = $104.0; salary = $132.2; value/salary = 79%), Jeter has been a relatively good investment. Marque players will almost always be overpaid, of course, but Jeter has actually lived up to his contract, at least broadly speaking.

  • dan

    Mike, your “translation” is a little off. The batting runs is above average. The replacement level adjustment is how many runs better than replacement an average player is. And defense is also above average (technically), but in the case of defense, average = replacement level.

    • Mike A.

      Eh, close enough. ;)

      • Tom Zig

        Brett Gardner, if he could hit like .275-.300, I’d be very comfortable with him in the leadoff spot. I dunno, I just really like Brian Roberts though lol

    • CB

      And defense is also above average (technically), but in the case of defense, average = replacement level.

      That’s not entirely accurate either with respect to Fangraph’s methodology for win value. It is true that UZR compares a player’s defensive performance to an average player at his position. However, that’s not what “replacement” is truly set at defensively. In fact, that exactly the reason why they use positional adjustment to determine the overall defensive value of a player.

      If replacement=average than a +5 SS would equal a +5 1b. But in order to create a more valid defensive replacement value than “average” positional adjustments were created – so in this sense “replacement” for a 1b is set at -12.5 and for SS it’s set at -7.5.

      You can disagree with the notion of positional adjustment, but in the methods advocated by Fangraphs/ Tom Tango, average is not truly equal to “replacement.” Average +/- positional adjustment for defense is what serves as the analog to “replacement.”

      • Mike A.

        Yeah, what he said.

      • E-ROC

        Um…layman’s terms. Baseball sabermetrics are tough to follow.

        • CB

          Sorry. I was responding to a sabermetrics argument with a sabermetrics response.

          Basically when it comes to offense players can be judged against other players playing their positions. Replacement level players are roughly a “AAAA” player.

          But defensively it makes less sense to judge players only against other guys who player their position. The reason why is that not all positions on the field are of equal value defensively. Is a SS who was +5 runs better than an average SS as valuable as a 1b who was +5 runs better than an average 1b? Many people say no – the SS is much more valuable than the 1b and to call them both “+5” defensive players is misleading. The SS has additional value given the difficulty of his position and the scarcity of players able to play that position.

          This is essentially what Fangraphs argues in their win value statistic.

          Catcher is a lot more valuable than a 1b and it’s harder to find a player to fill that position adequately. So just by playing catcher, the player has a defensive value. They give all catchers starting off value for defense of +12.5 runs. 1b is the opposite – they give them a starting value of -12.5 runs. This is what’s referred to as positional adjustment.

          So each position defensively has a different “replacment” value that depends on both the average performance at that position and the “positional adjustment” needed to fully asses the value of that position defensively.

          • E-ROC

            Thanx homie.

          • dan

            I agree with that, I think we’re talking about two slightly different things. Some people think that because the average hitter is 20 runs better than replacement level in 600 PAs, then you also add some runs above replacement for his fielding above replacement… And then add the positional adjustment. That, as you probably know, is wrong, and is what I was saying not to do.

  • Zack

    I love this, but i have a very hard time buying the value given to defense.

    Yes, its underrated, and yes, it plays a major part in winning. But as some above comments already said, in no way was Ellsbury worth $15M last season. That would make him basically the third most valuable player had he played for the Yankees last season. And 100% because of his defense.

    I’m sorry, but that is absurd. Yes, the guy plays a mean CF. But does that mean he’s worth that much? The combination of adjusting for position and the weight placed on defense means that 2008 Ellsbury>2006 Tex. I call BS.

    I think its very important to weight defense more than it has in the past, but this logic is a slippery slope. I wish their data on value went back further, because it has 2002 Bernie Williams as> Ellsbury as well, and again, that’s a load of hooey.

    • KW

      Hmm…Tex OPS’d about .890 in 2006, which while terrific, I guess it wasn’t significantly better than the average 1b. I think offense that year was also more prolific than in recent years, but I could be wrong. Giambi had an OPS of .970 that year, as a comparison. Tex’s defense also wasn’t nearly as good as it was this year and I believe it was slightly below average. Ellsbury’s bat was average, but on average, CFers suck at hitting much more than 1bs. And, the argument goes, his CF defense was sublime. All in all, I think the conclusion is that Ellsbury>Tex because of the positional adjustments.

      Bernie was a pretty darn good offensive player in 2002, how would that be a misconception that he’s better than Ellsbury? Or do you mean Tex? If it’s Tex, that’s probably due to the fact that Bernie had a better OPS playing CF.

      • zack

        I got that backward. It should read that Ellsbury was MORE valuable than 2002 Bernie.

  • Mike @ NYYU

    If I understand this correctly, it says that Molina was a MINUS in value for the Yankees last season.

    Molina was worth every penny he was paid and deserved a bonus on top of it.

    There is no stat that can measure how a catcher handles a pitching staff and the Yanks pitchers last season were quite a handful.

    Would Mussina have won 20 with Stewart/Moeller?

    Throw in veteran experience and “presence” in the club house, these “replacement level” type stats have no value to me.

    • dan

      If you read the introduction for these stats (see below), you’ll see that they explicitly state the full value of catchers (whether positive or negative) is now captured. They say you can add in a run value for game-calling, leadership, etc. if you want, and adjust accordingly. All catchers are assumed to be average defensively because the catcher fielding statistics aren’t so good yet.

    • D.B.H.O.F. p.k.a The Last Don

      “Would Mussina have won 20 with Stewart/Moeller?”

      The answer to that is NO. You can also ad Posada to that list and you still get this answer:


      Molina is the main reason we had a respectable season at all. Think of all the starts rookie, washed up, or bum pitchers got for the Yanks last year. I know Molina’s bat disappeared very early in the season but he ability to get the pitchers to pitch well was very very valuable and I do not buy any stat could ever show you that. What I really do not get is why “stat heads” feel the need to have a stat for everything, even if it lacks substance or ability to show what happened.

      • Mike @ NYYU

        I agree completely. Also the idea that

        They say you can add in a run value for game-calling, leadership, etc. if you want, and adjust accordingly.

        as nothing more than a side note, instead of a real value, is an insult to all quality game callers.

        Baseball is best analyzed with the head and heart, not spreadsheets and computers.

  • ko

    Batting lends itself to this type of analysis. Fielding doesn’t. Throwing out these questionable fielding values makes the analysis more real life and closer to what people actually value the players at.
    Also, where do they get these $ values from? Even if you took a team’s gross income from all sources, you couldn’t extrapolate a player’s value from that based on runs produced. For example, with the Yankees, there’s an intrinsic value to the franchise that figures into the team’s income no matter how many runs are produced. For example, I don’t see dividing A-Rod’s runs produced into the team’s gross income to come up with a dollar value for him for the year.

    • D.B.H.O.F. p.k.a The Last Don

      “Also, where do they get these $ values from?” I do not know the answer to that one, but just the fact that there is one is very absurd. I am assuming that the dollar value is suppose to be what they are worth to the league and not just one team. No way you could know what that value really is or should or could be. Just another silly made up thing by guys who play with stats like lil girls play with Barbies. I really think some folks would rather play the games on paper than play them in real life.

    • ChrisS

      Yeah, the average salary for some players is skewed by what the Yankees will pay versus what other teams can pay. That distorts the value category, I would think. The Yankees now have 4 of the 5 highest paid players in baseball and have bid against themselves in the past. What would they give Mauer if they already had him under contract and wanted to lock him up versus what would the Yankees offer him if he magically became a free agent?

      A player’s value is determined by market and available talent at that position.

      All that said, I think the defensive portion of this is pretty off. Defensive metrics are shaky enough as it is.

  • ryan

    Albaladejo is on fire in winterball 19 scoreless in closer role. Does he get the nod over Robertson and melancon for the open bp spot?

  • Pingback: Open Thread: Tex vs Giambi | River Avenue Blues

  • Pingback: Bang for the Buck: Pitchers | River Avenue Blues