Marginal Utility

by Ken Houghton

Have Lenny Dykstra and Beowulf Shaeffer ever been seen together?

Collateral question: Has Jim Cramer ever gotten anything correct?

Labels: baseball, Fraud, High Finance, Jim Cramer, Journamalism, Ponzi schemes

by Ken Houghton

The last time I saw a World Series game in person was 19 October 1993, in Philadelphia.

Dr. Black appears to have better luck than I at picking Phillies games to attend.

Or perhaps it's just a Random Walk.

Labels: baseball

by Ken Houghton

As if any more were needed.

Labels: baseball

by Ken Houghton

Via David Pinto at Baseball Musings.

Labels: baseball, LEGO

by Ken Houghton

At the end of a series of e-mails that were deliberately segued from football to baseball (or perhaps vice versa), I sent:

Did the Cardinals play the Bengals? I'll be David Eckstein ran for at least 100 yards.

The response was brilliant, warming the cockles of hearts everywhere, but especially over at Fire Joe Morgan:

Yes. He over-ran 1st base by 210 feet, trying to hustle out a bunt.

Labels: baseball, Fire Joe Morgan

by Ken Houghton

Wouldn't it be interesting if:

1. Evan Longoria was Eva's brother?
   
2. Ashley Dupre (nee Youmans) was the daughter of Floyd Youmans, who was Montreal's Opening Day starter in 1987.*

All right, maybe not.

*Yes, I'm still working on the Rose betting posts.

Labels: baseball, sports, Yankees

by Ken Houghton

Or will a minor-league sport (through all fault of its own management) get to close The House that Ruth Built?

Labels: baseball, hockey, Yankees

by Ken Houghton

Fifty-nine is probably younger than Derek Jeter's arm, though he will be sixty the next day. Good to see that Bud is earning his $15 million a year.

(I vaguely remember that he had a decent swing from Comic Relief in Radio City, but I can't find a YouTube video of it.)

Labels: baseball, Yankees

by Ken Houghton

It took years, but the final straw was hiring Dusty Baker to manage a young team with a lot of potential (and Griffey and Dunn, who I believe are a country music duo).

Fortunately, Fire Joe Morgan lays out the details on why the team would be doomed anyway.

Labels: baseball, sabremetrics

by Ken Houghton

If you haven't figured it out yet, I'm distracting myself from the vagaries of Health Economics and the trenchant, profound political "analysis" of the ProgLib Blogsphere—not to mention tomorrow's memorial—by going through the data from Coate (2008), available in PDF form here.*

I've found a minor mistake in the Coate data; at least two games allocated to the Chicago White Sox (AL) are actually Cubs games (NL). So the data presented below may have minor differences from the previous three posts.

One of the things about making a bet is that there are two possible motivations: one might be betting for a team, or one might be betting against the other team. So it might be worth looking (again, excluding bets involving the Reds) at which teams have large differentials (in both absolute and result basis). So, as a first run:

There is a difference between "0" and a blank in the above data. Using the example of the Atlanta Braves, Rose bet for the team 12 times: he won six and he lost six, leaving a net of zero (0).

By contrast, and following up to the issue raised in a previous post regarding his previous teams, there is NO indication in the data that he bet FOR the Montreal Expos, so there is a blank or null in that field.**

Also, in some cases, one might believe that a team has a significant advantage due to playing at home due, for instance, to the shape of the ballpark and the team developed to play in it.*** So, especially in the absence of other information, there might be a bet placed on a home team with such a park. Following is the bets on a general basis, broken down by league and home/road team being favored. (Again, the Reds are excluded.)

More ongoing.

*CREDIT UPDATE: Some data in this portion available from Baseball-Reference.com and/or Baseball Almanac.

**Similarly, he never bet against the Toronto Blue Jays, betting on them ten times during the period, winning seven and losing three for a net of +4. Other than the Canadian teams, all of the zeroes on the summary table indicate the net result of an even number of bets.

***There are teams that have had a significant advantage playing on the road, of course, but such advantages generally appear later in the season.

Labels: baseball, efficient markets, Rutgers

by Ken Houghton

In order to keep from discussing the Obama campaign's methods here—the Hilzoy/De Long axis keeps trying to tell us they are pure, and the commenters here appear to agree; why let matters of fact* get in the way?

So let's look a little more at the gambling activities of one Peter Edward Rose.**

The simplest way to answer the question most recently posted is to look at his cumulative returns. While we know the ultimate answer must be significantly negative, the accumulation is frightening in its consistency***:

After April 11th, Rose doesn't even approach break-even. There is a brief "winning streak" from the 15th to the 18th, and a fairly solid positive trend in American League bets from 15 April to 4 May (14 nonnegative days in 17 days of bets), but the time from May 5th to the 12th produces $30,000 in losses (three days—May 6, 8, and 9—lose $26,000, almost all of which is due to bad AL bets that more than negate the net $21,000 gain of the AL streak).

On the surface, on balance, and with few exceptions, the only obvious trend in the data Coate presents on Rose's betting is that he fails to consider the primary rule of betting: if you can't find the sucker at the table, the odds are good it is you.

*Link added here, as it was deleted at the original discussion, though a peculiar manifestation of The Obama Rules remains in response to it.

**Again, all data from Coate (2008), available in PDF form here. Dollar amounts do not foot for the National League; presumably this is an error of mine.

***All of the following data excludes the Cincinnati games, since we are working on the presumption, supported by the evidence presented, that Rose would not have bet against the Reds. The overall effect upon the data of adding those games does not appear as if it would be significant.

Labels: baseball, efficient markets, Rutgers

by Ken Houghton

Following up to the previous post, some more details.*

1. On an absolute basis, any presumption that Rose's working within the National League should give him special insight is not borne out by the data. Indeed, he performs notably better when betting on the American League:
   

   
   
   

2. His winning and losing splits almost perfectly. If he bets $2,000 or less on a game, he is likely to win. Bets of over $2,000 appear more likely to lose.**
   

   
   
   

3. The combination is disastrous.
   

   
   

Overall, the high-level data appears consist with a gambler whose instincts and information are near-perfect contraindicators.***

More later.

*All data from Coate (2008), available in PDF form here. Dollar amounts do not foot for the National League; presumably this is an error of mine.

**It would be prudent to assume that bets of $2,200 consist of a $2,000 and the 10% "vig" charged to the loser. However, the data indicates five (5) bets of $2,200, two (2) of which are winning amounts.

***Another example of this may be his consistent betting in favor of the Phillies, one of his former teams, which had a major P&L impact, as shown before:

but there is not a similar effect for his other former team, the Montreal Expos.

Labels: baseball, efficient markets, sports

by Ken Houghton

I walked past Douglas Coate's office a couple of times yesterday. Had I seen this paper (PDF) (h/t Tyler Cowen via Portfolio), I might have stopped in.

Curiously, especially if you exclude CIN games,* it appears that Rose was a better gambler when he didn't have access to information (i.e., better overall betting on the AL than the NL).

UPDATE: The distribution is interesting. Assuming the data is correct, The Philadelphia Phillies are responsible for 29% of his losses over the period. Indeed, Philadelphia, Detroit, and Houston account for over 60% of the dollar losses.

And to no one's great surprise, the same teams, along with San Diego, are the teams where Rose bets frequently and (presumably) the wrong way, with the Phillies again setting the standard:

Maybe more tomorrow.

*Which makes sense, since the consensus is that Rose always bet those, and only to win. Though the April 25th entry of "no bet" may be indicative something else.

Labels: baseball, efficient markets, Rutgers

by Ken Houghton

I have been watching in (somewhat mock) horror as Hillary Rodham Clinton gets triangulated out of the race for the Presidency and Roger Clemens—who for the past ten-plus years has shown the same trends and the same physical changes as Barry Bonds, in a much more fragile position—denies having used steroids.

But if Howard Bryant is to be believed—always an open question—Roger is about to get, well, rogered.*

I believe it was Gerald Wilkins, after "losing" a slam-dunk contest to Michael Jordan when the All-Star Game was in Chicago, who most poignantly observed, "It's difficult to win on the road in the NBA."

If Bryant's description of Wednesday's hearing is correct, Wilkens description applies aptly to Clemens's position :

As titillating, tawdry and undignified as the blood feud between Roger Clemens and Brian McNamee has been, the real showdown Wednesday before Congress is not between Clemens and McNamee, but between Clemens and former U.S. Sen. George Mitchell.

Let me be clear: The Mitchell Report, as it currently stands, is the Black Sox Scandal of the 21st Century. Despite that, as Bryant notes:

When Mitchell, union head Donald Fehr and baseball commissioner Bud Selig testified in Room 2154 of the Rayburn Building on Jan. 15, Mitchell was praised for his "fair and impartial" report, produced under the difficult circumstances of not having the players' cooperation. Mitchell's approach, methodology, findings and conclusions went unchallenged. No member of the committee probed with specific questions about the information gathering process on Clemens, even though each knew within a month that Clemens, and McNamee, would testify before them.

The normally-astute National Law Journal ran an op-ed two week's ago discussing the legend-in-his-own-mind Mitchell Report. It was a relatively balanced piece—certainly better than the ESPN coverage—but it still suffers from an attempt to be "fair and balanced." Especially egregious is this statement, whose sentiment is assumed throughout the piece:

It appears that Major League Baseball needed to have a George Mitchell not only conduct a vigorous investigation but also identify alleged wrongdoers.

The short answer is, well, yes, that would have been a good idea.

Let's be clear: the Mitchell investigation was not "vigorous" and did not "identify" any wrongdoers who were not found by other means.

The bulk of the "report" is based on two New York-based people who worked in the clubhouses of the Yankees (Brian McNamee) and Mets (Kirk Radomski). Note that the links are both to extant court cases that involve public testimony.

Mitchell had no subpoena power. He started by speaking with people under indictment and people they implicated as having dealt with them directly. His information came (primarily) second-hand from the Mcnamees and Radomskis and Jason Grimsley.

The result was by no means "vigourous"—it couldn't be, given the constraints. The result was what Mitchell always does: a report that made his paymasters (in this case, the baseball owners) happy that they had "identified the problem," that it was relatively limited, and that they themselves were not complicit.

And the thing is, Cohen and Gershman know everything I'm saying. While they open with:

Not surprisingly, the credibility of the now-famous Mitchell Report that publicly identified 85 baseball players as having used illegal drugs is about to be tested

they never support the claim in the report. Indeed, the closest they come is to say that it "would have appeared less credible" without including those 85 often-unsubstantiated names. Indeed, Cohen and Gershman are clear-sighted in the true purpose of the report:

It appears that Major League Baseball needed to have a George Mitchell not only conduct a vigorous investigation but also identify alleged wrongdoers. However, given the protocol he chose, there may be innocent victims whose protestations will be heard but, in most cases, not believed.

At least this time, some lawyers know the truth, and are telling it—even if it is by what they do not say. One can only wish Congress had acknowledged the same.

*Sorry, started this yesterday and don't feel like changing all the tenses throughout. But having now seen The Daily Show medley of clips of people discussing Clemens's bum and its bleeding, I don't feel much of a need to change anything else, either.

Labels: baseball, the rule of law

by Ken Houghton

Via ESPN came the claim that some Penn professors "refute[d]" the report that Roger Clemens's late-career performance for the past decade was not unique.

This gets us to the Freakonomics blog, with a guest post from Justin Wolfers, which gives credence to Daniel Davies description of the book's central conceit:

So, for example, sumo wrestlers are "cheating" because wrestlers who need to win a bout to stay in the top league do statistically significantly better when fighting wrestlers for whom the match is a dead rubber. Not "taking sensible steps to minimise the risk of injury". Not "following unwritten social conventions of the sport". Not anything else, but "cheating", and most likely doing so because of bribery and corruption from betting syndicates.

Why do I have such a reaction? From Wolfers say in his guest post:

There’s a pretty neat trick at work [in the report from Clemens's attorneys]: if you compare Clemens only to those who had a terrific last decade of their careers, then the last decade of Clemens’ career doesn’t look that unusual....

So we put together data on all 31 other pitchers since 1968 who started at least 10 games in at least 15 seasons and have pitched at least 3,000 innings...

To be clear, we don’t know whether Roger Clemens took steroids or not. But to argue that somehow the statistical record proves that he didn’t is simply dishonest, incompetent, or both. If anything, the very same data presented in the report — if analyzed properly — tends to suggest an unusual reversal of fortune for Clemens at around age 36 or 37, which is when the Mitchell Report suggests that, well, something funny was going on. [emphases mine]

and, from the NYT tarringarticle:

Other measures suggest Clemens performed similarly to his contemporaries. But these comparisons do not provide evidence of his innocence; they simply fail to provide evidence of his guilt.

and ends by giving the lie to their own lie:

Statistics provide powerful tools for understanding the world around us, but the value of any analysis invariably comes down to choosing a useful statistic and an appropriate comparison group. Statisticians-for-hire have a tendency to choose comparison groups that support their clients. A careful analysis, and a better informed public, are the best defense against such smoke and mirrors. [again, emphases mine]

You would think, from this, that the evidence presented by the Penn professors and Mr. Wolfers would be somehow sacrosanct, a Holy Grail, a definitive "smackdown" to make Brad DeLong proud.

Instead, we get "all 31 other pitchers since 1968 who started at least 10 games in at least 15 seasons, and pitched at least 3,000 innings." Now, some of these are Hall of Famers: Dennis Eckersley but maybe not Steve Carlton (career began in 1965), and if Carlton, then Doyle Alexander, Charlie Hough, Milt Pappas, Joe Niekro, Chuck Finley, Danny Darwin, Mike Mussina, and Jamie Moyer, to name a few from this site.

In short, most of those 31 pitchers aren't people whose career, up until the age of 33, compares to Clemens. Which is why—not to put too fine a point on it—the Wolfers graphic for walks and hits per innings pitched looks more like an Edgeworth Box than a statistical analysis.

The authors are careful not to name the pitchers in their sample. But they are perfectly willing to cast aspersions on the sample that named pitchers, since the pitchers named—Ryan, Randy Johnson—are generally presumed above suspicion and have early careers rather closer to Clemens's than the Charlie Houghs and Danny Darwins and Jamie Moyers ever were.

So the "statisticians" managed to show that Clemens's final decade is an outlier. But their own graphics shows that Clemens's career is an outlier from their smoothed data. You might think from this that they would know that their comparison was rather suspect.

From their text, they clearly knew it. Also from their text, they didn't and don't care. They got to cast aspersions, got published in the NYT, got pushed on ESPN, and got a nice writeup on the Freakonomics blog.

Let's look at their last sentence again:

A careful analysis, and a better informed public, are the best defense against such smoke and mirrors.

That would be nice. I wonder: do the authors intend to do one?

Labels: baseball, Economics, Journamalism, Statistics

by Ken Houghton

Following in its grand tradition of assuming the players are solely responsible for the MLB incentive system, MLB acted in completely expected fashion today:

Bud Selig was given a three-year extension as baseball commissioner through the 2012 season.

The unanimous decision made at Thursday's owners' meeting came two days after Selig and union head Donald Fehr testified before a congressional committee that both criticized baseball for its steroids problem and praised it for strides made the past two years.

That's two days after his "we weren't responsible, even though I 'take responsibility' schpiel:

As I said in my statement, I’ve thought about this thousands of times. I’ve been in this sport all my adult life. I agonize over that, because I consider myself, at the end, a baseball man. In the ’90s — you know, hindsight is always very beneficial. I watched things. I re-read all the articles that Senator Mitchell had. I take responsibiility for everything, let’s understand that. I take it –- for all the good things that have happened to make the sport as popular as it is today, and when we talk about something negative, there’s no question about that. I’ve agonized.

But I would also remind you, and who knows how long this has gone on –- the Senator said over 20 years –- we have come a long way in a difficult environment. My minor league program is going into its eighth year. So all of the great players in this sport have been tested eight years.

Do I wish we had reacted quicker? Should we have? Yes, one can make a compelling case. And I do a lot of introspective thinking, and I’ll second-guess myself. As far as responsibility, all of us have to take responsibility.

Yep, "all of us" have to take responsibility. The buck stops...with Barry Bonds? And there's good reason to doubt that those "thousands of times" have been for more than a millisecond:

2005 (March 17): “Do we have a major problem? No.”

2008 (Around 1:05 p.m.): “As far as I’m concerned, I don’t have a scintilla of doubt that the use of performance-enhancing drugs is a very serious matter for this sport — at its core. At its core.”

...Selig was asked if baseball could change its culture, in which players may be tacitly encouraged to do steroids, if only to stay competitive with those who already have. He responded: “I have a lot more confidence than I did three years ago.”

Fortunately for Bud S., but not for us, Congresscritters ask questions the same way they are asked them by the Tim Russerts and John Kings of the world:

He was not asked in follow-up questioning how that statement could square with the one before.

The late Doug Pappas's countdown clock needs to be updated, and we are all the poorer for that.

Labels: baseball, Risk Management

by Ken Houghton

Cokehead's favorite baseball player on Slappy and the "GREATEST MANIFESTATION OF EVIL ON THE FACE OF THE EARTH":

"Between them and the Yankees making sure we were updated every 15 minutes about when they were actually going to name their manager, I didn't give a crap. Bottom line was they're playing golf and making organizational decisions and we're still playing games."

Labels: baseball, LGandM, Yankees

by Ken Houghton

With Jeremy and Erin both celebrating, and even The Curse of Scott only having extended their playoff life one game, it's well past time for the final Yankees v. Mathematician v. Gamblers post of the year.

My original post was all too accurate, and the absurdity (not Joe Torre's decision) of starting Wang* on three days's rest instead of Mike Mussina only proved the initial point.

In the end, though, the Yankees ended up 3.5 games below where the bettors had them at the beginning of the year—which is quite an accomplishment, given the injuries and changes the team went through this year.

I freely admit being no fan of Joe Torre's;I remember his Mets team that battled Roger Maris for the single-season home run title. I generally give most of the credit for Torre's "success with the Yankees" to Woody Woodward and Brian Cashman. But even I have to admit that this year's team—aging, fading, increasingly weak defense, and without the dominant closer than Rivera was since the year after Jeffrey Maier—overachieved in retrospect.

By contrast, the mathematician predicted the Mets to win 90 games, two more than they did and virtually the same as the "over 89 1/2" SportsMemo predicted.

So we have one team that achieved 3.5 less wins than it was predicted to, in the face of its own injuries, and one that achieved 1.5 less despite major injuries to its closest rivals and which ended with Carlos Delgado declaring that they are "so good they get bored," even as they lose nine of their last ten home games.

If you're asking a fan which one was motivated, they can tell you the answer. Seems a pity that management doesn't know it.

The previous posts in the series are one, two (really a throwaway), three, four, and five.


*Or, as FoxSports referred to him, BLEEP.

Labels: baseball, LGandM, sabremetrics, Yankees

by Ken Houghton

The Avon girls* were passing out St. Derek of Pasta Diving (tm - Scott L) new cologne, Driven,** so I had to visit LG&M Thursday.

Well worth the trip.*** It's nice to see that that messy "Iraq" thing has been solved and we can go back to important work:

[Senator Mitch McConnell]’s “deeply disappointed” the new Democratic-controlled Congress has not tackled the “tough issues,” like Social Security…

Besides their “preoccupation” with the Iraq war, Democrats spend their time dreaming of “taxation, regulation and litigation. It’s in their DNA.”

It's a good thing, as I noted Wednesday, that Jane GaltMegan McArdle assures us that we can ignore anything a politician says; otherwise, I might start thinking that the Republicans are trying to distract us from the war and pay for their exorbitant spending with the Greenspan Committee's payroll tax increase that both (1) was scheduled to pay the Boomer retirements and (2) even now was rather accurate in its demographic projections, and moderately pessimistic about productivity and revenue gains.

Headline Reference should be clear to EOB readers. See here for album picture.


*Phrase used advisedly. While at my age anyone Suzanne's age or younger can fairly be considered a "young woman," these two appeared barely old enough to carry a Trapper-Keeper.

**The cologne was named, one guesses, because it is the answer (along with "He was...") to "How did Derek Jeter get to a ball hit four steps to his right?"

***I have to assume that this

Given rural Kentuckans' well known hatred for Federal entitlement programs, dismantling Social Security is a sure winner for Mitch, but it's still kind of surprising to see a Republican in a border state try to shift the focus from the war to domestic issues...

is a joke (possibly unintentional), since without those Federal entitlement payments, the state of Kentucky would be at least $500MM a year poorer [PDF]. If we assume they are rational, then cutting SocSec and Medicare payments should be the last thing on a Kentuckian's mind.

Labels: baseball, Eat Our Brains, Iraq, LGandM, policy wonk, Politics, social security

by Anonymous

Quinn is obsessed with baseball. He's decided that he's going to be the third baseman for the Mariners when he grows up. We're encouraging this, of course, not only because of the thought of being kept in our old age in a style to which we haven't become accustomed, but because this option isn't really all that much less likely than his prior career choices: 3 (at age 2), a jack-o-lantern (at age 3), and the driver for the M&M [nas]car (at age 4).

At any rate, he spends a considerable amount of his considerable free time listing the teams. One of yesterday's lists was by mascot, and was broken down into five types: bird teams (Orioles, Blue Jays, Cardinals), fish teams (Devil Rays, Marlins), mammal teams (e.g., Cubs, Tigers), "people" or occupation teams (e.g., Mariners, Pirates, Brewers, etc.), and others (e.g., Reds, Rockies).

When he got to the Los Angeles Angels at Anaheim, or whatever they're called today, Quinn announced without hesitation, "oh, that's a fish team." I stifled a smile and gave myself a virtual pat on the back for a parenting job well done.

One wonders how he reconciles the halo on the team gear, not to mention the absence of any angelfish, with his world view.

Labels: baseball, Family