Jason Bartlett For Real? - at driveline mechanics
maybe someone posted this already, and if so, pardon my oversight
2 months ago
benfunke
9 comments
0 recs |
Comments
Okay, I've got questions off this for the sabermetically inclined here
Looking at the summary numbers in the story, Barlett’s is roughly 44.5 runs above replacement that are converted after the PT adjustment stated and the x10 factor for wins to a WAR value of 3.8. With the standard conversion of $4+ MM per win we get the “value” of the player.
But wait a minute. 25 of the 44.5 runs in the equation – about 55% of the value – are in fact the constant value of an AL replacement player? Why is this number included, if the goal is to reach the value EXCEEDING that of a replacement player. In real world salary terms, the value of a replacement player likely doesn’t exceed about $750 K – a reasonably typical average salary for FA bench fill. Using 25 from the run value above, a full time replacement level player is thus worth 25 / 10 X $4.75 MM (or thereabouts as the value constant), or $11.875 MM!
Second, there is the positional adjustment, 7.5 of Bartlett’s 44.5 total run value. I understand the premise of the positional adjustment, but the fact is Bartlett isn’t going to replace or be replaced by a guy from another position – say a C or 1B – but by another SS, who also gets the adjustment. This inherently states that any SS is on average worth .75 (7.5 / 10) X $4.75 MM more than the average player, or roughly $3.6 MM. But there should be no premium when replacing a SS with a SS – which to my feeble old school mind is a highly likely occurance.
To me, these adjustments should differ. All players should hav the inherent value of a replacement player, let’s say $750 K. Take that off the current average major league salary of, what, $2.3 MM, and value above replacement should be calculated as the average win value above replacement these surplus dollars share out to.
I’ll readily admit I haven’t gone back to see how these formulas are derived initially – my bad – so Dave Cameron or Tom Tango or some other sabermetric titan can rip me to shreds if they’d like. But there is value at replacement level – the current formula seems to ascribe all value to above replacement runs / wins. I’m leaving aside the actual calculations of replacement levels, or the adjustment factors that seem to comprise a high percentage of the value in the first place.
To at some extent, and on the surface, this reminds an old schooler like myself of voodoo economics. Again, I’m responsible for educating myself on the philosophical / formulaic underpinnings of the whole thing. And I did once review something akin to that by Cameron that made sense to me. But anybody who wants to enlighten me in a positive way have at it. RJ????
All I know is in the real world, Bartlett’s value is what he would be paid as a free agent. And baseball free agents may have about the most free labor market of just about any in capitalism. And Bartlett wouldn’t make no $20 million per.
by nyyfaninlaaland on Sep 3, 2009 7:40 PM EDT reply actions 0 recs
Postscript
I do recall Cameron’s take did have substantial discussion of WAR deriving from the value of marginal wins, or those that exceed the expected wins from a team comprised completely of replacement players. So there are in my expectation adjustments built into the system relative to some of the concerns expressed. Perhaps someone can provide links – or I’ll just read the DRB glossary.
by nyyfaninlaaland on Sep 3, 2009 7:47 PM EDT up reply actions 0 recs
Hey NYY, just finished reading this, great points
I was kind of surprised by that too. So I went here and saw that the league leader had a replacement of 20.4, which is still about 2 wins (or $9M). I think what this comes down to is how high of a WAR would you expect out of a replacement level player? If 2 is your average player than it must be less than that. Now the league leader in replacement is Scut who plays a lot of SS. I think they should get that boost since we’re trying to “level the playing field.” I don’t think there is anything voo doo about it. A replacement level SS is more likely to hang on at another position than a RL COF. SS is such a tougher position that either the glove or the bat could be attractive at 2B or COF or CF or 3B.
As far as your second point about the positional adjustment, I think it is fine. No matter how good Teix is at 1st, you cannot compare his defensive responsibilities to that of a SS. That needs to be reflected somewhere in the equation. As far as SS to SS, you’re comparing apples to apples, what’s the problem? Jeter’s 6 WAR season is certainly benefiting from his 20 replacement and 5 positional adjustments.
Let’s say a replacement level player on the free market is 1.5 WAR/$7M. This seems high, right? That is the point of the free market. That is why no one can complain about Burrell’s contract (even though he wasn’t the 2 WAR guy he needed to be), because guys pay their dues in a crooked system all their lives so that they can get the big money deals after 6 years in the majors. I don’t really have any problems with WAR, except that it maybe inflates defense a little. I don’t think defense is equal to offense on a 1:1 basis, though they may not be figuring it that way. If that is the case then I apologize. As far as the rest of WAR, I think it’s the best league-wide comparison tool that I have seen.
Embrace Eternity
by Sandy Kazmir on Sep 4, 2009 10:32 AM EDT up reply actions 0 recs
hey
I’m the author of the Bartlett post linked here…
Maybe I’m missing your point, but by definition, a “replacement level layer on the free market” is 0 WAR. “Wins Above Replacement” means wins above that guy. Let’s call him Billy Woomquist. Billy should, in principle, not cost any talent or mooney beyond the league minimum to acquire.
I'm not a sabermetrician, but I do play one at Driveline Mechanics.
Can't get enough of me? Check out my Twitter feed.
by devil_fingers on Sep 4, 2009 11:20 AM EDT up reply actions 0 recs
You're right Devil, second cup of coffee might wake me up
Embrace Eternity
by Sandy Kazmir on Sep 4, 2009 11:34 AM EDT up reply actions 0 recs
Great read by the way
I’ve put together one using the Trade Value Calculator here that uses estimated WAR. Perhaps you could read it and give me some pointers? LINK Or maybe you could update with the figures you arrived at.
Embrace Eternity
by Sandy Kazmir on Sep 4, 2009 11:37 AM EDT up reply actions 0 recs
I am not an expert...
But I think it’s because the run values for offense and defense are calculated on a scale with 0 being average. Adding 25 runs for “replacement level” really means readjusting the scale so 0 is replacement level (thus 25 runs is average). This does not mean a replacement level player is worth 11.875 mil, because to be replacement level, they would have to rate -25 on offense + defense. But by adding the replacement level adjustment they are 0 instead of -25, thus the stat is WAR instead of WAA (wins above average). An average player is worth 11.875 mil as a free agent (ignoring positional adjustments), because they would rate 0 on offense + defense.
Also, the positional adjustment is so you can compare across positions. When comparing two SS, they get the same positional adjustment, so it cancels out anyway. But a SS gets a much bigger positional adjustment than a RF, so to be equal in value the RF has to be that much better on offense+defense.
by ChiBurbRaysFan on Sep 4, 2009 12:47 AM EDT up reply actions 0 recs
Wow, is this Mandarin?
I’m either depressed, discouraged or annoyed. I’m not sure which. I had no idea I knew so little about a entire realm of baseball.
by terp12 on Sep 4, 2009 10:30 AM EDT reply actions 0 recs
just enjoy the game
we nerds have our own weird ways of extending our enjoyment of sports
(winky face)
I'm not a sabermetrician, but I do play one at Driveline Mechanics.
Can't get enough of me? Check out my Twitter feed.
by devil_fingers on Sep 4, 2009 11:20 AM EDT up reply actions 0 recs


















