December 22, 2005

From The Mailbag

JAWS, Revenue Sharing, and Sandwich Picks

by Baseball Prospectus

Worst of the Best

I like your articles looking at the Hall of Fame ballot each year, but I disagree with some of your methods. It isn't the JAWS score but that you use the average of all HOFers, dropping the worst ones. Players who meet these standards are obviously in, the standards that need to be met are those of the borderline guys.

I understand that there are plenty of mistakes in the HOF (Lindstrom and Kell over Hack and Groh?) However, not dropping the bottom guy would go a ways towards solving the problem, getting the standards closer to what the in/out line should be in the right people were in the HOF.

--Mark Shirk

Mark,

The "right line" as you suggest is an entirely arbitrary construct that's subject to each person's taste. The averages come out differently depending upon which method is used to determine them. The first two years I did the study I didn't drop the bottom guy at each position; you can easily dig up those articles in the BP archive to see the results.

However, the more I studied the data, the more I became convinced that my standards have been a bit too low, particularly with respect to the way Veterans Committee selections weigh down the positional averages. An article I did about the recent VC ballot showed about a 25-point gap between the two sets of enshrinees (note that the numbers were determined by the older, 5-consecutive method and don't directly translate to the current article--the effect is still essentially the same with the new method). As ticked as I am that the BBWAA won't elect Bert Blyleven or Rich Gossage, they do a reasonable job by NOT electing a lot of borderline candidates who won't improve the Hall, while the old VC made a mockery by admitting, in some cases, the wrong brother of a duo.

With the current positional averages in mind, I defy you to convince anyone that the eight VC-elected guys I deselected (stats shown below) are relevant when considering the case of, say, Don Mattingly or Albert Belle. They're essentially outliers, and many of them are at least 10 points behind the next lowest score at their position.

POS       #  BRAR  BRAA  FRAA   WARP   PEAK   JAWS
C        13   410   196   74    97.6   59.4   78.5
1B       18   738   483   -2   100.4   60.9   80.6
2B       17   570   295   88   114.1   67.1   90.6
3B       11   656   374   63   108.8   62.6   85.7
SS       20   415   137   87   102.4   62.0   82.2
LF       18   745   470  -15   105.2   59.7   82.4
CF       17   715   466   -8   108.6   63.8   86.2
RF       22   780   504   21   112.4   61.5   86.9

                       BRAR BRAA  FRAA  WARP  PEAK  JAWS
Roger Bresnahan    C   325  170   -80   58.3  40.8  49.6
George Kelly      1B   242   41    81   50.4  39.3  44.9
Johnny Evers      2B   284   63    23   69.4  46.0  57.7
Freddy Lindstrom  3B   272   88    -6   49.5  42.3  45.9
Travis Jackson    SS   227   22   -22   57.6  46.9  52.3
Chick Hafey       LF   366  217   -50   49.0  41.8  45.4
Lloyd Waner       CF   287   39   -46   55.8  37.6  46.7
Tommy McCarthy    RF    92  -82    14   23.8  29.8  26.8

In general, electing borderline candidates due to lower averages doesn't improve the content of the Hall of Fame, it compromises it further. Your mileage may vary, but the general consensus seems to be towards tighter standards, not looser ones.

--Jay Jaffe

Valued at $1, Give or Take $59 Million

Could you guys do me a favor and explain baseball revenue sharing for both myself and a friend of mine?

I have a friend who told me that the Cubs have their broadcast rights on the books of the Tribune at $1 and thus because they have their own network they are able to hide revenue sharing dollars from being shared.

On the flip side, I read an article in the N.Y. papers about 3 weeks ago that Major League Baseball values the Yankees YES Network at $60 million for 2005 and that in fact they were looking to have the figure revised upwards.

Is it possible that the Cubs are hiding such amounts from baseball revenue sharing?

--C.T.

C.T.,

It's not only possible, it's quite likely. As the late great Doug Pappas reported two years ago, the Cubs regularly report less local media revenue than their crosstown rival White Sox, despite the fact that until this October you couldn't find a White Sox fan in Chicago with Google Earth. From there, let's allow sports economist Andy Zimbalist to pick up the story, from his book "May the Best Team Win":

So, what's going on? The Cubs are owned by the Tribune Corporation, which happens also to own WGN. The Tribune Corporation transfers revenue away from the Cubs and correspondingly lowers the costs of WGN. According to Broadcasting & Cable, the industry's authoritative source, the Cubs' local media earnings were $59 million. [In 2001; by 2003 this figure was $63 million. -ND] If the Cubs had reported this figure instead of $23.6 million, then their reported $1.8 million loss would have become a $33.6 million profit in 2001!

There's a long tradition of these hide-the-revenue shenanigans in baseball--George Steinbrenner once paid himself a $25 million "consulting fee" to negotiate his own TV rights deal, and no one has believed the Braves' media figures in decades. But the temptation to fib on your reporting form has only gotten stronger as rights fees have gone through the roof (some estimates have the Yankees' cable rights worth as much as $200 million a year) and the revenue-sharing rate has climbed to around 40%.

Bud Selig has tried to keep these sorts of abuses in line by appointing a revenue oversight committee, and, as you noted, even threatened to audit the Yankees' reporting if they didn't come clean, though it's been two years now and that Daily News story has been the only peep about the audit's progress. Until MLB starts opening up its books for real, it's probably best to assume that all teams operate by Paul Beeston's credo when he ran the Blue Jays: "Anyone who quotes profits of a baseball club is missing the point. Under generally accepted accounting principles, I can turn a $4 million profit into a $2 million loss, and I can get every national accounting firm to agree with me."

--Neil deMause

Rating Kaat and Glavine

Read your article and agree with you on many points. I want to ask about a few other pitchers with regards to their WARP3 and JAWS scores. Specifically Tom Glavine and Jim Kaat. I'm sure Glavine will get in with two Cy Young awards, but am interested to see his JAWS score relative to the others, given that he and Randy Johnson may be the first pitchers of the new era without the 300 win barrier to gain entrance. Kaat I'm interested to see because the number of wins he has is comparable to Tommy John and Bert Blyleven.

--Timothy Bumpus

Tim,

Kaat: 102.2 career WARP3/56.7 Peak/79.5 JAWS, falling slightly short. What really kills his case for me is that he's actually at -5 PRAA for his career, despite 1067 PRAR. That's a lot of hanging on, and it doesn't really measure up to Blyleven. It's interesting that he's close to John, with a higher peak of about 1 win a year but less career value. Ordinarily I'd prefer that combo to the other way around, but the PRAR/PRAA stats are definitely in TJ's favor, as are the "intangibles."

Glavine: 124.9/61.3/93.1, a no-brainer in his favor. The two pitchers he falls between right now are Fergie Jenkins and Nolan Ryan. Good company.

--J.J.

Bottom of the Ninth

Thanks for the great series explaining things. Looking at the draft order you linked, I am still confused by the order of the sandwich round. From your description, shouldn't each team's picks be in consecutive slots? The last several slots in the sandwich rounds intermix the Red Sox and the Cards with a Marlins pick mixed in. Can you help explain any further?

--C.D.

C.D.,

I've gone over the CBA and the Major League Rules (MLRs) a few times and I'm quite confused. At first I thought that I might be holding an old version of the MLRs. My copy still has an "Order of Selection" section that requires the AL and NL to alternate picks, and that clause was scrapped a couple years ago. I asked someone with a fresher copy to track down any other edits to this section of the rules and the report I got back wasn't helpful. The newest version of the rules doesn't have an explanation of the sandwich round order.

I pinged some contacts at a couple clubs and the answer I got explained that the compensation round order goes something like this: teams pick in reverse order of their finish in the previous year's standings but teams with two picks have to wait until every other team with a free agent compensation pick goes at least once, and teams with three picks have to wait until everyone with two picks goes twice, etc. After all the compensation picks for free agents the sandwich round moves to compensation picks for unsigned 1st rounders from the previous year (like the pick the Orioles got for losing Wade Townsend).

I got confirmation from a club official that this rule exists in writing and that it came from the Commissioner's Office. Apparently the Commissioner's Office occasionally has to interpret the trickiest of rule problems (like this sandwich pick order of selection) and they issue their rulings in memos to the teams that are then collected in a binder of precedents. So not only are the Major League Rules secret, but there's also a binder of rule interpretations that I'd never even heard of.

Another reader showed me his blog and pointed out another fascinating part of the sandwich picks. Teams with Type A free agents have a disincentive to resign their players. When you lose a FA you get the signing team's regular pick and the bonus sandwich pick. When you sign someone else's FA all you lose is your regular pick. If you swap your FAs for someone else's you both swap regular picks and everyone nets those bonus sandwich picks. Since the sandwich round takes place at the end of the first round those picks can be quite valuable.

Look at Boston from last year. They lost Pedro Martinez, Derek Lowe, and Orlando Cabrera and replaced them with David Wells, Matt Clement and Edgar Renteria.

If they'd just resigned their old players the Red Sox first picks would have looked something like this (assuming all the other compensation picks went the same way):

1st Round: 28th overall selection
2nd Round: 73rd overall selection
3rd Round: 105th overall selection

As a result of the way they flipped their FAs for new ones they ended up with these selections:

1st Round:      23rd overall - Angels' compensation to Boston for Orlando Cabrera
                26th overall - Dodgers' compensation to Boston for Derek Lowe
Sandwich Round: 42nd overall - for Pedro Martinez
                45th overall - for Orlando Cabrera
                47th overall - for Derek Lowe
2nd Round:      57th overall - Mets' compensation to Boston for Pedro Martinez

Would you rather have picks 28, 73, and 105, or would you rather have 23, 26, 42, 45, 47, and 57? Three picks in the first 105 or six in the first 57? Nice little loophole.

--Tom Gorman