Jump to content


Yo! You're not logged in. Why am I seeing this ad?

Photo

ABCs of FA Pitching (2004/05 review)


This topic has been archived. This means that you cannot reply to this topic.
7 replies to this topic

#1 philly sox fan


  • SoSH Member


  • 9748 posts

Posted 05 November 2005 - 10:45 PM

One of the notable characteristics of last winter’s free agent season was the plethora of copycat 3 yr/21-25M contracts that followed the Mets early signing of Kris Benson. Seemingly any solid, healthy mid-rotation starter was in line for a contract of at least three years (and no more than four) for at least 7M per year (and all the way up to 10M for the Yankees). It’s easy to use those contract parameters as defining a class of “B” free agent starters.

Behind that group there was another larger somewhat motley group of pitchers who could only get 1-2 year commitments due to any number of questions about their health or age or just the fact that they aren’t very good. Just from the type of contracts they signed you can easily categorize these pitchers as “C” free agent starters.

And, of course, there was a tiny select group of “A” pitchers whose names you may recognize. I thought it would be interesting to look at how each of those groups from the 2004/05 FA class performed in 2005 and then to see how this year’s class of FA pitchers might segregate into those same categories.

Contract totals and the list of potential FAs are both taken from Cot’s Baseball Contracts.

http://mlbcontracts.blogspot.com/

I used a straight AAV for each pitchers 2005 salary except for Woody Williams and David Wells whose contracts were heavily incentive based. I used their actual 2005 earnings. I think AAV of a long-term contract is generally a better baseline than actual 2005 cost.

Performance was measured by VORP because it’s easily available, context adjusted and likely more or less as good as any other run based metric.

I used “$/10 VORP” as a simple measure cost efficiency. It’s the cost in millions for roughly 1 win unit of VORP.

“BPVal” is a simple market valuation recently developed by Nate Silver at BP. The formula is: Salary = (WARP^2 * $212,730) + (WARP * $402,530).

“Diff” is the difference between “BPVal” and AAV. Positive numbers indicate the pitcher produced more value than his cost.

Part I: A look back at the 2004/2005 FA pitching class

A is for didn’t that bum used to pitch for the Sox?
Player       Type   Years    Total    AAV     VORP   $/10 VORP   BPVal     Diff
R Clemens     A       1      18.0    18.0     80.6      2.23     27.20     9.20 
P Martinez    A       4      53.0    13.25    66.1      2.00     16.46     3.21
                       Total: 71.0    31.25   146.7      2.13     43.65    12.40

There were two familiar ace level pitchers on the market. Clemens is a unique case because he has, apparently, restricted himself to one year commitments from a single team. Both pitchers had excellent seasons both in terms of raw production and in terms of their cost efficiency. Additionally, each pitcher was significantly undervalued for 2005. Silver’s valuation methodology is pretty shaky at the very high end production so Clemens’ “BPVal” is probably much less accurate than it is for the other pitchers in the study, but the direction is certainly correct. Perhaps he was worth 20M instead of 27M. Either way the Astros were happy to have him.

I think it’s important to point out that both pitchers had financial upside in addition to on field production upside. The salaries of star players are so high that I think we get caught up in the belief that at best star players can earn their money with great production or they are financial problems. In fact, truly great players can have a fair amount of financial upside in addition to their obvious great on field production.

B is for we don’t need no steenkin budget
Player       Type   Years    Total    AAV     VORP   $/10 VORP   BPVal     Diff
B Radke       B       2      18.0     9.0     31.9      2.82      7.85    -1.15
J Lieber      B       3      21.0     7.0     28.9      2.42      7.59     0.59
D Lowe        B       4      36.0     9.0     25.2      3.57      8.38    -0.62
K Benson      B       3      22.5     7.5     22.5      3.33      5.66    -1.84
M Clement     B       3      25.5     8.5     21.6      3.94      6.83    -1.67

O Perez       B       3      24.0     8.0      8.6      9.30      1.66    -6.34
C Pavano      B       4      39.95    9.99    -1.3      neg       0.62    -9.37
J Wright      B       3      21.0     7.0     -9.8      neg       0.04    -6.96
R Ortiz       B       4      33.0     8.25   -20.4      neg       0.25    -8.00
E Milton      B       3      25.5     8.5    -25.0      neg       0.04    -8.46
                     Total:  266.45   82.74    82.2     10.07     38.9    -43.82
                     Good:   123.0    41.0    130.1      3.15     36.31    -4.69
                     Bad:    143.5    41.74   -47.9      neg       2.61   -39.13

I included Radke in this group despite the fact that he only received a two year contract because the shorter length of his deal was apparently his decision. In terms of talent and AAV, he clearly belongs in this group of “B” pitchers. He was arguably the most desirable pitcher on the market after Clemens and Martinez.

There were ten pitchers who received “B” pitcher contracts. However, the last three pitchers – Wright, Ortiz, Milton – probably should not have received a “B” pitcher contract. The fact that a dumb team might give a pitcher a contract he didn’t deserve makes it difficult to just lump all of these pitchers together and call them “B” pitchers. The market didn’t define those pitchers as “B” quality, three specific dumb teams did.

Five pitchers had good, solid years. They produced at least 20 VORP and were good mid-rotation starters. Their cost efficiency ranged from 2.4-3.9M per 10 VORP which is noticeably higher than the 2.1M figure for the A level pitchers. And that makes sense because these pitchers were generally paid 50-67% of Martinez “A” contract and only generated 33-50% as much production. That’s a tough way to be cost efficient or a good value and that’s reflected in the “Diff” column where only Lieber managed a small positive value. In general, these pitches were overpaid by roughly 1M in good seasons. More so than the “A” pitchers it’s the “B” level pitchers who don’t have much room for a financial upside.

Five of these pitchers were busts including two, Perez and Pavano, who were generally considered to be deserving of a “B” contract. Pavano didn’t deserve the specific extravagant contract he received, but at something closer to the group average of 3yrs/24M he would have been an ok signing and yet still been a major bust.

The ten pitchers received contacts that totaled 266M and paid out nearly 83M in 2005. The group only produced 82 VORP which translates to a terrible cost efficiency of 10M/10 VORP. Their “BPVal” of 38.9M was just 47% of their AAV. I don’t trust the “BPVal” for Clemens, but using that figure the “A” pitchers produced a “BPVal” that was 140% of their AAV. Star players have more financial upside than you think and solid players have plenty of financial downside even though their contracts are much smaller.

If you ignore the busts completely you find that the successful “B” pitchers produced a “BPVal” that was 89% of their AAV. That’s close enough to a break even point. The five busts produced just 6% of their AAV. And they’re still owed 100M. Ouch.

C is for carbon and crap
Player       Type   Years    Total    AAV     VORP   $/10 VORP   BPVal     Diff
K Millwood    C       1       7.0     7.0     52.3      1.34     11.29     4.29
E Loaiza      C       1       2.9     2.9     42.1      0.69      8.38     5.48
P Byrd        C       1       5.0     5.0     37.3      1.34      8.65     3.65
D Wells       C       2        *      9.0     24.1      3.73      7.85    -1.15
M Morris      C       1       2.5     2.5     18.9      1.32      2.79     0.29

C Lidle       C       2       6.3     3.15    10.0      3.15      3.47     0.32
W Williams    C       1+       +      7.0      7.4      9.46      2.19    -4.81
S Estes       C       1       2.5     2.5      7.0      3.57      2.64     0.14
O Hernandez   C       2       8.0     4.0      6.1      6.56      2.34    -1.66
W Miller      C       1       1.5     1.5      5.9      2.54      1.53     0.03
I Valdez      C       1       1.5     1.5     -0.7      neg       0.25    -1.25
P Wilson      C       2       8.2     4.1    -12.2      neg      -0.13    -4.23
A Leiter      C       1       8.0     8.0    -12.9      neg       0.39    -7.61
J Lima        C       1       2.5     2.5    -30.8      neg       0.39    -2.11
                     Total:   80.9    60.65   154.5      3.93     52.02    -8.63
                     Good:    30.4    26.4    174.7      1.51     38.96    12.56    
                     Bad:     50.5    34.25   -20.2      neg      13.06   -21.19

*Wells: original value of 8-18M, after 2005 that is now 13-18M. I used 13M as the total
+Williams: original value of 3.5-14.5M, after 2005 that is now 12-14.5M, I used 12M as the total

The first thing to note is that the top 3 pitchers all outperformed every “B” pitcher by VORP. The five best pitchers from this FA class were the two “A” pitchers and the three best “C” pitchers.

Millwood, Loaiza, Wells and Morris were so good and/or so cheap that their cost efficiency was excellent at 0.69-1.3M. Wells’ performance puts him in the “good” group of “C” pitchers, be he was also the most expensive and as a result his cost efficiency and value were more in line with the good “B” pitchers.

The “bad” group includes a few decent values in Lidle, Estes and Miller (though if he earned any of his 3M in incentives, then not so much), but it doesn’t really help your team all that much to “win” one of these cheap contracts and end up with a pitcher who can barely clear 10 VORP for the year. Because these contracts were generally on the cheap side the busts were on the small side except for Leiter, Wilson and Williams.
I think expressing the “BPVal” as a percent of AAV is a decent quick way at looking the relative financial success of these classes of pitcher. Here’s a quick table looking at each class by %AAV. I split the “B” and “C” groups into “tot” for total and “good” and “bad” in the same way I did above.

         %AAV
A         140

Btot       47
Bgood      89
Bbad        6

Ctot       86
Cgood     148
Cbad       38

Total:     77

No big surprises here. Successful “A” pitchers are tremendous values. “C” pitchers in general are likely to be decent values because they’re so cheap. “B” pitchers in general are pretty poor values, though if you can pick the right ones you can come pretty close to breaking even.

Overall, pitchers only returned 77% of their AAV. I ended up going back and doing these same kinds of calculations for relievers and position players and both of those groups were higher. I’ll hopefully post those results soon-ish.

In total, MLB committed 418M to these pitchers. Their AAV is 175M. They produced 383 VORP in 2005 for a pretty poor “$/10 VORP” of 4.55M. They had a “BPVal” of 135M which translates to a year 1 loss of value of 40M. These pitchers are owed an additional 367M over the next three years. If they maintain a return of 77%, then they’ll produce another 84M in wasted payroll in that time. I don’t know for sure, but I assume years 2-4 for pitchers are worse than year 1 so it could potentially be much more than that.

And actually, let’s split these pitchers into the one year and multiyear contracts.

Player       Type   Years    Total    AAV     VORP   $/10 VORP   BPVal     Diff
R Clemens     A       1      18.0    18.0     80.6      2.23     27.20     9.20 
K Millwood    C       1       7.0     7.0     52.3      1.34     11.29     4.29
E Loaiza      C       1       2.9     2.9     42.1      0.69      8.38     5.48
P Byrd        C       1       5.0     5.0     37.3      1.34      8.65     3.65
M Morris      C       1       2.5     2.5     18.9      1.32      2.79     0.29
S Estes       C       1       2.5     2.5      7.0      3.57      2.64     0.14
W Miller      C       1       1.5     1.5      5.9      2.54      1.53     0.03
I Valdez      C       1       1.5     1.5     -0.7      neg       0.25    -1.25
A Leiter      C       1       8.0     8.0    -12.9      neg       0.39    -7.61
J Lima        C       1       2.5     2.5    -30.8      neg       0.39    -2.11
                              51.4    51.4    199.7      2.57     63.5     12.1

These pitchers produced a “BPVal” that was 124% of their AAV. A big chunk of that is Clemens, but even without him the total would be 109% of AAV. If you can pick the right “C” pitchers, there’s a lot of potential value there.

Player       Type   Years    Total    AAV     VORP   $/10 VORP   BPVal     Diff
P Martinez    A       4      53.0    13.25    66.1      2.00     16.46     3.21
B Radke       B       2      18.0     9.0     31.9      2.82      7.85    -1.15
J Lieber      B       3      21.0     7.0     28.9      2.42      7.59     0.59
D Lowe        B       4      36.0     9.0     25.2      3.57      8.38    -0.62
K Benson      B       3      22.5     7.5     22.5      3.33      5.66    -1.84
M Clement     B       3      25.5     8.5     21.6      3.94      6.83    -1.67
O Perez       B       3      24.0     8.0      8.6      9.30      1.66    -6.34
C Pavano      B       4      39.95    9.99    -1.3      neg       0.62    -9.37
J Wright      B       3      21.0     7.0     -9.8      neg       0.04    -6.96
R Ortiz       B       4      33.0     8.25   -20.4      neg       0.25    -8.00
E Milton      B       3      25.5     8.5    -25.0      neg       0.04    -8.46
D Wells       C       2      13.0     9.0     24.1      3.73      7.85    -1.15
C Lidle       C       2       6.3     3.15    10.0      3.15      3.47     0.32
W Williams    C       1+     12.0     7.0      7.4      9.46      2.19    -4.81
O Hernandez   C       2       8.0     4.0      6.1      6.56      2.34    -1.66
P Wilson      C       2       8.2     4.1    -12.2      neg      -0.13    -4.23
                             366.95  123.24   183.7      6.71     71.09   -52.15
                             313.95  109.99   117.6      9.35     54.63   -55.36

The pitchers who received multiyear deals (for a total of 367M) produced just 58% of their AAV. In year one of those deals MLB as an industry wasted 52M. Pedro accounts for one third of the VORP at roughly one tenth of the cost. If you subtract Pedro and just look at the “B” and “C” multiyear deals there is a total commitment of 314M with a one year AAV of 110M. With a “BPVal” of just 54.63M these pitchers returned a fraction less than 50% of their AAV. With 204M left on these contracts, we might expect to see teams reap just 102M of value from their investments. Or maybe less.

I don’t think it’s terribly surprising that one year deals are better values than multiyear deals, but it raises some interesting questions. The Angels and Indians were both reportedly aggressive pursuers of Matt Clement. When he turned their offers down they moved on to Paul Byrd and Kevin Millwood, respectively. Both of those pitchers dramatically outperformed Clement at less cost. Clearly in year one those teams came out way ahead by “settling” for Byrd and Millwood. However, after year one these teams have to again try and fill that one rotation slot while the Sox are locked in with a solid middle of the rotation pitcher. Will that force the Angels and Indians to just sign a different “B” level pitcher this year? Will they try and hit a home run in the “C” pitching market again? If so, what are the chances that they can succeed with that strategy two years in a row?


Part 2: A look forward at the 2004/2005 FA pitching class

These pitchers are taken from the following list from the Cot’s Contract site.

http://mlbcontracts....t-pitchers.html

Just for a tiny bit of perspective I added 2005 IP, ERA, FIP and XFIP. The latter two are different attempts to assess the underlying skill performance that might be hidden by context. I grabbed them from The Hardball Times because they’re useful and easily available. These stats certainly aren’t meant to be any kind of thorough analysis of these pitchers 2005 performances.

A is for “a twilight of his career”, not “the twilight of his career”
                           IP    ERA     FIP     xFIP
Roger Clemens      A      211    1.87    2.85    3.44

Roger Clemens is really, really good. He might actually retire this year. If not signing him to a one year contract for almost any amount would be a good deal.

B is for booyah because Stuart Scott is my personal hero
             
                            IP    ERA     FIP     xFIP
AJ Burnett         B      209    3.44    3.09    3.30
Kevin Millwood     B      192    2.86    3.77    3.99
Jarrod Washburn    B      177    3.20    4.39    5.01
Jeff Weaver        B      224    4.22    4.45    4.28
Mark Mulder        B      205    3.64    4.28    4.13   (reasonable CO)
Javier Vazquez     B      216    4.42    4.04    3.49   (right to demand a trade)

We already know this is a thin year for FA pitching and that can be seen in the relatively small group of likely “B” level pitchers. Last winter ten pitchers received “B” level contracts though arguably only seven deserved one. This winter only the top four in this group look like definite recipients of a “B” level contract to me.

Burnett will probably get a B+ Pavano-type contract. I don’t want to rehash the Burnett debate that’s pretty much been going on since mid-summer. He’s got great stuff. He was very good last year. He had recurring elbow injuries that culminated in TJ surgery two years ago. There will likely be a lot of aggressive bidding for his services. Very soon he’ll be able to afford platinum nipple rings for every day of the week. It’s good to be AJ.

Millwood was a notch below Burnett last year. Interestingly, each of the last two winters he was a FA and he ended up settling for one year deals. Someone who didn’t offer him a multiyear deal last year is going to end up signing him to a pretty big multiyear deal this year. I would think that little fact will make the team that wins the right to give him tens of millions of dollars have some scary second thoughts.

Washburn scares me as much as anybody on the FA market and not just because his peripherals are so far from his very good ERA. He suffered periodic bouts of forearm stiffness this season that seemed ominously similar to the issues Mike Hampton had before he was finally diagnosed with a ruptured elbow ligament. I have an irrational fear that whoever signs him will end up paying him Eric Milton money to rehab from TJ surgery for 18 months. Pass.

Weaver was durable and mediocre. With Boras as his agent it wouldn’t surprise me if he ends up signing very late with the idea that some desperate team will cave in and give him a Derek Lowe type contract. Oddly enough that strategy worked last year both for Boras/Lowe and the Dodgers.

I think it’s pretty much a guarantee that the Cards will pick up Mulder’s option.

I don’t think Vazquez will be traded, and although it looks his season was better than his ERA and record he doesn’t appear to be worth the 2yr/24M left on his contract either. That wouldn’t necessarily be as much of an issue if he was part of a trade for another overpaid player like Manny Ramirez, but as a stand alone contract it’s not that great.

B/C is for cheating on the original categories
             
                            IP    ERA     FIP     xFIP
Esteban Loaiza     B/C    217    3.77    3.32    3.86
Paul Byrd          B/C    204    3.74    3.98    4.78
Matt Morris        B/C    193    4.11    3.97    4.02
Kenny Rogers       B/C    195    3.46    4.11    4.88
Scott Elarton      B/C    182    4.61    5.11    5.19
Jason Johnson      B/C    210    4.54    4.39    4.61

I don’t think any of these pitchers have quite the pedigree necessary to get a 3yr/21M contract, but if there is a lot of money sloshing around the industry they might. I would guess these guys would slot in more in the two years for 7-14M total range. They could be solid values at those prices, but I’m not sure they really fit the Sox needs.

Loiaza had a very good year and has reportedly already turned down a 2yr/8M offer from Washington. Is he going to get a 3 yr/15-18M contract from someone? He’s probably a decent candidate to sneak into the bottom of the “B” range.

Byrd has had too many injures to get more than a two year contract, although the fact that he had some success in KC might make him a foolishly attractive target for them as a veteran rotation anchor.

I may be way off base on Morris because I have seen some generic mentions of him as a fairly desirable FA, but he was terrible in the second half and I don’t think his stuff really came back after his shoulder injury. His full year line is pretty solid, but I would think any team that signs him to a multiyear deal will regret it.

Rogers had a decent year, but he has been pitching on short money, short commitment contracts since 2002. I don’t see any reason for that to change.

I tossed in Elarton and Johnson mostly because they’re younger than most FA pitchers. Neither had an especially good year, but I bet some team will give them 2yrs/8M or so to be a (hopefully) dependable #3/4 starter. I doubt the Sox would be interested.

C is for none of these guys will be Sox so who cares
                           IP    ERA     FIP     xFIP
Tony Armas         C      101    4.97    5.64    5.74
Brian Anderson     C       31    6.75    5.31    4.48
Pedro Astacio      C       67    6.04    4.78    4.24
                           60    3.17    4.13    5.01
Kevin Brown        C       73    6.50    3.65    4.42
Shawn Estes        C      124    4.80    4.75    4.57
Byung-Hyun Kim     C      148    4.86    4.66    5.02
Al Leiter          C       80    6.64    5.64    6.13
                           62    5.49    4.57    5.59
Jose Lima          C      169    6.99    5.75    5.67
Joe Mays           C      156    5.65    5.07    4.91
Brian Moehler      C      158    4.55    4.01    4.30
Jaime Moyer        C      200    4.27    4.44    5.17
Ramon Ortiz        C      171    5.36    5.48    4.93
Kirk Reuter        C      107    5.95    5.33    5.96
Bret Tomko         C      191    4.48    4.18    4.79
Ismael Valdez      C       51    5.33    5.08    5.63
Jamey Wright       C      171    5.46    5.17    4.91

I don’t really expect the Sox to offer enough to sign any of these as their #5 starter, but I’d assume that they’d be able to get more elsewhere than what the Sox might be willing to offer to make one of these guys their insurance #6 starter.

Astacio was surprisingly good in SD, but his peripherals were all over the place.

Brown had a great FIP because his BABIP was a ridiculous .391. Some of that was probably bad luck, but a whole lot of that was because he was legitimately getting knocked around. I’m guessing his .391 BABIP is about as much due to luck as Curt “line drive” Schilling’s .381.

Moehler’s overall line looks decent enough, but he struggled quite a bit in the second half.

Tomko probably will not realize the potential he had as a younger pitcher, but he’s turned himself into a durable (792 IP from 2002-05) league average(ish) innings eater. He can be an asset at the back of a rotation.

I'll have a Reliever review up shortly. A Position player review maybe tomorrow. A team review maybe Mon or Tues. None of those are going to have any looks at the 2005/05 FA class though.

#2 mr guido

  • 3111 posts

Posted 06 November 2005 - 12:49 AM

I don’t think Vazquez will be traded, and although it looks his season was better than his ERA and record he doesn’t appear to be worth the 2yr/24M left on his contract either. That wouldn’t necessarily be as much of an issue if he was part of a trade for another overpaid player like Manny Ramirez, but as a stand alone contract it’s not that great.

<{POST_SNAPBACK}>

The Yankees were so desperate to rid themselves of Vazquez and land Johnson that they sent $9m over in the deal, meaning the remaining contract is more like 2/18 from the D-backs POV.

Interesting analysis and certainly some curious numbers. My first reaction would be to say that if the vast majority of free agent starters & relievers signed last year came up with a negative valuation, chances are the problem lies with BP's valuation rather than (solely) with the market. In other words if the league paid $174.64 for 383.4 VORP, then the de facto value of one VORP is $.46 even if BP thinks it should have cost $134.59 and $.35 per VORP. No?

We all saw the market change last winter so it seems silly to accept a metric that claims last winter's players were overpaid when really that just reflects the new market.

#3 yecul


  • appreciates irony very much


  • 14248 posts

Posted 06 November 2005 - 12:53 AM

This really illustrates the value that top talent can provide even if they're overpaid. Someone like Pedro or Manny might be a couple mil over their true value, but will produce more than the "savings" a lesser player will provide... and on top of that they're just as likely to be overpaid themselves...

Of the available starters I would only pursue (or at least investigate) a few of them. A lot of the Sox plans will depend on what they want to do with Arroyo (trade bait?) and Papelbon (AAA, rotation, or pen). I'm inclined to think both remain on the team, especially with the GM opening potentially giving the team's moves a little pause.

Weaver might be interesting if he didn't have Boras as a client. A reasonable short/cheapish deal for him is somewhat unrealistic. Maybe you can get him late on a make-good contract as he looks to establish his value entering next offseaon (a la Millwood). Boras does like to take this approach. I don't see it happening.

Vazquez would also be a fine pickup. As a trade target he'd be attractive if, as philly mentioned, you could move a contract to mitigate his cost. I'd be much more inclined to pick up pitchers like him and Weaver who are still young, provide lots of innings, and can slot in the middle/back of any rotation with potentially to give top production. Upside baby.

Byrd would be fine at the back end if they moved Arroyo. Ditto Loaiza. I would only want them on short/cheap money. PB is more likely to take than then EL, but after this season they might bump up a notch or two on the pay scale.

I would consider Tomko as an Arroyo replacement. Or perhaps in the role they intended for Halama as the swingman. Likely more expensive and/or not attractive to him, however. Would be good for injury insurance and Papelbon backup I suppose.

#4 Eric Van


  • Kid-tested, mother-approved


  • 10990 posts

Posted 06 November 2005 - 03:57 AM

There was hardly a pitcher in MLB shafted more by VORP than Clement, whose bullpen quite demonstrably let in 5.1 extra runs of inherited runners. That's 24% of his VORP. His $/10 true VORP was actually 3.18. You'll have to recalc BPVal and Diff.

As always, great stuff. And I haven't even read it yet.

#5 OCD SS


  • SoSH Member


  • 6846 posts

Posted 06 November 2005 - 10:41 AM

Considering Pedro, I'm curious as to whether the context adjustment in VORP fully accounts for the switch from the AL to NL. Perhaps he bounced back after a tough '04, but some of his peripherals didn't have the same rebound. I tend to think that he might not have been as good a value if he would've had to work through the AL East again.

Of course, he probably still would've provided better value than some of our other starters this year. It will be interesting to see how his contract plays out over it's full time span.

#6 philly sox fan


  • SoSH Member


  • 9748 posts

Posted 06 November 2005 - 12:04 PM

Interesting analysis and certainly some curious numbers.  My first reaction would be to say that if the vast majority of free agent starters & relievers signed last year came up with a negative valuation, chances are the problem lies with BP's valuation rather than (solely) with the market.  In other words if the league paid $174.64 for 383.4 VORP, then the de facto value of one VORP is $.46 even if BP thinks it should have cost $134.59 and $.35 per VORP.  No? 

<{POST_SNAPBACK}>


That's definitely a possible. I wouldn't take the precise numbers too seriously. It's the general trends that are more significant.

In terms of these pitchers, the biggest drivers of the value shortfall are guys like Pavano, Wright, Leiter, et al who were obvious expensive busts by any common sense value method. While it is in retrospect true that NY paid ~17M for -10 VORP that wasn't what they had envisioned. Right or wrong, they paid those guys on the beleif/hope that they would continue to pitch as well as they had.

And that's basically the premise that Silver used in his article:

Guzman Mulligan

Rather than use PECOTA to project performance, I reverted to a 'naïve' projection system. This system establishes a player’s baseline WARP projection based on a simple weighted average of his WARP scores over the past three seasons, with 2004 weighted at 50%, 2003 at 30%, and 2002 at 20%.

This projection system isn’t going to be as accurate as PECOTA, but it isn’t intended to be. Rather, it’s meant to be a representation of how typical major league teams actually value players in the market; they assume the player is going to do pretty much what he’s done in the recent past, unless maybe he’s really old.


I think that premise is a correct one. And to the extent that "BPVal" overstates the market one of the main reasons it does so is because teams expected players to keep playing well and they didn't. I'd guess that a lot of the value gap comes from underperformance in 2005 both in terms of Silver's WARP projections and the teams wishcasting.

And actually that might be an interesting add on project. Look at these players - or maybe just the top 25 or 50 - and compare thier production and cost in 2004 to thier production and cost in 2005.

I would assume that 2004 production was higher and 2005 cost was much, much higher. That's the primary danger of the FA market.

At least for high end players you can "win" a contract like the Sox did (so far) with Varitek and you end up with a player who maintained his production, but at a 50% increase in cost.

Or you can "get killed" by a contract like the Yankees did with Pavano where the player's produciton craters and you're stuck with a premium cost contract.

#7 philly sox fan


  • SoSH Member


  • 9748 posts

Posted 06 November 2005 - 12:07 PM

There was hardly a pitcher in MLB shafted more by VORP than Clement, whose bullpen quite demonstrably let in 5.1 extra runs of inherited runners.  That's 24% of his VORP.  His $/10 true VORP was actually 3.18.  You'll have to recalc BPVal and Diff.

<{POST_SNAPBACK}>


That's a good point. For any number of players - and pitchers are probably especially vulnerable due to bullpen and defense support issues - there can be tweaks up or down for the valuations.

But the the groupings are both borad enough and distinct enough that those tweaks probably wouldn't change our overall assessment of how the player performed. Despite the late season swoon and attendent carping, Clement was at the very least a decent value signing.

#8 philly sox fan


  • SoSH Member


  • 9748 posts

Posted 06 November 2005 - 12:09 PM

Considering Pedro, I'm curious as to whether the context adjustment in VORP fully accounts for the switch from the AL to NL.  Perhaps he bounced back after a tough '04, but some of his peripherals didn't have the same rebound.  I tend to think that  he might not have been as good a value if he would've had to work through the AL East again. 

Of course, he probably still would've provided better value than some of our other starters this year.  It will be interesting to see how his contract plays out over it's full time span.

<{POST_SNAPBACK}>


The first thing is something we'll never know. His translated peripherals did improve with respect to 2004 so some of his rebound should be taken as real.

The second thing we'll just have to see. But I'm interested enough in the "year 1" vs "year 2" issue that I'll probably uppate this next year.