Making Money (June 23, 2003)
I'm re-reading Voros' excellent final Primer article where he concludes the relationship between $ and wins (and remaining linear regardless of market size). If all the non-winning variables remain the same the marginal rev $/ marginal win is 2.65. (Though he does have a curved relationship if we include the playoffs.) Paying out about 70% of the marginal revenue to marginal salaries, and that marginal salary / marginal win comes in at 1.85.
Why .7 and not the average of .6 or 1.0? Again, we are talking about marginal, so the .6 would be the absolute minimum. There is a risk associated with taking on more salary (revenue might not come in), and so, you have to give the team a reason to take the "junk bond" player. As well, you might have increased expenses to cover the extra attendance the extra wins will carry. 0.7 works for me, though you could argue for .75 (or 2 million$/win) or .80 (2.15 million $ / win).
So, what does it mean? My best guess is that a replacement team will win 49 games (.300). If the average team has a 70 million$ payroll, that makes the replacement level payroll: (49 - 81) x 1.85 + 70 = 10. The minimum allowed payroll is about 10 million$. So, it looks like this model will work. The fans have properly established the market value of a win.
If winning wasn't important, if fans came no matter what, the marginal $/win would approach zero. In that case, all players would be paid the minimum salary, and teams would pocket the huge profits.
While in 2003 they may not have paid for players the 1.85 going rate, they will have to soon. Why? Because that's what the fans are generating. They are generating 2.65 million $ / marginal win. Teams can't hope to only spend 1.3 on the players, and pocket the rest. The (free) market won't allow it.
--posted by TangoTiger at 05:01 PM EDT
Posted 6:58 a.m.,
June 24, 2003
(#1) -
unc84steve(e-mail)
I'm going to have to find the original article. But saying:
"If all the non-winning variables remain the same the marginal rev $/ marginal win is 2.65. (Though he does have a curved relationship if we include the playoffs.)"
Seems similar to "Besides that, Mrs. Lincoln, how was the play?" (I don't know if you do smilies here :-)
Please. Teams in MLB are supposed to chase "trophies & rings!" (the problem occurs when they stop). Joe Garagiola, Jr. drives me crazy, but those playoffs were fun (at least in 2001). Many teams delude itself into shooting the moon (aka "playoffs") to get to that "curved relationship" portion where playoffs help marginal revenue--that has got to count. (Big time)
Hey, trust me, I'll be there for all the "anti marketing" conspiracies you want, but MLB hasn't officially canceled the playoff yet, have they?
--Steve
Posted 8:48 a.m.,
June 24, 2003
(#2) -
tangotiger
You don't have to look for the article as it is in the link of the title of this thread.
Your quote of mine was me referring to Voros' equation of
Team Revenue = (W% * $430,169,580) + (Metro Population * $3.46) - (Teams in Metro Area * $27,962,685) + (Home Playoff Games * $2,446,043) + (Per Capita Income of Area * $2,655.60) - $160,287,379 + ($22,906,159 if the team’s stadium is less than two years old).
So, the non-winning variables are: population, teams, per capita income. I meant "all things equal" among these three, and let's only focus on the relationship between winning and revenue.
I also said that he has an extra variable for playoffs, which will force a curved relationship for the higher achieving teams. According to the Voros equation, each home playoff game adds 2.5 million$. If you are the "upper echeleon" team, maybe you expect to add an extra 5 million$ of which you'd give 4 million to the players. So, instead of having an 80 million$ payroll to finance an 87 win team, you really have 84 million$ payroll, or 2 mlliom$ instead of 1.85.
I also didn't say anything about any conspiracies, but rather that there's a fundamental equilibrium point that would force the free market to center around.
So, what exactly am I saying that you disagree with?
Posted 6:00 p.m.,
June 24, 2003
(#3) -
unc84steve(e-mail)
I'm sorry, I need to apologize.
First, reading Voros McCrackin's essay was revealing. I tend to see the complications in everything, while Voros asks if "obvious" matter. This is similar to his model of assigning the roles of pitching vs. defense.
Second, I jumped to the conclusion that you (or he) was neglecting an important part of the marginal revenue equation because it seemed so obvious to me that teams like my hometown D'backs got a next season bump from the actual playoff berth (as opposed to the winning pct.)
I see with further review that in the comments section the last exchange with Art asked that specific question.
So anyway to summarize my understanding of the article & data, is that it concentrates on MARGINAL salaries, wins & revenue. That is completely different than the ABSOLUTE salaries, wins & revenues. Of course, the Yankees & LA have higher starting points than KC & Milwaukee do, but that involves different baselines which has nothing to do with marginal calculations.
I assume Carlos Beltran has the same MARGINAL effect on wins in KC as he would for the Yankees. But perhaps as people grossly exaggerated the effects in the WL columns of free-agents (10 to 20 instead of 3 to 6), I'm sure people exaggerate the revenue effects of a star on revenue.
(Rationalizing) I didn't see the effect on attendance but perhaps small markets get more excited when the local teams make a run and get some extra wins (as opposed to the jaded big cities). This would be reflected in extra attendance, merchandising and TV where all eyes would be on the local heroes.
This could lead to Carlos Beltran getting as good an offer in KC as in a big market literally because of the "big fish in a small pond" argument. There may be more T-shirt buyers in LA or NYC, but in those cities they have Shawn Green, Nomo, Lo Duca; Jeter, Soriano, Giambi to also choose from.
Posted 5:06 p.m.,
June 25, 2003
(#4) -
tangotiger
Steve, thanks for writing back.
I am surprised that the effect of revenue works as it did, rather than as a multiple of the other factors (like population, etc). Makes life easier though that way... And the playoff impact wasn't as great as you would have thought, though I am surprised.
It's good to remember that Voros' equation is based only on 3 years of data, and so, we would have issues there. It would be worthwhile to re-run for the last 10 years.
Posted 5:48 p.m.,
June 26, 2003
(#5) -
Walt Davis
Upon re-reading that article, it occurs to me that per-capita income is not quite the correct variable. Per-capita income is much higher in the NY metro area than it is in KC ... but so is the cost-of-living. We should probably add a COL variable to the equation or somehow adjust PCI for the COL.
Posted 8:00 p.m.,
June 26, 2003
(#6) -
tangotiger
I suppose what we want is "disposable net income". NYC residents and workers also pays a city tax.