Steven Engelhardt

The problems with my regression which “predicts” a baseball team’s win-loss record are the following:

1. The regression is extremely sensitive to the input parameters, many of which are quite unstable throughout the year. For example, one extra point (0.001) of team OBP results in 0.6 extra predicted wins.
2. The standard error of the prediction is quite high (4.68).
3. The regression doesn’t seem to be much better than linearly extrapolating a team’s current win-loss record to the entire season.

To demonstrate my point, a run of the regression against the current Cubs team statistics (0.371 team OBP, 3.81 team ERA) now predicts a record of 113-49, a swing of 23 wins in a matter of 15 days. This record would put them 3 wins shy of the regular season win record—first achieved by the 1906 Cubs (in a 152-game season) and again by Lou Piniella’s 2001 Seattle Mariners.

It’s clear that this new prediction is the consequence of the Cubs’ hot streak. Let’s hope they can keep it going.

Well, the original source of the error was somebody the WSJ quoted, but they took it the next step without correction:

Until recently, for instance, the Lincoln Navigator offered 128 options on its console alone.

“You know what 128-factorial is—it’s a lot of combinations,” Mr. Mulally joked at a conference recently, mocking the number of designs theoretically resulting from mixing-and-matching the options. (Answer: 3.85620482 x 10 to the 215th power.)

Spector, Mike. “Ford Eyes More Cuts as Recovery Advances; Earnings Improve”. The Wall Street Journal 23 April 2008: A1, A14.

128-factorial is the correct number of combinations of options only if the order in which you pick the options is important. It almost certainly is not. Instead, assuming each option is independent and can be either on or off, the correct number of combinations is 2^128—3.4 x 10^38. Mr. Mulally was off by a factor of 177.

Roger Lowenstein has written a fascinating article for the New York Times Magazine entitled Triple-A Failure which explores the role of rating agencies in the credit crisis. It even explains the process rating of an actual mortgage-backed security:

The business of assigning a rating to a mortgage security is a complicated affair, and Moody’s recently was willing to walk me through an actual mortgage-backed security step by step. I was led down a carpeted hallway to a well-appointed conference room to meet with three specialists in mortgage-backed paper. Moody’s was fair-minded in choosing an example; the case they showed me, which they masked with the name “Subprime XYZ,” was a pool of 2,393 mortgages with a total face value of $430 million.

…

Moody’s did not have access to the individual loan files, much less did it communicate with the borrowers or try to verify the information they provided in their loan applications. “We aren’t loan officers,” Claire Robinson, a 20-year veteran who is in charge of asset-backed finance for Moody’s, told me. “Our expertise is as statisticians on an aggregate basis. We want to know, of 1,000 individuals, based on historical performance, what percent will pay their loans?”

…

In the frenetic, deal-happy climate of 2006, the Moody’s analyst had only a single day to process the credit data from the bank.

…

Mortgage-backed securities like those in Subprime XYZ were not the terminus of the great mortgage machine. They were, in fact, building blocks for even more esoteric vehicles known as collateralized debt obligations, or C.D.O.’s.

We built a financial house of cards on ratings agencies’ models, and these models turned out to be disastrously wrong.

Guess who People magazine voted the Sexiest Man Alive in 1992?

Read the rest of this entry »

In the course of preparing for Level 1 of the CFA Exam, I learned a few interesting things about the Fed I did not previously know:

• The Fed does not directly control the Federal funds rate. Instead, the Fed publishes a target for the Federal funds rate and attempts to hit it by using open market operations.
• It is a relatively recent development for the Fed to target the Federal funds rate. Before this, the Fed would often target the money supply.
• The Federal Reserve Bank of New York has a special place above all other regional Federal Reserve Banks. The representative from the New York Fed is a permanent member of the Federal Open Market Committee (FOMC), and all open market operations are conducted by the New York Fed under direction from the FOMC.

While reading the Wall Street Journal this morning, I thought I caught them in a (relatively rare) factual error:

[Barack Obama's] subsequent [2004 U.S. Senate] victory was helped along by newspaper disclosures of embarrassing material from the divorce papers of a Democratic opponent.

Kaufman, Jonathan. “For Obama, Chicago Days Honed Tactics“. The Wall Street Journal 21 April 2008: A1.

I thought this passage referred to Jack Ryan, the at-the-time inevitable Republican nominee. Ryan’s candidacy was destroyed as his divorce papers revealed that his ex-wife alleged he forced her to go to sex clubs. However, the article later indicated that this passage referred to Blair Hull, a Democratic primary candidate whose divorce papers revealed he allegedly beat his ex-wife.

The Wall Street Journal did not escape completely unscathed, however, as I spotted this passage from the same article:

[In 2005, Obama] and his wife bought a mansion in Hyde Park for $1.65 million, $300,000 below the asking price.

Kaufman, Jonathan. “For Obama, Chicago Days Honed Tactics“. The Wall Street Journal 21 April 2008: A11.

Obama’s house is in Kenwood, not Hyde Park.

Changes in relative prices lead consumers to change the items they buy. For example, if the price of beef rises and the price of chicken remains unchanged, people buy more chicken and less beef. Suppose they switch from beef to chicken on a scale that provides the same amount of protein and the same enjoyment as before and their expenditure is the same as before. The price of protein has not changed. But because it ignores the substitution of chicken for beef, the CPI says the price of protein has increased.

CFA Institute, comp. Economics: CFA Program Curriculum, Volume 2. Boston: Pearson Custom Publishing, 2008.

I hate this argument. It’s no different from saying:

Because the price of beef went up, I bought less beef, so the price of beef didn’t go up.

Furthermore, “provides … the same enjoyment as before” is impossible to measure and unlikely to ever be achieved. I like beef more than chicken; no amount of chicken will ever provide me the same enjoyment as a good ribeye.

An article was brought to my attention today that indicated that spot and futures prices are failing to converge in certain futures markets. Below is an excerpt:

A futures contract is an agreement to deliver a specific amount of a commodity — 5,000 bushels of wheat, say — on a certain date in the future. Such contracts are important hedging tools for farmers, grain elevators, commodity processors and anyone with a stake in future grain prices. A futures contract that calls for delivery of wheat in July may trade for more or less for each bushel than today’s cash market price. But as each day goes by, its price should move a bit closer to that day’s cash price. And on expiration day, when the bushels of wheat covered by that futures contract are due for delivery, their price should very nearly match the price in the cash market, allowing for a little market friction or major delivery disruptions like Hurricane Katrina.

But on dozens of occasions since early 2006, the futures contracts for corn, wheat and soybeans have expired at a price that was much higher than that day’s cash price for those grains.

For example, soybean futures contracts expired in July at a price of $9.13 a bushel, which was 80 cents higher than the cash price that day, Professor Irwin said. In August, the futures expired at $8.62, or 68 cents above the cash price, and in September, the expiration price was $9.43, or 78 cents above the cash price.

Corn has been similarly eccentric. A corn futures contract expired last September at $3.36, which was a remarkable 55 cents above the cash price, but the contract that expired in March 2007 was roughly even with the cash price.

Henriques, Diana B. “Odd Crop Prices Defy Economics“. The New York Times: 28 March 2008.

Why aren’t arbitrageurs buying grain on the spot market, selling futures, and making delivery?

I may be going out on a limb here so early in the season, but I just ran the Cubs’ current team OBP of 0.339 and ERA of 4.03 through the regression to come up with my prediction for this year’s win-loss record: 90-72.

Update 2008-04-08 11:13AM: Looks like I’m in good company. Baseball Prospectus is predicting 91-71.

1. Diagnosing IIS NTLM authentication failures is a real pain, until…
2. You use Microsoft’s Authentication and Access Control Diagnostics package, which can tell you…
3. You need to enable HTTP Keep-Alives for NTLM to work.