Bet the Farm: Man vs. Machine

This week's Bet "The Farm" kicks off with new contributor Zhihao Zhang, a Stanford grad who has worked for Vegas casinos and wrote his Stanford thesis on gambling. Both our prognosticators are on the right side of .500 -- will Zhihao's machine-like efficiency in tracking line movement or Daniel's solo quest to actually know his teams rule the week?

Daniel has generously allowed me to share some of my input regarding Stanford betting lines. Before I delve into UCLA +5 @ Stanford, here's some quick background information about myself as well as insight into my sports betting philosophy.

I've always been interested in the statistics/probability of casino games and interned as a table games analyst at the Wynn for two summers. I graduated in 2008 after a memorable senior year of sports ( That's also approximately when I started getting interested in sports betting.

When it comes to choosing a side, I pay very little attention to players/matchups. I leave that work to people who are a lot more knowledgeable than me. When the sports books set a line, their main goal is to attract equal action on both sides. In their eyes, they have two types of money coming in. Smart money, from individuals/groups who have sophisticated methods to determine what the true line should be and will bet large amounts when they feel they have an edge, and square/public money from the average sports bettor who believes that by watching ESPN, they are able to evaluate a team's chances of covering the spread. Sometimes these two groups have completely opposite takes on a line and my goal is to figure out which side the smart money is on.

This leads to the idea of being a contrarian, or going against the public, which is perfectly illustrated by last week's Stanford-Washington game. As evident from consensus websites, if you polled the average sports fan (or most people on the Bootleg), Washington +7 was akin to the sports books handing out free money. Washington had just toppled the great USC, played a traditional powerhouse LSU tough and was #24 in the AP polls. They would definitely cover and 75% of people in the Yahoo Pick'em pool had Washington winning outright. [Ed: Technology has made it much easier to determine where the public money is coming in, and by implication, the smart money. Wonder if there's profit to be exploited there?\] That opening line alone was enough to alert the contrarian in me and convince me that the smart money was on Stanford to win by a TD since a line that large was begging the public to bet on Washington. Subsequent movements that resulted in the line moving from Stanford -7 to Stanford -9, despite the public being overwhelmingly on Washington, only further reinforced that notion. Another example of taking a contrarian side is betting against Tiger, because as much as I like him, the public loves him even more and there is no way (in my mind at least) that he has a 33 percent chance of winning a major against the best players in the world.

So, where does all this leave us with this week's line, which opened at UCLA +6 @ Stanford and is currently at UCLA +5? Before the season started, the game would have been close to a pick'em (i.e. no team favored), based on a couple of sports books that issued lines for select games in the 2009 college football season. Stanford vs. UCLA wasn't one of those games but there were enough data points to impute a line. Given that UCLA is undefeated and had a nice win on the road against Tennessee, this would imply that Stanford is performing much better than expected. The convincing shellacking of ranked Washington (former Trojan-slayers) could also figure to be resonating in the public's mind. Based on a sampling of online sports books, about 60% of the bets placed on the game have been on UCLA to cover. This is pretty consistent with the line movement and unlike last week, it appears that the smart money does not have a strong opinion either way. Unless there's any late line movement, I figure it will close at UCLA +5, which will give Stanford about a 63 percent chance of winning.

Finally, I'm going to take a page out of Daniel's book and make three picks every week. Since I rarely bet on spreads anymore and instead prefer the futures market because it's less efficient and allows for greater edges with lower variance, this will be a good way to keep track (albeit with a small sample size) of how good my contrarian picks are. Last week, I had Nevada, Oregon and Stanford (ended up 2-1). [Ed: we didn't run a column because we knew this was in the pipeline and were working out the details.] This week I like: Syracuse +6.5 vs. South Florida, Kentucky +16 vs. Alabama, Duke +16.5 vs. Virginia Tech. You might notice that all the teams I'm picking against are highly ranked (Alabama, Virginia Tech) or coming off a big win (South Florida). Such is life as a contrarian.

Back to Daniel now…

Pretty cool to have an insider telling you how to bet, huh? Guess that's why I'm calling this piece man vs. machine – Zhihao has all the insider knowledge and formulas; I'm stuck all alone looking at, God forbid, the actual teams in order to place a bet. Don't cry for me yet, this season hasn't gone half-bad. Here's what we have for this week:

Cherry picking:

Two overs I like: over 47 in that Duke/VT game and over 45.5 in Oregon State/Arizona State. 45.5 is the lowest total of any of this week's games (and is also the Stanford/UCLA total), and which of Virginia Tech, Duke, Oregon State or Arizona State have killer defenses? Similarly, let's take Arkansas and Texas A&M under 66.5. I'm a contrarian too – it's way too early in the season to know much of anything definitive about any of these teams, so when the house puts itself out on a limb by projecting each team to score 33 points, I'm going to happily go under that. I also like Miami at +260 – they have better than a 28 percent chance of winning that game.

This season: 10-9, -1.1 units on the money line.

Survivor Pool

We got lucky with Georgia hanging on vs. Arizona State, but we're in good shape now – saving our Washington States, Cals and USCs until we need them later in the season. As long as there are lopsided non-conference matchups, sign me up, because it eliminates one team instead of two, and this is a long season. If you map out your picks for another 10 weeks, things get dicey quickly.

So this week, same idea: Notre Dame over Washington. I was actually surprised to find the Irish double-digit favorites, but they do have a potent offense even without Michael Floyd that should have its way with Washington's questionable D. Hopefully we get some more upsets somewhere in the chain and more people start losing – it's been pretty boring thus far!

Boot Pool

Not to jinx things, but your hardworking Bootleg insiders appear to know a little bit about football –yours truly, msqueri and now cardinalmaniac08 (Zhihao) find themselves no worse than tied for fifth in a pool of 56 or so. We're all running ahead of the Vegas lines too. This week, I'm putting Georgia over LSU and Michigan over little brother, ahem, Michigan State, as my one and two, Michigan because they're an underdog (though I think QB Tate Forcier will be good to play, which would move the line to a pick'em I'd think) and Georgia because we know nothing about LSU. I've got Oregon as an obvious ten over Washington State and Notre Dame as my nine over Washington, as those are this week's only double-digit lines. Of my remaining picks, I'm putting my two projected winners with biggest history of flakiness at the bottom. Arizona State is obviously a three then over Oregon State. Stanford, welcome to the four spot – I'm still scarred by collapses of yesteryear against UCLA. Oklahoma, with a healthy Sam Bradford and a sick defense, is a legit top-five team – I like them over Miami big. Eight. Florida State is the biggest mismatch left, they're my seven at Boston College (very overrated this season – they're not that good), and next is Mississippi over Vanderbilt at six. That leaves USC over Cal at five, because while USC is better, either team could come out on all cylinders or totally choking for air – and we've seen both sides of both squads already. Honestly, games three through seven are a total jumble, which is how it should be this time of year. Here's hoping to some luck from the football gods.

Big Three
Oklahoma -7.5 at Miami
USC – 5.5 at Cal
LSU +3 at Georgia

Georgia's shown its warts both offensively and defensively in battles with Arizona State, Arkansas, South Carolina and Oklahoma State. LSU squeaked by both Washington and Mississippi State, the two toughest teams it's faced. Reminds me of one of the funniest comments on The Bootleg. Everyone was calling for the backups, someone asked why, and the commenter said, "Because we don't know that they suck yet." We've seen Georgia suck and so we're underrating them here. They're 3-1 against a good slate, LSU hasn't beat anyone, I'm going with Georgia.

Think what you want about the Trojans, but USC is solid enough defensively to shut down Jahvid Best. They are also uncannily good in big games under Pete Carroll. I don't know if Cal can score if Best is a non-factor, and I do know that a healthy Mays and Barkley could further aid the Trojan cause. USC, lock of the week.

Finally, Oklahoma lost by one to BYU, 14-13. They've pitched two shutouts in their other contests and their D is statistically the best in the nation. Miami has explosive potential, but when they've faced a good D, as in last week in Blacksburg, they got rattled. I think the Bradford brouhaha, Miami's supposed reemergence and Oklahoma's narrow upset all have people overrating the Canes in this one, so contrarian says Boomer Sooner. Plus, I'll be at the game, pulling hard to make this a reality.

This season: 6-6 (plus 10-9 gives the 16-15 overall mark)

Are you fully subscribed to The Bootleg? If not, then you are missing out on all the top Cardinal coverage we provide daily on our award-winning website. Sign up today for the biggest and best in Stanford sports coverage with (sign-up)!

The Bootleg Top Stories