Monday, June 07, 2010
In the meantime, I'm going to try my hand at writing up my travel adventures on a new blog. Hope you find it interesting and provide feedback.
Friday, January 15, 2010
But once in awhile, someone comes along and manages to fool even the luminaries of baseball analysis into thinking he makes a legitimate point.
The comments on the original post do a pretty good job of summing up its flaws, so this won't take too long. Basically, the author took data from one year of MLB games and determined that in that year, there was little or no favorite/underdog bias in the closing lines for each game. So, if your strategy is to bet either the underdog or the favorite in all 2430 games, paying full juice, you're very unlikely to show a profit on the year. His conclusion: no betting model can be demonstrated to beat Vegas.
Needless to say, I was disappointed to learn that the repo men will soon be here for my Lexus and stock portfolio, since they were purchased with money I couldn't possibly have won over the past four years. While I wait for them to show up, though, let's bet on some baseball! I'll book your action, since the house never loses.
The home team wins about 54% of baseball games, making -117 a fair line on the home team. I'll let you take any away team on the board for +100 or any home team for -130. Unless home teams run really hot or cold, I'm going to crush you over the course of the season. Right?
Maybe, if you're a degenerate moron. But let's pretend for a minute that you're not: how would you approach this proposition? You stay away from the close games and save your money for when C.C. Sabathia is facing Livan Hernandez. Assuming you have the discipline not to wager blindly, I will go broke in no time flat even though my book is "unbeatable".
I hear you saying, "Now just a dang minute. Chris Moore is telling me that the only way to bet baseball is to back all 2430 favorites or all 2430 underdogs each season. And what if the line is close to even money and the favorite becomes the dog? Do I have to bet on both teams? HELP!"
Fortunately for the pros, nobody is holding a gun to their heads. They can lay off most games, bet only when they see an edge, and still enjoy a reasonable ROI over a large volume. 'Cause knowledge is power!
I eagerly await Chris Moore's next column, about the MIT Blackjack team and their incredible run of luck at another unbeatable game. Hopefully between then and now someone will explain to him the difference between "your" and "you're".
Postscript: The comments for these articles include extended discussions of season win over/under bets as a potentially profitable alternative to individual games. One popular data point is the 2008 Rays, who were tabbed for 85-90 wins by various forecasts. People are quoting anywhere from 72 to 74 wins as the "Vegas line".
I probably monitored this prop more closely than anyone, since I placed about forty separate wagers on the Over. The line actually opened at 68.5(!) in mid-February at BetCRIS. Within twelve hours I and other sharps had bet it up to 71.5. It grew slowly from there, eventually closing at 76.5*. Was the Over was no longer a profitable play? No, the books were drawing roughly equal action on both sides of 76.5, so they moved it no further. If I hadn't exhausted my futures budget, it would have closed at 78 instead.
mgl commented that we "will NEVER see a line like" this again. This may be true: the 2008 Rays were a perfect storm of young talent bouncing back from a season where they underperformed and got very unlucky. I may never again see a team be so underestimated heading into a season. However, I don't share his general pessimism regarding season wins lines. In 2007, the White Sox over/under was anywhere from 86.5 to 89.5 against a PECOTA forecast of 72 wins. They actually won 72 games. Why didn't this cause the 2008 bookies to open Tampa Bay at 88 wins instead of 68.5, or Seattle at 76 wins instead of 87.5? Because the sports betting market doesn't work that way. Beating these lines is tougher than it was two years ago, but things don't change that quickly.
*Some books adjust the vig on season wins lines instead of the total. At books offering a line of 73.5, the price on the over was -200 or worse.
Wednesday, December 23, 2009
Patriots: Yes -1200 (98%)
Bengals: Yes -1500 (98%)
Steelers: No -330 (87%)
Titans: No -400 (97%)
Texans: No -1800 (98%)
Jaguars: No -400 (85%)
Unfortunately the maximum bet is a risk of $500 on each, but a couple hundred bucks of EV is nothing to sneeze at, plus you can clear $150 of bonus. Happy hunting.
Edit: Throw in Dallas +180 to win their division (43%)
Sunday, December 06, 2009
"History says that Brett Favre throws a lot of interceptions and that Adrian Peterson fumbles the ball. The only stat you can't predict in a football game is the turnover stat, and if there's gonna be an issue with the Minnesota Vikings--we saw glimpses of it tonight--it's gonna be they're going to turn the ball over down the stretch--maybe in the playoffs--and they may be the better team on paper, but I...'cause the turnover history, it may cost them a chance of getting in the Super Bowl."
Friday, August 07, 2009
There is nothing in John Smoltz's 2009 peripherals to suggest he can't be a solid mid-rotation starter for a contender. His 8.33 ERA is a mirage of sample size and pitching in terrible luck. Every contending team currently sports a staff with at least one starter--and several relievers--worse than Smoltz.
Will the GMs of those teams start a bidding war for Smoltz's services? Nope, they're still using ERA to evaluate pitchers. In their defense, it looks like BBTF was fooled as well.
Monday, July 27, 2009
That day has come, courtesy of Baseball Prospectus. Once the home of cutting-edge statistical analysis, they now misinterpret basic statistics to draw inaccurate conclusions.
Here are the article's major points:
- For pitchers with a large discrepancy between FIP and ERA in the first half of the season, the correlation coefficient (r-value) between first-half ERA and second-half ERA is .33, whereas the r-value between first-half FIP and second-half ERA is .35. Thus, ERA is "equally as likely" to indicate performance going forward.
It's true that there is little difference between .33 and .35. However, this statement alone means nothing. Let's look at a hypothetical group of pitchers:
|1H ERA||1H FIP||2H ERA|
The correlation between first-half ERA and second-half ERA is a perfect 1, whereas the correlation between first-half FIP and second-half ERA is a completely imperfect -1: a lower FIP actually indicates a higher ERA going forward!
Does this mean ERA is a better predictor than FIP? Of course not. Anyone can look at the above numbers and see that 2H ERA matches up very well with 1H FIP and not at all with 1H ERA. Yes, this example was contrived, but the same effect is at work with the real numbers. The lesson: Don't believe everything an r-value tells you.
- Pitchers with a discrepancy between their 1H FIP and ERA, as a group, had a 3.34 1H ERA, a 4.64 1H FIP, and a 4.60 2H ERA. This compares to a control group with a 4.40 1H ERA, 4.34 1H FIP, and 4.35 2H ERA.
Now, you might think that this means FIP is way, way better than ERA at predicting future performance. But wait...
- The 2H ERA sample has a higher standard deviation (1.42) than the 1H ERA (0.83) and the 1H/2H FIPs. This explains everything!
It explains nothing. I can't believe I have to point this out, but as the average ERA of a group increases, the standard deviation of ERAs within the group tends to increase with it.
Seidman reminds us that this is the "SAME group" of pitchers. So let's do it his way and make two groups of the SAME pitchers: Group A is every starter's ten best starts from 2008, and Group B is every starter's ten worst starts. Naturally there is going to be a huge discrepancy in group ERA--it might be something like 1.50 for Group A and 8.00 for Group B.
What about the standard deviations for the groups? Should we expect them to be equal, since these are the SAME pitchers? Of course not. Group A is going to contain a lot of ERAs between 1.00 and 2.00, while Group B will be spread more thinly between 6.00 and 11.00.
Similarly, we simply cannot expect a group with a 4.60 ERA to have the same standard deviation as a group with a 3.34 ERA, even if it is the SAME guys. (Okay, I'll stop with the caps now.)
What about the 2H ERA having a higher standard deviation than either FIP sample? ERA naturally has a higher standard deviation than FIP, because FIP has much of ERA's variance stripped from it. The reason 1H ERA has a similar standard deviation to the FIP samples is that the average 1H ERA is much lower than either group's FIP, reducing the standard deviation as we saw above.
Mason Malmuth once wrote that the real handicap of a bad poker book is that the reader cannot distinguish between good advice and bad, and as a result will develop bad habits without knowing it. If BP doesn't screen its content better than this, it's going to suffer from the same problem.
Thursday, June 25, 2009
Keith, You seem pretty jaded about the whole steroid issue, do you have any comments about the example it sets for young people who are heavily involved in athletics?
Well, maybe if the media would stop harping on the subject and implying that steroids make you a superstar, kids wouldn't get the idea that they're worth using.