Bad Conclusions about Fixing Need to Get Fixed
Joe Duffy (JoeDuffy.net)
Wolfers asserts that point shaving is occurring in about five percent of games that involve large pointspreads. If only the good professor read any of our sports gaming articles such as “It is Good to Pick Bad” he wouldn’t have come up with such preposterous conclusions, knowing there is a very logical reason large underdogs cover more than 50 percent of the time.
To the surprise of no sharp sports handicapper, Mr. Wolfers found that big underdogs cover a disproportionate percentage of close spread outcomes. Wolfers’ statistical “big dog” is a team getting 12 or more points. As the NY Times puts it “There is a strange dearth of games in which 12-point favorites win by, say, 13 or 16 points. And there are a lot of games that they win by 11 points or slightly less. There is just no good explanation for this.”
Poppycock. While I am sure the Ivy League lecturer is a competent economist, he needs to take Professor Joe Duffy’s “Sports Handicapping for Dummies” class, ah I guess whenever I offer one.
Without question the first factoid, on the first day of class will be that the point spread is not and will never be a prediction on the outcome of the game. It is a “prediction” on what the spread needs to be to get as close to an equal amount of money on each side. Second factoid in my lecture will be that the public loves betting superior against bottom shelf squads and this of great magnitude and indisputable fact is accounted for in the line.
Professor Wolfers research paper claims the spread is “a market-based forecast of home team's winning margin”. Don’t get caught up in semantics as the “winning margin” could also be a negative number. Damn numbers crunchers. However his misconception of what the spread actually is proves fatal to his conclusions. The “market based forecast” is predicting perception, not eventual reality.
Let me give you an analogy. Let’s say Professor Wolfers wanted to do a study on which movie critics are the most and least accurate at predicting the Academy Award winners. He studied 20 critics’ lists from the last 15 years. But as an example, 14 of the critics were listing who they felt should win, while six were actually predicting who they think would win. In many cases there is a huge differentiation. But if the professors’ study were oblivious to this, it makes his entire paper greatly flawed if not totally irrelevant.
As we mention in the above “It is Good to Pick Bad” article, I have flat out had gamblers tell me, “If I am going to lose, I’d rather lose betting (the vastly superior team) than betting on (the grossly inferior team).” Never in my life have I heard a more self-fulfilling prophecy.
Quite cognizant of the betting patterns of the gambler, the oddsmakers adjust accordingly. If Duke is playing
We exploit that indisputable fact on a regular basis with our “official outlaw line”. Premium customers are told several times a week that many plays are based on the deviation and that “The official outlaw line is what the odds would be if there was no consideration given to public perception and balancing the action. It is the more accurate line from the standpoint of the bettor.”
Well Professor, if you only asked I could have told you with large pointspreads there is an unbalanced digression when it comes to “off lines”. It is because of public betting inclinations that big dogs get too many points more often than any subset. Here is gambling economics—if big favorites covered 50 percent of the time, the books would lose money because square players love betting big favorites. As a result one has to pay retail not wholesale often when betting big favorites by laying an extra point or two.
So “off lines” will involve what should be a 14 point spread instead being 16 much more often than four being six or two points. The difference between 14 and 16 points means a lot of big dogs will cover close spread games.
Nothing, I mean nothing in the professor’s paper should surprise any sharp player except his defective supposition. Furthermore I strongly doubt that most fixed games will involve a team covering by only one or two points. Too much is left to chance if that happened.
In my educated opinion, most fixed games would involve covering by large margins, a pattern the Quaker educator admits to not finding with his fallacious forensics.
I’m not claiming in any way that there is no chicanery involved in the outcome of games. But I am saying the only thing Wolfers proved is he has no idea what a point spread is based on.
Ironically my highly unscientific study tells me that most gamblers and handicappers suspect just the opposite in college sports—if anything too often coaches run up the score for no reason.
I know nothing about fixing games, but I suspect those who could, would be wealthy alum who have a lot of influence with the coaching staff. We already know about the improprieties of alumni and boosters in recruiting. I can’t help but highly doubt no booster has asked for any favors in return for those thousand dollar handshakes.
A simple don’t ask, don’t tell conversation could go along the lines of “coach, I am going to make a donation to the athletic program on Monday. I sure will be feeling much more generous though if you guys win this Saturday by 15 or more points. Can you help me feel charitable coach?” Wink.
Do I have any substantiation of my suspicions? No, nothing that rises above the water cooler burden of proof, but I’d argue it’s a notch above what the professor’s so easily refuted evidence presents us.
The biggest irony of his paper is that he may not realize he may have uncovered something even more significant: forensic handicapping.
While a lot of handicappers take off baseball season, some literally many just figuratively, The Center of the Handicapping Universe Godspicks.com is looking forward to another long year of baseball dog and small favorite winners. Joe Duffy’s baseball prowess goes back to his “JD of the ACC” Cadillac Club scorephone days.
No comments:
Post a Comment