My S*** Doesn't Work in U.S. News & World Report

For that minority (sliver?) of readership here interested equally in academia, economics, and Michael Lewis' book Moneyball, Volokh co-consipirator Ilya Somin (read his Reason archive here) last night analyzed whether, as in the Oakland A's, recent rankings slippage by former media darling George Mason University is reason to question the underlying strategy of identifying certain market inefficiencies in the hiring process.

My 2003 review of Moneyball here. Reason has a long, rich relationship with the gang at George Mason; most recently we posted a Reason.tv interview with the university's Tyler Cowen.

Comments

[John]

Billy Bean has never built a team that has won a playoff series. Meanwhile in the last four years a small market team, St. Louis, and a small payroll team, the Chicago White Sox, have won a world series and the quintisential small market team Tampa Bay beat Boston and the Yankees to get to another. Also, Bean's A's teams were at the center of the steroid scandals of the 90s and early 00s. The pervasive use of steroids among bay area baseball players makes you wonder if maybe Bean's theory that you draft for OBP and let the power come later was perhaps based on some flawed assumptions.

This is not to say "moneyball" is necessarily a bad book or theory. It is possible that Bean is just bad at applying his own valid theories. But one thing is for sure, Billy Bean is a lousy general manager.

[hmm]

The first to implement effect is bound to slowly lose its benefits as it works for the original person or group to implement the strategy. You see this in the stock market every time a new trend is identified through out history. As everyone else catches on you lose your advantage and the strategy becomes less effective for all. I wonder what the benefits for the players or how the system being adopted throughout the industry impacted all the players.

[John]

How was the "moneyball" movie going to work? A story of a young genius GM using roids to pump up his players only to build a decent team that gets its ass kicked by the late 90s Yankees every year, doesn't exactly sound like Hoosiers.

[Matt Welch]

"Billy Bean" was the gay player; the Oakland GM has an "e" at the end.

[robc]

John,

Cam Bonifay is a lousy general manager. Bean is, at least, above average.

[robc]

Damn it John. Your inability to spell screwed me up.

[John]

RobC,

Considering how the As are doing I don't see how he is above average. Average maybe, but not above. I would be curious to see him take over a team like the Dodgers where he had money to burn and no excuses.

[robc]

Making the playoffs a number of years with no money is an above average skill, regardless of what you do once you get to the playoffs.

Playoffs series are virtually random. 162 game seassons evens out many of the breaks.

[robc]

An "average" GM would have a team that made the playoffs once every 3.5 years (AL). That is without monetary adjustments.

[Matt Welch]

An "average" GM would have a team that made the playoffs once every 3.5 years (AL). That is without monetary adjustments.

It's a bit different in a four-team division, no?

[Pro Libertate]

John,

Tampa Bay is a small payroll team, not a small market team. The Tampa Bay area is the 13th largest television market in the U.S.

Overall, I think statistical analysis and throwing out baseless conventional wisdom is a great idea. But it's not a substitute for managing or playing the game. The problem is, of course, that the measured variables only tell part of the story. It's often the unmeasured variables that make the difference. For instance, the statistical wisdom on stolen bases is probably not entirely correct, because it discounts the effect a base stealer on first has on many pitchers.

[R C Dean]

Does an average (.500) team ever make the playoffs?

[robc]

Matt,

Actually, no, the math isnt that different.

You win win the division once every 4 years on average, and in non-division winning years, win the WC in 1 out of 11 of the remaining years.

So, in 44 years, you would expect 11 division titles plus 3 additional wild cards, for 14 playoff trips. That is 7/22, so 1 every pi years. 🙂

[robc]

Does an average (.500) team ever make the playoffs?

If their division is bad enough, yeah. 🙂

[robc]

RC,

2005 SD Padres won the NL West at 82-80. That is the worst so far.

[robc]

In the 1994 strike year, the Texas Rangers were leading the AL West with a 52-62 record when play stopped. If there had been a playoff that year....

[robc]

The 2nd worst to make the playoffs were the 82-79 1973 NY Mets.

There have been a handful of 83-85 win teams make the playoffs, both in 2 and 3 division leagues.

[robc]

Hooray to baseball-reference.com

[Confused H&R-er]

What does this have to do with Professor Gates?

[Pro Libertate]

Gates used sabermetrics to determine the best course of action to take with the police.

[Seitz]

Billy Bean has never built a team that has won a playoff series.

FAIL. John must have been asleep in 2006.

Postseason:
Lost AL Championship Series (4-0) to Detroit Tigers
Won AL Division Series (3-0) over Minnesota Twins

[Russ 2000]

2005 SD Padres won the NL West at 82-80. That is the worst so far.

Cardinals of '06 had 83 regular season wins, worst of any WS champ (that had a full regular season).

[Seitz]

In the 1994 strike year, the Texas Rangers were leading the AL West with a 52-62 record when play stopped. If there had been a playoff that year....

Kee-rist! I had forgotten how bad the Angels were that year. 47-68 at the strike (coincidentally, playing exactly to their pythag record). Spike Owen was their everyday third baseman! And he was one of their best hitters!! Six regulars with OPS+s under 85, and four of those under 65. That's world class awful right there. No pitchers with an ERA under four. I don't mean starters, I means pitchers! Not even a reliever with an ERA under four.

Man, thank God for the strike.

[John]

Seitz,

I got married that october so I kind of was asleep and out of the country for much of it. I stand corrected.

"Cardinals of '06 had 83 regular season wins, worst of any WS champ (that had a full regular season)."

I often wonder if Jim Leyland didn't throw that series so his mentor Tony LaRusa could win one. There was no way that Detroit team should have lost to St. Louis. What a joke of a series that was.

[robc]

There was no way that Detroit team should have lost to St. Louis. What a joke of a series that was.

I feel like this has already been covered in this thread...Oh yeah...my comment from 10:27:

Playoffs series are virtually random.

[robc]

If I counted right, Beane has taken the A's to the playoffs 5 times in 11 years, which is well above 11 divided by pi.

[Pro Libertate]

Wild cards have no rational place in a playoff following a 162-game season, nor does anything less than a best-of-seven series make any sense.

[Matt Welch]

Playoffs series are virtually random.

Do not agree, at all. That Cardinal team was much better than its record -- it was filled with great veterans who'd been injured for swaths of the year, but healthy in the playoffs. Similar to the Marlins of a few years earlier. It's rare, I think, when the WS champion isn't among who you would describe pre-playoffs as one of the best 3 teams in baseball.

[Pro Libertate]

I dunno about that. The Atlanta Braves had to be the best team in baseball for most of the 90s, but they have exactly one World Series championship to show for it.

There's something to be said for the old way--win the pennant, and you're in. No playoff nonsense unless there's a tie.

[John]

"I dunno about that. The Atlanta Braves had to be the best team in baseball for most of the 90s, but they have exactly one World Series championship to show for it."

Atlanta never had a dominant closer and that showed in the playoffs. They had thier best teams in 91, 95 and 96. The won in 95, but lost in 91 and 96 to teams that had better bull pens and dominant closers. In 1999, they were greatly inferior to the Yankees and got swept.

[Matt Welch]

I dunno about that. The Atlanta Braves had to be the best team in baseball for most of the 90s, but they have exactly one World Series championship to show for it.

There is also an argument that Bobby Cox never adapted his regular-season managing style for the playoffs.

[Pro Libertate]

Atlanta didn't have the same great closer from year to year, but it had several during that period--Rocker (who was dominant for a while), Pe?a, Stanton, et al. Frankly, I don't think Atlanta was inferior to any of the teams it lost to until the end of its run. They had bats, relief, and likely the best starting rotation ever.

[John]

"There is also an argument that Bobby Cox never adapted his regular-season managing style for the playoffs."

Joe Tore completely out managed him in 1996. That Yankee team was not the dominant team of 98 and 99. It was very good but not great. Cox had Maddax Smoltz and Glavine and still couldn't win.

[John]

"Frankly, I don't think Atlanta was inferior to any of the teams it lost to until the end of its run. They had bats, relief, and likely the best starting rotation ever."

The Toronto team that beat them in 1992 was awfully good. And the Yankee team in 1999, with a starting rotation of Clemmens, El Duque Hernandez, David Cone and Andy Petite with Mariano Rivera was one of the great teams of the century. The Braves didn't stand a chance that year.

[Seitz]

Playoffs series are virtually random.

This may be true to a certain extent, but there is definitely a difference between playoff baseball and regular season baseball.

Take a team with five good, but not great starters, and a bullpen full of good, but not dominant relievers. That kind of team can put up a great regular season record. But that depth is nullified in a short series with a three or four man rotation and a shortened bullpen/bench. Some teams mop up on poor teams, and kill mediocre pitching, but struggle against good pitching. Playoffs and regular season aren't exactly the same animal.

Then there are matchup issues. John Lackey has a career ERA of 3.81, but it's almost a run and a half higher against the Red Sox, who happen to kill the Angels in the playoffs. The Sox take a lot of pitches, and Angels pitchers don't do well against teams that make them throw a lot of pitches. Meanwhile, the Angels, with a philosophy of putting the ball in play and putting pressure on the defense (at the expense of power and patience) routinely bitchslap the Yankees. Some teams are just better built to beat certain other teams.

In 2006, the Cardinals were perfectly designed to exploit the Tigers' inability to successfully complete a throw from the pitcher to the third baseman. /sarcasm

[brotherben]

Whatever happened to Steve Avery?

[Seitz]

Whatever happened to Steve Avery?

Over 750 ML innings before the age of 24. Injuries, loss of velocity, and that's all she wrote.

[Pro Libertate]

brotherben,

He tore a muscle in his throwing arm and never really recovered. I think he retired in the early Aughts.

[brotherben]

IIRC, Avery was pretty handy with a bat for a pitcher. I have fond memories of him trying to learn how to brush batters off the plate. Doink. Doink. Doink.

[Mike M.]

Personally, I lost all my respect for Billy Beane when he fired manager Ken Macha after that '06 season (their first ALCS appearance in 14 years), gave no reasonable explanation for it, and then hired his personal best friend (and Best Man at his wedding) Bob Geren.

True intellectuals and people who believe in excellence don't engage in this sort of blatant nepotism.

[John]

"The Sox take a lot of pitches, and Angels pitchers don't do well against teams that make them throw a lot of pitches. "

There is a limit to that strategy. You have to actually hit the ball once in a while. Since the demise of Ortiz and the departure of Manny, the Sox have a bad habbit of just taking pitches and not much else. J.D. Drew is the worst about standing up there and not swinging and finding himself down 0-2. That fact and the American League's discovery that Jason Bay can't hit a curve ball have doomed the Red Sox offense this year.

[rbenchley]

The Red Sox offense isn't exactly "doomed" this year. They're sixth out of thirty teams in runs scored this year, plus their pitching is pretty good as well so they have the best run differential in the AL.

[Brian Courts]

Playoffs series are virtually random.

Do not agree, at all. That Cardinal team was much better than its record -- it was filled with great veterans who'd been injured for swaths of the year, but healthy in the playoffs. Similar to the Marlins of a few years earlier. It's rare, I think, when the WS champion isn't among who you would describe pre-playoffs as one of the best 3 teams in baseball.

Matt, I don't think that the WS champion usually coming from one of the best three teams is enough to "not agree at all" with the randomness comment. Randomness doesn't mean all the teams that make the playoffs have the same chance, it just that the outcome of any particular series has a large degree of randomness in its outcome and therefore is of limited value in assessing which is truly the "better" team.

To give some concrete numbers to this discussion, look at the expected results in a couple scenarios. If you measure the better team by its probability of beating the inferior team in any given game, then a team that is 55/45 superior would win a 7-game series only about 6 out of 10 times. Even if you take a truly dominant 67/33 team (and in baseball, beating even regular season opponents at 67% is unheard of, much less another playoff team) will lose a 7-game series about 1 in 5 times. So if a team more dominant than any ever is in baseball would lose one in five, it's no big surprise when the lesser team in a more realistic match-up beats the better team.

Of course you're right that the better team will usually win, over the long run, but that doesn't refute the point that the result of any given series is highly random.

[John]

"Of course you're right that the better team will usually win, over the long run, but that doesn't refute the point that the result of any given series is highly random."

See e.g. 1960 Pirates, 1988 Dodgers, 1985 Royals, and 2006 Cardinals as examples of inferior teams that managed to beat substantially superior teams in a seven game series that prove your point.

[John]

"The Red Sox offense isn't exactly "doomed" this year. They're sixth out of thirty teams in runs scored this year, plus their pitching is pretty good as well so they have the best run differential in the AL."

I wouldn't say doomed in the sense they are a bad team. But doomed in the sense that good pitching will kill them in big games. They got smoked in game 7 against TB last year. They haven't produced a big hit in a key game since the 2007 WS. I really think they are the third best team in that division. After Youkalis and Pedroia there is no one in the lineup who would concern you. That is a problem especially when you compare it to Tampa Bay and New York's lineups.

[Matt Welch]

See e.g. 1960 Pirates, 1988 Dodgers, 1985 Royals, and 2006 Cardinals as examples of inferior teams that managed to beat substantially superior teams in a seven game series that prove your point.

Four times in a half-century? Color me unimpressed! (Also, the Cards weren't *really* that much worse than the Tigers.)

That 1988 Dodgers team, though, were stinky enough to prove the randomness theorem all by their lonesome.

[Pro Libertate]

Kirk Gibson limping to the plate. Friggin' awesome, and I hate the Dodgers.

[Brian Courts]

Four times in a half-century? Color me unimpressed!

I doubt that is an exhaustive list. Again, unless you think there are many times in the post-season when a team is more than a 55%/45% favorite, it's going to happen about four times every 10 series. Of course those aren't huge underdogs, but the point remains, judging a superior team simply by its winning a seven-game series is a dubious proposition at best.

[Voros McCracken]

The big reason why the Moneyball movie is not working out is because I'm not in it.

I'm box-office gold, I tell ya.

[han]

I have to say this is good