Ross Ohlendorf: Did Advanced Stats Help?

In the wake of Ross Ohlendorf’s arbitration win, it would be easy to claim that this was a win for advanced stats. That may not be the case, however. Yes, Ohlendorf recorded only one win last season, but this is the not the first time a pitcher has won an arbitration hearing with one win. You only need to go back to 2009, when Shawn Hill won his hearing against the Nationals despite a 1-5 record the previous year, to see that it’s not necessarily always about wins and losses.

But while it’s not just wins and losses, advanced stats are not yet commonly used in arbitration. Just a few years ago, there were stories about arbitrators who needed RBI explained to them. And while in some situations these days you will see advanced stats used, Ohlendorf’s advanced metrics don’t necessarily help him make his case. A large factor in arbitration is your bulk, and so even with an advanced bulk stat like WAR, Ohlendorf doesn’t necessarily stack up better than recemt comparable players.

On a more basic level, it’s not hard to see why the Pirates would have felt confident taking Ohlendorf to a hearing. Last season, he had two stints on the disabled list, and as a result only accumulated 108.1 innings pitched in his platform year. Add in that lone victory, and that during his career he was optioned back to the Minors for more than two months, and the Pirates were armed with a couple of black marks on Ohlendorf’s record. They also had Edinson Volquez’s recently signed contract on their side of the midpoint.

In some areas, Volquez and Ohlendorf match up well. A player’s first arbitration hearing, be it as a super-two or a regular three-plus in terms of service time, is the one time when career and platform season (the season most recently completed) are weighted equally. Certainly, each side can argue differently, putting more emphasis on what strengthens their case most. Using Volquez as a comp is a good example of how the two sides would likely have argued differently.

The Pirates would be more inclined to point out that career ERA, WHIP and K/BB for the two are nearly identical, but since Volquez has more than double the wins and was a Game 1 starter in the playoffs, Ohlendorf shouldn’t earn more than him. Ohlendorf, on the other hand, would point to the fact that it took Volquez two seasons longer to accrue barely more innings pitched than Ohlendorf did, thanks to Volquez’s injuries and performance enhancing drug suspension. Additionally, Ohlendorf would point to the fact that he had a better platform year per conventional statistics like ERA, IP and WHIP (he wouldn’t point to platform year FIP, since Volquez’s 4.00 trumps Ohlendorf’s 4.44 handily). But where Volquez presented the Pirates an opportunity, there simply weren’t a ton of cases for them to use. Here are some recent comp players for Ohlendorf – pitchers whose platform season was 2006 or more recent, who were in their first arbitration hearing and who were primarily starters during their platform season. First, platform (or 3+ season) numbers:

3+ Year      
Player 3+ Yr Next Yr $   IP   ERA   WHIP   K/BB   FIP
Jer. Weaver 2009 $4,265,000   211.0   3.75   1.24   2.64   4.04
M. Pelfrey 2010 $3,925,000   204.0   3.66   1.38   1.66   3.82
Erv. Santana 2008 $3,800,000   219.0   3.49   1.12   4.55   3.30
J. Saunders 2009 $3,700,000   186.0   4.60   1.43   1.58   5.17
J. Verlander 2008 $3,675,000   201.0   4.84   1.40   1.87   4.18
E. Bedard 2006 $3,400,000   196.1   3.76   1.35   2.48   3.59
J. Cueto 2010 $3,400,000   185.2   3.64   1.28   2.46   3.97
D. Braden 2010 $3,350,000   192.2   3.50   1.16   2.63   3.80
C. Capuano 2006 $3,250,000   221.1   4.03   1.25   3.70   4.04
J. Jurrjens 2010 $3,250,000   116.1   4.64   1.39   2.05   4.19
J. Guthrie 2009 $3,000,000   200.0   5.04   1.42   1.83   5.31
J. Hammel 2010 $3,000,000   177.2   4.81   1.40   3.00   3.70
D. Cabrera 2007 $2,875,000   204.1   5.55   1.54   1.54   5.01
S. Olsen 2008 $2,800,000   201.2   4.20   1.31   1.64   5.02
J. Lannan 2010 $2,750,000   143.1   4.65   1.56   1.45   4.47
K. Slowey 2010 $2,700,000   155.2   4.45   1.29   4.00   3.98
P. Hughes 2010 $2,700,000   176.1   4.19   1.25   2.52   4.25
J. Maine 2008 $2,600,000   140.0   4.18   1.35   1.82   4.40
W. Rodriguez 2008 $2,600,000   137.1   3.54   1.31   2.98   3.62
D. Bush 2007 $2,550,000   186.1   5.12   1.40   3.05   4.57
C. Vargas 2006 $2,500,000   167.2   4.83   1.41   2.37   4.85
S. Feldman 2009 $2,425,000   189.2   4.08   1.28   1.74   4.31
B. Morrow 2010 $2,300,000   146.1   4.49   1.38   2.70   3.16
T. Gorzelanny 2010 $2,100,000   136.1   4.09   1.50   1.75   3.92
Ohlendorf’s # 2010 $2,025,000   108.1   4.07   1.39   1.80   4.44
P. Maholm 2008 $2,000,000   206.1   3.71   1.28   2.21   4.15
C. Gaudin 2007 $1,775,000   199.1   4.42   1.53   1.54   4.69
Midpoint   $1,712,500          
E. Volquez 2010 $1,625,000   62.2   4.31   1.50   1.91   4.00
Pirates #   $1,400,000          
J. Patterson 2006 $850,000   40.2   4.43   1.11   4.67   3.25
J. Litsch 2010 $830,000   46.2   5.79   1.46   1.07   5.44
S. Hill 2008 $775,000   63.1   5.83   1.75   1.70   4.06

And career:

Career thru 3+ Yr        
Player 3+ Yr Next Yr $   IP   ERA   WHIP   K/BB   WAR
Jer. Weaver 2009 $4,265,000   671.2   3.73   1.25   2.00   12.9
M. Pelfrey 2010 $3,925,000   683.0   4.31   1.46   1.56   8.0
E. Santana 2008 $3,800,000   706.2   4.42   1.29   2.61   12.0
J. Saunders 2009 $3,700,000   571.1   4.22   1.37   1.78   6.9
J. Verlander 2008 $3,675,000   600.0   4.11   1.33   2.18   10.7
E. Bedard 2006 $3,400,000   476.0   4.08   1.44   2.12   10.7
J. Cueto 2010 $3,400,000   531.0   4.27   1.35   2.31   5.5
D. Braden 2010 $3,350,000   473.1   4.20   1.33   2.13   7.3
C. Capuano 2006 $3,250,000   561.2   4.20   1.33   2.44   8.0
J. Jurrjens 2010 $3,250,000   550.1   3.52   1.30   1.97   9.0
J. Guthrie 2009 $3,000,000   603.0   4.27   1.32   2.01   6.4
J. Hammel 2010 $3,000,000   561.2   5.06   1.48   2.24   8.1
D. Cabrera 2007 $2,875,000   661.1   4.99   1.53   1.43   8.7
S. Olsen 2008 $2,800,000   579.1   4.63   1.45   1.81   3.7
J. Lannan 2010 $2,750,000   566.1   4.10   1.41   1.39   3.9
K. Slowey 2010 $2,700,000   473.1   4.41   1.28   4.57   6.9
P. Hughes 2010 $2,700,000   369.0   4.20   1.27   2.48   6.4
J. Maine 2008 $2,600,000   464.2   4.18   1.31   1.97   4.3
W. Rodriguez 2008 $2,600,000   584.1   4.79   1.41   2.10   6.7
D. Bush 2007 $2,550,000   630.1   4.53   1.26   3.23   9.2
C. Vargas 2006 $2,500,000   532.1   4.92   1.44   1.81   1.9
S. Feldman 2009 $2,425,000   430.2   4.58   1.39   1.45   4.6
B. Morrow 2010 $2,300,000   344.0   4.19   1.43   1.97   4.8
T. Gorzelanny 2010 $2,100,000   558.0   4.68   1.49   1.60   5.8
Ohlendorf’s # 2010 $2,025,000   354.0   4.40   1.39   1.89   1.9
P. Maholm 2008 $2,000,000   601.1   4.30   1.41   1.84   6.7
C. Gaudin 2007 $1,775,000   359.0   4.46   1.57   1.42   2.8
Midpoint   $1,712,500          
E. Volquez 2010 $1,625,000   388.1   4.36   1.48   1.86   5.6
Pirates #   $1,400,000          
J. Patterson 2006 $850,000   423.0   4.09   1.31   2.55   6.2
J. Litsch 2010 $830,000   342.2   4.10   1.32   1.90   3.3
S. Hill 2008 $775,000   206.1   4.93   1.46   1.94   3.0

There simply aren’t a lot of cases on the Pirates side of the ledger, and except Volquez, those that are had poor platform seasons. That lack of precedent in his case – there just aren’t many examples of a player with only around 350 innings pitched but also with 20-plus starts in their platform season – opens the floor for more creative arguments. So while Ohlendorf doesn’t stack up with most of these players on his side of the ledger in terms of games started or innings pitched (or FIP or WAR), there is a window to an effective argument in traditional rate stats. He has better or similar ERA, WHIP and K/BB numbers than many players on his side of the midpoint, such as Mike Pelfrey, Brandon Morrow, Jeremy Guthrie, Scott Olsen, Joe Saunders and Scott Feldman, Daniel Cabrera, John Lannan and John Maine.

Again, Ohlendorf didn’t have the career bulk of these players, and that is why those players earned or will earn more than him. On the other hand, Ohlendorf and his team seem to have known that, as they didn’t ask for similar amounts of money. And with those players on his side of the midpoint, Ohlendorf had more comparable players from which to build his case than did the Pirates.

Ross Ohlendorf and his people did need to show that wins and losses aren’t the only thing that counts in their arbitration hearing against the Pirates, but it wasn’t necessarily to their advantage to use advanced statistics to illustrate that point. They well may have, as only the people in the room on Tuesday will know what was said and what wasn’t, but it certainly would not have been necessary. Filing an appropriate salary number and taking advantage of a lack of precedent by using rate stats like ERA, WHIP and K/BB may have been just as, if not, more important.

Print This Post

Paul Swydan is the managing editor of The Hardball Times, a writer and editor for FanGraphs and a writer for the Boston Globe. Follow him on Twitter @Swydan.

21 Responses to “Ross Ohlendorf: Did Advanced Stats Help?”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. hunterfan says:

    “They well may have, as only the people in the room on Tuesday will know what was said and what wasn’t”

    Is there a gag order on the arbitration hearing?

    Vote -1 Vote +1

  2. Telo says:

    It’s still kind of baffling to me, that at the very highest level of “NEEDING TO KNOW HOW EFFING GOOD A BASEBALL PLAYER IS”, these folks still don’t look at advanced stats… let alone know they exist (assumedly, from the RBI anectodote.)

    I understand that service time is the driving factor in all arbitration cases, but once you have comparables, as you showed, you then look to performance at that level of service time to make the call.

    Millions and millions of dollars every year… baffling.

    Vote -1 Vote +1

  3. Andrew says:

    Are players allowed to represent themselves? From what I’ve read, he’s more than capable of presenting his own case to an arbitrator.

    Vote -1 Vote +1

  4. Tom Au says:

    One thing I’ve noticed is that FanGraphs places Ohlendorf’s value at about $5 million in 2009 and $3.5 million in 2010 (if he were a free agent). A first-year “arb” player is supposed to make 40% of his free agent value, which would be $2.0 million if you use 2009 (Ohlendorf’s position) or $1.4 million if you use 2010 (Pirates’ position).

    One could argue that Ohlendorf’s injury-plagued 2010 was “atypical,” and therefore that a $5.0 million value for 2009 was a better measure of his potential going forward. The team would argue the reverse, for $1.4 million. The arbitrator’s decision supports the player.

    Vote -1 Vote +1

    • Telo says:

      Yea, like Bill said, that “value” is just a players WAR multiplied by whatever a “win is worth” for that given year, which at present is in the 4.5-5mil range. For arb it’s an entirely different ballgame, and your “would be FA” market value has basically nothing to do with how much you receive in arb. It’s much more to do with comparable players in service time, and then performance.

      Vote -1 Vote +1

      • Ed says:

        You’re both right.

        Pre-free agency arb salaries are strictly based on what players with similar service time and performance make.

        Historically, that’s averaged out to 40%/60%/80% of free agent values for the 3 arb years.

        There are definitely exceptions, but for the most part, either way of looking at it gets you a similar value.

        Vote -1 Vote +1

  5. Bill says:

    The fangraphs value is what he would be projected to get as a free agent. I’m not sure what the adjustment is for arb eligible guys like Ohlendorf, but it’s significant.

    Vote -1 Vote +1

    • Tom Au says:

      I gave you the adjustment. It is a multiplication by (roughly) 40% of the free agent value. And both figures (Ohlendorf’s and the Pirates’) worked out to about 40% of the 2009 and 2010 values respectively.

      Vote -1 Vote +1

      • Jeff says:

        Ohlendorf was a Super Two rather than a standard 3 year arb player. Is there an accepted % scale for Super Two players with 4 arb years?

        Vote -1 Vote +1

      • JRoth says:

        A guy at Bucs Dugout looked at this a few months ago. Super-2s tend to get around 30% of FA value, and then go (roughly) 40-60-80. One thing his research showed is that the 40-60-80 scale is only very approximate – probably ±5% at each stage (not just for the range of players, but even for the average; IOW, it could be more like 44-57-85 on average, but people love round numbers, so).

        Vote -1 Vote +1

  6. ofMontreal says:

    Beyond precedent setting numbers and comps… we are talking Pirates here. Ohlendorf is arguably their 2nd best pitcher and someone they are counting on to take a regular turn this season. And the Pirates are nearly touching bottom in terms of salary obligations. So why are they fighting over a couple of hundred grand with Ross? With an attitude like that, I expect them to trade him out of spite. And things are supposedly turning around in Ptown.

    Vote -1 Vote +1

    • Ryan S says:

      It’s simple: Because being a Super Two, his win in abitration will increase his salary over the next three seasons which makes it a more difficult decision whether or not to keep him around.

      Vote -1 Vote +1

      • JRoth says:

        Exactly right: a $600k difference in his Super-2 season could easily add up to $3.5-4M over the 6 years of control, which isn’t a lot, but isn’t chump change, either.

        Vote -1 Vote +1

  7. Tigerdog says:

    Once and for all, let’s get the fact straight that the primary considerations in an arbitration proceeding are NOT statistics. The primary factors- the parameters within which an arbitration award will be made, are the level of experience, his own salary history, and the salaries of players with a like amount of experience. Once in a blue moon, you get a Ryan Howard, who can make a case that he deserves to be paid more than any other player of similar experience, and they you have a new upper limit, but the vast majority of players will fall within a pre determined range based on players with the same basic description in terms of experience, the position they play, and the number of games that they play, before they even begin to look at performance.

    When the arbitrators do begin to look at performance, they’ll look to special recognition awards first, because those are specifically referred to in the language of the CBA. When it comes to Ohlendorf’s “poor performance” the case is immediately made that they kept playing him. He filled the role that they put him in. Losing teams lose games. If he’s no good, then why play him and not pay him?

    This article gives an excellent break down, first sorted by experience and position, of the comparables that will be used by an arbitration panel. If you immediately rush to the statistics before establishing the parameters, you’re going to be lost.

    Vote -1 Vote +1

  8. What a great and informative article and discussion. I am just starting to learn about all the salary arbitration stuff, and see that it is very complicated. I look forward to reading more of your articles.

    Vote -1 Vote +1

  9. MGL says:

    “Once and for all, let’s get the fact straight that the primary considerations in an arbitration proceeding are NOT statistics. The primary factors- the parameters within which an arbitration award will be made, are the level of experience, his own salary history, and the salaries of players with a like amount of experience.”

    I know little about salary arb in MLB, but if what you say is true, then there would be little correlation between players’ arb salaries and their WAR as well as their future FA salaries (which is all about talent/value). There would be some – because, if by “experience” you mean playing time, more experience equals better talent/performance – but not much. Also, if what you say is true, then the idea of using 30/40/60/80 (percent of a player’s FA value) would be useless.

    Given the research that went into the 30/40/60/80 construct, I highly doubt that what you said (the above quote) is true. But, I could be wrong. If someone wants to do a little work, simply run a regression of arb salaries versus WAR or subsequent FA salaries, holding “experience” constant. Or do a “poor man’s regression” by breaking down players with similar experience and salary histories into two groups – high WAR and low WAR – and see what their arb awards are as a group. I also don’t really know what you mean by “his own salary history.” By definition, all pre-arb players are typically paid the league minimum. If a player has a contract that pays them more than that, they are typically locked up through the arb years anyway. Occasionally a (high draft pick) player will have a high salary given to him upon signing, or occasionally a player will be given a small raise for “good-will” purposes prior to their arb years.

    Vote -1 Vote +1

  10. Maury Brown says:

    Many heads are going to spin here, but to do a real world comp analysis, remove all the advanced stats. Maybe get crazy and throw in OPS.

    Seriously, the process is what the process is. It’s a great discussion to throw in WAR, etc. but at the end of the day, it’s simply an exercise in a “what if” discussion.

    Finally, none of this removes the great work Paul did on the article (we are working together via my data sets, and vetting material). One can hope we get to sabermetric nirvana in salary arb hearings, but it’s more than a ways off.

    Vote -1 Vote +1

  11. rohit says:

    The primary factors- the parameters within which an arbitration award will be made, are the level of experience,

    Vote -1 Vote +1

  12. adieLo says:

    good job! Impressed by your artcle. Anyway,each players will get more better and better by their experience in the field, out in the field. And how they prove themselves out there, with their performance, the money rolls on and on and on. well sources say on vabaseballdotcom all possible info why this happens.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>