November 27, 2007
Questions and AnswersQuestions and Answers
Today we're going to revisit few recent topics with the help of reader mail, and start discussion of a new area with the unintentional help of a new general manager. Without further ado:
Equivalent Average and Normalized Statistics
This one isn't a letter, but rather a quote I read recently in Alex Carnevale's excellent The Week In Quotes column. It's from new Pirates GM Neal Huntington, discussing some of the statistical measures that he'll be using in the Pittsburgh front office:
We are going to utilize several objective measures of player performance to evaluate and develop players. We'll rely on the more traditional objective evaluations: OPS (on base percentage plus slugging percentage), WHIP (walks and hits per inning pitched), Runs Created, ERC (Component ERA), GB/FB (groundball to flyball ratio), K/9 (strikeouts per nine innings), K/BB (strikeouts to walks ratio), walk percentage, etc., but we'll also look to rely on some of the more recent variations: VORP (value over replacement player), Relative Performance, EqA (equivalent average), EqOBP (equivalent on-base percentage), EqSLG (equivalent slugging percentage), BIP% (balls put into play percentage), wOBA (weighted on base average), Range Factor, PMR (probabilistic model of range) and Zone Rating.
The quote was originally an answer in a question and answer piece at MLB.com. Huntington presents an impressive toolbox, one that includes some of our favorites. Still, the way that EqA is listed alongside EqOBP and EqSLG brings up an issue, about the possibility of confusing EqA (Equivalent Average) and EqBA-which is Equivalent Batting Average).
Please note that I'm not saying that Huntington doesn't understand this distinction-I'd bet that he does. I'm just saying that the two statistics' names and appearance are close enough that the situation is ripe for confusion. The prefix Eq- denotes statistics created by Clay Davenport, where a player's actual performance has been adjusted to strip away contextual issues like the ballpark the player was playing in, the total offensive level of the league, and the league's difficulty relative to the majors. You'll also see these referred to as "translated" or "normalized" statistics.
The benefit of this approach is that every player-season in major league history can be compared on the same scale-you can look at players from the Deadball Era alongside their counterparts in the home run-happy mid-'90s just by viewing the "Translated Statistics" sections of their respective DT Cards on the web site. Just about any hitting or pitching statistic can be translated in this fashion, and in previous installments of Toolbox we've discussed Equivalent Strikeouts (EqK) and Equivalent Runs Scored and Allowed (EqR/EqRA). Equivalent Batting Average is simply the player's batting average once it's been translated to a neutral ballpark and a league with a league batting average of .260.
In the case of Carl Yastrzemski, there were two seasons in which he finished with a .301 batting average-1968 and 1974. Both of those seasons (his entire career, actually), his home ballpark was Fenway Park. Nonetheless, you can't really consider the two seasons equivalent in terms of batting average. In 1968, a .301 average was good enough to lead the American League. This was the lowest league-leading batting average ever, as the AL as a whole hit .230 that year, in what might have been the offensive low point in the game's history. By 1974, the pitching mound had been lowered by almost a foot, and the league batting average was a much healthier .258. The difference is reflected in the EqBA numbers-Yastrzemski's .301 season in 1968 produced a .325 EqBA, the same batting average six years later was only worth a .292 EqBA.
While EqA is also a normalized statistic, it's one that tells us much more than EqBA. Equivalent Average is a rate stat that incorporates almost all of the major components of offense, including hits, total bases, walks and steals. The result is scaled to look like batting average-the average is .260, .300 is good, .220 is bad. However, it's actually more like OPS, in that it incorporates the elements of on-base percentage and slugging percentage, and correlates to runs scored much better than batting average (or OPS, for that matter) does.
Free Agent Compensation Notes
Most of the feedback on my last column came before Alex Rodriguez and the Yankees kissed and made up. So in the questions that follow, when readers mention Rodriguez, imagine that they're talking about the next-most prestigious free agent currently available-which is who, Geoff Jenkins? Shannon Stewart? Not a lot of plums left in that barrel.
Getting on to the first question, Dan Budreika asks:
Can you explain what would happen if a free agent reaches a deal with another club before he is offered arbitration by his former club? I know the deadline to offer arbitration is in December. For example, what would happen if A-Rod reaches a deal with the Angels in two weeks and the Yankees have not yet offered A-Rod and Bora$ arbitration? Would [they] still net compensation picks?
If a compensation-eligible player is signed prior to the clubs' deadline to offer arbitration (December 1), the team who loses the free agent gets the compensation, as if arbitration had been offered. For example, the Braves have already signed Tom Glavine, a Type A free agent, away from the Mets, so the Mets will receive the Braves' first-round pick (the Braves will pick in the bottom half of the June amateur draft, so their pick is not protected), as well as a supplemental first-rounder.
Because there's always a chance that a free agent won't be offered arbitration by their old ballclub, it's been widely considered wasteful to sign a compensation-eligible player prior to the deadline unless:
However, given that the signing team is now only penalized for signing Type A free agents, and there are fewer of those now than there were before, it will be interesting to see if more teams are aggressive about signing free agents prior to the arbitration deadline.
In the next question Reader C.H. imagines shenanigans resulting from all the sandwich round picks that could wind up moving around in the free agent market:
I really enjoyed your article and Unfiltered post about the free agent compensation system, and it made me think of a silly hypothetical question. Let's say the Red Sox agree to re-sign Lowell and the Yankees agree to re-sign A-Rod. What's to stop the Red Sox and Yankees conspiring so that the Red Sox sign A-Rod, the Yankees sign Lowell, and the teams make a straight-up trade of the two players, thereby switching first-round picks, but gaining additional compensatory picks before the start of the second round? Am I understanding the rules correctly? I'm sure this collusion is unlikely, but it seems like a win-win situation for both teams if their first-round picks are very close to each other.
You understand the compensation pick rules correctly-if the Yankees and Red Sox had signed each other's third basemen, they would have traded off draft picks (though not necessarily each other's first-round picks, as we'll discuss below, that depends on the teams not signing other free agents that have higher compensation scores) and in addition, the two richest franchises in baseball would have gotten a little bit richer, because each would get a supplemental first-round pick. Boston's supplemental pick would be higher than New York's, because Lowell had a higher score than Rodriguez.
The tricky part would be the two teams trying to trade the players back to each other. Taking the more obscure rule first, it would require coordination not only between the two teams, but also between the teams and the free agents-under the Collective Bargaining Agreement (CBA), Article XX(B)(6)(a), to be exact-new free agent signees can't be traded without their consent prior to June 15 of the first year of their contract. Teams acting in concert with other teams or players acting in concert with other players with regard to free agency are both forbidden by the CBA, so this deal would be a double no-no.
Although some believe that MLB has been lax when it comes to policing interactions between the teams during Bud Selig's watch, one of the few times it's happened was last year, when the Phillies tried to circumvent the waiver rules by swinging a deal with the Devil Rays to claim pitcher Bobby Livingston off of waivers from the Mariners, and then sell him to the Phillies. The situation's not completely analogous, but it's possible that in this hypothetical situation the league could invalidate the signings, the trade, the compensation picks, or even the players' free agency rights.
There is a completely aboveboard strategy that doesn't require any collusion or trades, which takes advantage of the same principles: let your compensation-eligible players sign elsewhere (after you've offered them arbitration, of course), and then replace them with compensation-eligible players from other teams. You don't run into problems with the free agent quota system because a team is always allowed to sign as many Type A or B free agents as they themselves have lost, and you're virtually guaranteed to come out ahead on the draft picks. This is because a team can only lose their own draft picks as compensation for a free agent signing, not compensation picks received from other clubs or sandwich round picks.
Boston effectively did this after the 2004 season. In the short term, signing Matt Clement, Edgar Renteria, and David Wells to replace Pedro Martinez, Orlando Cabrera, and Derek Lowe was pretty disastrous. In the long term, however, the Red Sox lost the 28the (first round), 76th (second round), and 108the (third round) overall picks, but gained two first-round picks (23rd and 26th overall), three supplemental first-rounders (at 42nd, 45th, and 47th overall), and a second-round pick. Some of the selections they made might sound familiar to you today: Jacoby Ellsbury came aboard with the pick the Red Sox got from the Angels for Cabrera, and the supplemental picks became Clay Buchholz, Jed Lowrie, and Michael Bowden. Those are four out of the top six prospects in the Red Sox organization, per Kevin Goldstein's recent Top 11 list.
MLB and the MLBPA, The Collective Bargaining Agreement (courtesy of Maury Brown's Biz of Baseball web site): Most of the transaction rules about free agents and compensation are contained in Article XX of this document.