THE BOOK--Playing The Percentages In Baseball

I wrote the following on JC’s blog, regarding salary growth:

Why is 10% not sustainable?  Payroll has grown at a 10% clip from 1985 to 2005.  Also 10% from 1990 to 2005.  As high as 9% from 1996 to 2005.  While 6% may seem more appropriate, I’d counter that MLBAM money will be enormous.  MLBAM is already worth more than the Yankees I’d guess.  And one day, possibly within the next few years, MLBAM will be worth 50% of the entire league.

Teams will have lots of money floating around, and, while spending it won’t give them a bigger share of MLBAM, they won’t be able to resist not spending it.  They are going to budget spending 55% - 60% of their revenue on payroll, even if 45% makes more sense because of MLBAM-revenue sharing.

We need to keep in mind, however, that the new CBA has set the inflation for minimum salaried players at 22% over the course of the deal.  If we’re giving FA salaries a 10% rate of annual inflation, but the league minimum salary is only increasing at about 4.5% annually, then we’re projecting the middle class of talent to become more and more overpriced. 

Despite the new money coming into the league, the pre-arb players have to be considered some sort of rational cap on spending. Teams aren’t going to be able to justify $6 or $7 million for a one win player when their more intelligent competitors are kicking their tails with a couple of highly paid superstars and a whole bunch of no-names. 

I’d argue that the 10% rate of inflation isn’t sustainable because the GMs responsible for that historical figure are going to get fired for being terrible at their jobs.

Tom, where do you get a half win per year decline from?  I think it is closer to 2 runs (.2 wins).  If it were .5 wins, then a league average player at, say, 30 years old, would be replacement level in 3-4 years. That does not sound right. 

Plus I don’t think I have ever determined whether annual decline was a percentage of a player’s “talent” or a fixed number (in runs).  Do you know the answer?

In fact, looking at my lwts aging cuves, it appears to be exactly a 2 run a year decline from age 28 to 38.  I am not sure if that is defense plus offense or just offense.  I think that is both (Supelwts).

Nope, that is offense only.  Superlwts indeed declines by around 4.5 runs per year from age 28 to age 38, although it is a little less per year in the early 30’s than in the late 30’s.  Plus, I still don’t know if that is a fixed number of runs for all players (on the average of course) regardless of their initial lwts or it is a “percentage” of their initial lwts, or perhaps some other kind of function.

There’s a table and graph at the end here:
http://www.tangotiger.net/aging.html

which shows that the aging is not linear, and is rather steep in the mid-to-late 30s.

The fielding-aging, as you know, begins a bit earlier.  And, if the triples/SB aging offer a good clue (say one-third of the fielding weight):
http://www.tangotiger.net/agepatterns.txt
then the fielding aging is probably even more severe.

As well, the worse you are, the less playing time you’ll get, so that’s another component, when looking at it from a “replacement” perspective.

Marcel has two aging components in steps 4 and 5 here:
http://www.tangotiger.net/archives/stud0346.shtml

The step 5 age adjustment shows that the farther away your current age is from 29, the more severe the age adjustment.

All-in-all, for the little work that I need from it, a 0.5 change in win per year is just what a handy-dandy chart asks for, especially since almost all FA are over 30.  I suppose if I was more serious about it, I’d do more work on it.

David, what you say makes sense, rationally.  Which basically means it’s invalid with MLB.

I think owners will simply set a budget as a percentage of total revenue, and will spend to that, whether it makes sense or not.  As long as the revenue stream is there, the owners will give players their share, even though they don’t deserve it, from an economic standpoint.

"As long as the revenue stream is there, the owners will give players their share, even though they don’t deserve it, from an economic standpoint.”

Whether they deserve it or not is certainly debatable.  You’re assuming a player only ‘deserves’ the marginal revenue he generates for a team by increasing it’s on-field performance.  But there is a substantial amount of money flowing to all teams from TV contracts, merchandise, etc.  The players certainly believe they deserve a significant share of that, and I’m inclined to agree.  Arguably, the players ARE the product we all pay to see; the owners are just administrators, who deserve only a modest fee for organizing the games.  The truth is the owners bring relatively little of value to the productive process, in either financial or intellectual capital. 

And in practice, what would happen if the players’ share of revenue fell very far is they would strike and extract a new agreement that restores the old percentage, or something close to it.

Agreed with Guy.

***

Right, I agree that he only deserves the marginal impact.  At the extreme, if all teams pooled all their revenues (cable, playoffs, local), then all players would be paid exactly the same, which means, the bare minimum.  (i.e., Wal-Mart).  But, if that happens, players will simply find a different sport, vocation, or a new league would sprout up.

As for the owners being “administrators”, the same rule applies to a managers of a consulting firm like Perot (ex-future-president).  The service is the workforce (the consultants).

So, there are other universal considerations, and those forces will act to make the player’s take in the 50%-65% range.

Well, MGL already answered his own question before finishing post #3, but a league average player at 30 being replacement level by 33-34 sounds exactly right to me.

I remember an old study where Bill James was looking at attrition rates, and I think more than half of the players who played regularly at 30 had been replaced by 34.

And as to the question of why 10% inflation is not sustainable:

It may continue for a while, maybe 5,10, 20 more years, or it may grind to a halt in 2008.  I can’t say for sure when, but if 10% baseball inflation keeps up then in 100-150 years baseball salraies will equal the entire GDP of the U.S., assuming that grows around 3%.  I don’t think even Cuba spends their entire GDP on baseball.

If I’m lucky and live 60-65 more years, and if 10% continues that long, then I might see the average American spend as much of their income on baseball as I do.

I’m estimating I spend 100X the percent of the average American.

Rally raises a good point.  At some point salaries will hit a ceiling in terms of a percentage of revenue, and then can only grow as fast as revenue.  That could certainly be well above inflation for a long time, but not 10%. 

Tango or Studes (or anyone else):  do you know how the payroll/revenue ratio has changed over the past few decades?

Sounds right to me, Rally.  I think in 100 years, MLBAM will be its own country.

***

Back to MGL’s question: don’t forget, if at your peak you are a league average player, you’re darn tootin’ that you’ll be out of the league in 4-5 years, from that point.

A guy who comes into the league as a league average player at age 22 will hit replacement level at around age 39.  A guy who is MVP-like at age 22 will hit replacement level at age 43.  A replacement-level at age 22 will rehit that level at age 35.  (Around the peak, the change in wins level is closer to the 0.3 wins mark.)

(All this based on a little model I’m working on.)

I read it in the Zimbalist book, or maybe Heylar (both great reads) about 15 years ago or so, and they provide the revenue numbers.

I’m saying that the payroll/revenue rate will remain flat, and that as long as revenues increase at 10% per year, so will payroll.  I don’t see the situation of revenues outpacing payroll.

MLBAM and cable provide an enormous amount of potential, and 10% will be sustainable for quite a while.

Can one use the model to evaluate existing FA contracts, or just those signed at this point in time?  Specifically, with yet another round of Manny trade rumors, could we use the model to say that at 2/$38, Manny is set to be paid for +4.75 wins?

You can use it for existing FA contracts, but only to evaluate the contract from this point forward.  So, if Manny has 2/38 left, then he’s being paid for +4.8 above replacement.