Peter Out

Peter Gammons has an article on the why it is best not to put all your eggs in one nest, er, player (Teams don't always get what they pay for). The article stinks of someone else's research (specifically Doug Pappas' whose salary numbers start in 1985. Besides it is preceded by an article on the Braves' rotation options for 2003 with Maddux and Glavine apparently gone and in typical Gammons-speak not one stat, barring captions, is employed throughout).

Besides, the impetus for the article is the potential of the Phils signing Jim Thome for $15 M (a sum that cannot be proposed to Thome officially until tomorrow but is already a forgone conclusion in the media):

[B]ut the days of the $17 million-$20 million contracts may become antique concepts. The dangers of a team in small- or medium-sized markets putting a large contract in the hands of a single player it cannot do without have come back to haunt the team time and again.

Second, will Thome's $15 M contract, assuming that will be the Phils' offer and that he accepts, be indeed 15% of their team payroll? They would have to have a payroll over $100 M and this year they were under $60 M and are in the process of jettisoning a lot of high-priced veteran dead weight (Terry Adams, Ricky Bottalico, Robert Person). Let's look at next year's guaranteed contracts: The Phillies have $8.5 devoted to Bobby Abreu in 2003 (plus $3 M signing bonus), which will escalate to $15 M by 2007. Mike Lieberthal will make between $7.25 and $7.5 M a year over the next three seasons. Jose Mesa is owed $4.5 for 2003, Rheal Cormier $2.9, and Turk Wendell $3.25. Tomas Perez signed for $1.3 M over the next two years. Pat Burrell is arbitration-eligible this year and will get a big bump up from $1.275 M in 2002 (let's say $5 M). That's about $53 M for seven players. They have a few young players who will get substantial increases in the upcoming seasons due to arbitration. They are also pursuing David Bell (in the $5 M per season range) and a few big-name starting pitchers.

Let's say that they sign Thome at $15 M and get at least one big-name free agent in the $5-10 M range. That would give them at least $73 M devoted to 9 players. Let's say they give Randy Wolf , Jimmy Rollins, Marlon Anderson/Placido Polanco, and Vicente Padilla at least $750 K each. Say, of the rest (12 players), half play at league minimum (just raised to $300K), and the other half play at about twice that, that gives them an additional $5.8 (the Phils had 6 players under $300K this year).

The team salary would about $82-87 M, short of the $100 M figure. That's for this year. Let's assume that the Phils are building for their moving into their new stadium in 2004. In a couple of years, Abreu will be approaching the $15 M range, and Marlon Byrd, Jimmy Rollins, Pat Burrell, the second baseman (Anderson or Polanco), as well as much of the starting rotation will be maturing and will get substantial increases due to arbitration. The Phillies may continue to pursue high-end free agents next winter to create a buzz for the move. It is not inconceivable that the $18 M shortage could be made up in a year or two as the Phils' young players mature and the team tries to make a run at the playoffs.

Third, he's examining the symptom and not the cause. It's not that signing players for a lot of money is bad. It's signing one player for a lot and then having a bunch of scrubs that is bad. Take a look at the Florida Marlins for example (this is using both Sean Lahman's database and Doug Pappas' salary data):

Team Year Tm Payroll Max Salary Max %   W   L  Pct Pos
FLA  1993 19,705,545  4,225,000 21.44% 64  98 .395 6
FLA  1994 21,633,000  4,625,000 21.38% 51  64 .443 5
FLA  1995 24,515,781  5,625,000 22.94% 67  76 .469 4
FLA  1996 31,132,000  6,135,000 19.71% 80  82 .494 3
FLA  1997 48,583,000  7,000,000 14.41% 92  70 .568 2
FLA  1998 41,492,667 14,936,667 36.00% 54 108 .333 5
FLA  1999 21,085,000  7,000,000 33.20% 64  98 .395 5
FLA  2000 20,072,000  7,000,000 34.87% 79  82 .490 3
FLA  2001 35,762,500  7,000,000 19.57% 76  86 .469 4

Note that in the Marlins championship year of 1997, their maximum salary ducks below 15% for the only time in their existence. But note also that it is a substantial increase over past maximum salaries for the team. The '98 team has a max. salary (Gary Shefield's) that is 36% of the total, but this is an aberration-the next three years the max salary holds steady at $7 M, but the payroll goes down. So does this tell us that the Marlins won by shaving bloated salary? No, the won by adding a lot of bloated salary in 1997. They lived with those consequences for a short time and then pared down all of the team payroll except for the $7 M contract (Alex Fernandez, whom they couldn't rid themselves of).

The Marlins before and after 1997 were a similar team according to Gammons' study. But in reality they were worlds apart. The 1993-96 version was a young team that had a low salary base with a couple of free agent signings that exceeded 15% of the total payroll. In 1997, they had a relatively large team payroll but no one player getting the lion's share. In 1998, they dealt with the excesses of '97, and from 1999 until today, they have been a small market team that has had a couple of ill-advised free agent signing that they couldn't divorce themselves from.

Gammons' study is lacking cause and effect. Without that the numbers lose their meaning. The context must first be determined by asking questions like: Is this a young team that is rebuilding in earnest or one that made choices in the past that are still haunting them? Is this a well-balanced veteran team with a number of players deservedly being paid a healthy salary or just a collection of low paid scrubs? By comparing players' salaries to the rest of his team you lose the context. Looking at the team payroll within the league or looking at the change from the previous year(s) would be an interesting study because it would provide context.

I do give him props for trying. Welcome to sabermetrics in the 1980s, Peter.