February 15, 2007
Give 'Em Enough ROPE
Marginal Wins Revisited, RevisitedLast week's article expanding on Doug Pappas' Marginal Payroll/Marginal Wins formula generated lots of reader mail, as expected. Everyone who wrote in seemed to like the idea of building a better metric for judging teams' efficiency at spending on players; there was less agreement, however, on how to do so.
The loudest gasps were prompted by the article's final conclusion, that by looking at teams' ROPE scores (Return on Payroll Expenditure, which is essentially the old Wall Street term "return on investment" with a cute acronym tacked on) we can determine that nearly every team in baseball is spending more on payroll than it gets back in new revenues, many by a factor of more than two to one. To be honest, this shocked me when I saw the numbers--I've long held to the position that sports team owners are in it to make a profit, first and foremost, and will only spend on players what they figure they can recoup from the increased ticket sales, TV rights deals, etc., that result from a winning ballclub.
That's not what ROPE shows, though. As I wrote last week, aside from the handful of teams that are getting a decent return on their investment, "Every other team in baseball would have been better off, from a revenue perspective, by fielding a minimum-wage team and taking their lumps on the field." That prompted reader W.L. to ask:
My thought is that if a team fielded a minimum-wage team, then their ticket revenue would decline, since many fans wouldn't pay to go see such a lousy team...ditto for concession revenue, etc. ...so shouldn't marginal revenue take this into account?
Absolutely. According to Nate Silver's research in Baseball Between the Numbers that sparked this whole thing, revenues for a losing team will indeed decline. If the Texas Rangers, to pick an example, had released their entire squad and called up a team of replacement-level talent from Oklahoma, they could have expected to see revenues plummet by nearly $27 million. But they also would have saved $59 million on payroll, for a net profit of $32 million. By that measure, then, Tom Hicks is leaving money on the table by making an effort to win the AL West.
Of course, unless you're the Marlins, it's not so simple to jettison everyone making more than the minimum. Reader B.S. notes that while it's easy to calculate ROPE in hindsight and smack GMs on the wrist for money poorly spent, that's not how decisions are made in the real world:
In reality, at the end of the previous season, but even at the very beginning the next season, the best the GM can do is make a guestimate or forecast of how his team will do in the coming season--as he contemplates what roster moves to make. ...
Second, you assume that a GM has the ability add or shed payroll willy-nilly, without being locked into long-term payroll commitments. In reality, he may be stuck with a whole lot of extra payroll at the end of 2006, a team with the talent to win a forecasted 70-80 games next season, and not have the guts to dump players on the market for fear of alienating his fan base, nor the flexibility of simply buying more talent and eating the salary on his current roster of aging stars and ne'er-were-very-goods.
These are excellent points, and it was too flip of me to say that GMs are dumber than a box of rocks. (Okay, I actually said "spending money really badly," but you could read between the lines.) Nate further points out that in the real world, GMs face what he calls the "playoff bubble": "This year, you sign a guy because you think he'll put you over the playoff hump. Next year, lo and behold, you find that you're still on the playoff hump and sign another guy under the same premise, with the old guy still on your books. And so forth. Basically, the sunk costs become something of a vicious cycle."
Reader M.G. took issue with my use of baseball's reported revenue figures, writing:
Particularly for those teams (like the Braves) owned by publicly traded corporate parents, it's difficult to imagine such widespread waste. Off the cuff I'd expect about half the teams to have a ROPE above 1.0, and half below. I wonder if the problem here is the "hidden" money. Thanks to the regional sports networks and their generally non-public books, MLB clubs may have succeeded in obscuring the true value of a marginal win.
I've written before about hidden revenues--one of my favorites is the time George Steinbrenner paid himself a consulting fee to negotiate his own cable deal--and I'm sure including them would boost the revenue figures here. But how much? When Doug Pappas looked at MLB's reporting of media revenues back in 2001, he found only a couple of cases where revenues were likely to have been substantially understated; even if we assume that more teams have been able to offload profits to RSNs in the meantime, we're still talking about a relatively small amount of money. The total "ROPE gap" you'd need to fill to get the average team in the black is more than a billion dollars a year--if teams were really hiding that much money, I think even Bud Selig would notice.
Other questions included why I didn't why I didn't chart "marginal dollars per marginal weighted third-order wins" to avoid dinging GMs for bad luck (because I was trying to evaluate return on investment, and teams don't get a revenue boost from winning the third-order division crown) and whether postseason wins should be added to the marginal win total (no, if only because the postseason is such an unpredictable crapshoot, though Nate's figures do include a bonus for an average postseason appearance).
Then there were the readers who suggested that it's wrong to apply a single marginal revenue formula for the whole league, since marginal revenues vary too wildly from team to team--though they couldn't agree how: One reader insisted that big-market teams are right to spend more on players because they have more ways of making money off a playoff appearance, while another wrote that small-market teams have more to gain, since teams like the Yankees and Red Sox are going to sell out their home parks regardless, while the Royals need to actually win games to get people to show up. (Speaking hypothetically, of course, no one alive ever having seen the Royals win games.)
For the record, Nate says that his investigations have found no significant correlation of any kind between marginal revenue potential and market size. If true, this is potentially even bigger news than the finding that teams are overspending on payroll, since every sports economist on the planet will tell you that the main reason why big-market teams spend more on free agents than small-market ones do is because acquiring a top player earns them more in marginal revenue. This clearly calls for further investigation.
So, great, now I have to write another article. Stay tuned.