The proprietor of VegasWatch fortunately happens to be a friend of the show. This is not through any planning of his, and to be sure if he had known that befriending me in college would lead to me publicly linking him to this website, he probably would have reconsidered.
Then again, he responded to my email which contained questions about his website and he dutifully answered them. Here is the result of that communication.
-Can you sum up what your site does in a few sentences and explain how you got the idea for the site?
The site basically tries to take an intelligent, analytical look at gambling on sports. A lot of it is looking at futures odds — which team will win the NCAA tournament, the AL West, etc. — and trying to figure out ways to accurately handicap those events. I focus less on trying to find good bets on individual games, as I think those lines are a lot more efficent, and it’s just generally less interesting. There’s some other stuff on the site too, but the majority of it trues back to that.
A few years ago, the second year that Florida won the tournament, North Carolina was (correctly) seen as one of the best teams in the country. I had no problem with this, but I couldn’t shake the feeling that they were “inconsistent”, and this would hurt them if they had one bad game in March. Now, “consistency” is obviously an idea about which there’s a lot of BS thrown around, but I tried to quantify it using teams’ performance against the spread. I sent this work to a few MSM writers, and SI.com’s Luke Winn ended up including it in one of his articles. After that I kind of figured that if I was going to be doing this stuff anyway I might as well have a place to post it, so I went to blogger.com and that was that.
For the record, that was the year that a horrendous 15-minute stretch ended up costing the Tar Heels a Final Four berth.
-We’ve talked about how you didn’t really begin following college basketball seriously until earlier this decade. Why the sudden interest in that sport specifically?
I’m still not entirely sure. I guess it started my freshman year of college, when I had an absolutely absurd amount of free time on my hands, and started watching a lot of college hoops after baseball season ended (poorly). The whole format of the sport drew me in — gathering information on teams for four months, and then factoring all that into picking the tournament.
-You often use the Vegas lines to modify and refine projections from complex algorithms like PECOTA and the Pomeroy Ratings. I have encountered people who believe the Vegas lines are not accurate because they exploit the popularity of teams and the biases/ignorance of gamblers. What is the reasoning behind using the lines to make predictions?
The idea that Vegas lines are shaded/not accurate is a common one, and I think it really gives people the wrong idea. Yes, a lot of dumb people bet on basketball, and this does effect the line somewhat. But a lot of smart people bet on hoops as well, and they’re betting a lot more money than the guys who don’t know what they’re doing. If the lines were way off because of the average bettor’s biases, we’d all be rich. They’re not though — the margins are very small, particularly when you’re dealing with the closing line, which has been bet to where it “should” be.
-I hear you talk often about the lines “moving” – which, as I understand it, is how much the initial lines changed based on how people bet on them. For those who are unfamiliar, what do these movements tell us?
At about 6pm EST every day, lines are released for the next day’s CBB games. Contrary to the closing lines, these “openers” are not always dead on, so the people generally referred to as sharps (basically professional gamblers who wager large sums of money) bet on the side they feel has value. If Ohio St. is favored by six and a very large bet is placed on them, the line may move to tOSU -6.5 or so.
Line movements also can occur when a large volume of smaller bets, placed by more “average fan” gamblers, come in on the same team, but I’m of the opinion that that is not nearly as common.
-As sites like yours and The Moneyline Journal and others gain in popularity, as well as the increased availability of information elsewhere online, is it going to be harder to find value? Or is the number of casual gamblers so overwhelming it won’t matter?
The second one. Betting on sports is just such a unique situation; so many people think they know enough to make money doing it, and the vast majority are 100% wrong. You look at Villanova being a small favorite on the road against Cincinnati, and it seems SO OBVIOUS that they’re going to cover the spread, you couldn’t talk people out of that no matter how hard you tried.
Even if someone who has the opportunity to reach a broad audience was educating people about this stuff — say, if ESPN’s Chad Millman didn’t suck at his job — I don’t think it’d make a huge difference.
-What’s wrong with the RPI and the AP Poll?
The issue with the RPI poll is that it doesn’t consider margin of victory at all. That’s really it. If it did, it’d be KenPom, and the correlation between NCAA tournament seed and actual skill level would be a lot higher.
The polls have a similar issue with MOV — they don’t completely ignore it, but it’s not into account nearly as much as it should be. Then they have the added handicap of coming up with rankings at the beginning of the year, and then moving teams down a few spots every time they lose. I’m not sure how anyone could think this is an accurate way to judge anything.
-Baseball Prospectus writers have developed a “Secret Sauce” – elements to a baseball team that become disproportionately important in the playoffs, or are the markers of teams more likely to succeed once the playoffs begin. Have you found, or do you suspect, that there is an equivalent to this in college hoops?
Not really. I suspect that performance in road/neutral games is a better predictor or tourney performance than home games, but don’t really have any strong evidence to back that up. Recent performance is more predictive than Nov/Dec performance, but people overrate that anyway so that’s rather unhelpful.
-How far does Purdue go without Hummel, and without him, how would you rank the Big 10’s top 4 teams?
I’m pretty comfortable with a ranking of Ohio St., Wisconsin, Michigan St., Purdue. Since Purdue was the best team in the conference prior to Hummel’s injury, the gap between #1 and #4 still isn’t huge. Unless they win the B10 tournament, the Boilermakers are going to get screwed with their seeding because the rules in this situation are so arbitrary, and they looked crappy when everyone was paying attention in the Michigan St. game. With very little confidence I will say Purdue wins two tourney games and then loses in the Sweet 16.
-In your opinion, which mainstream media NCAAB commentators and analysts are the best and in what capacity? And if you feel comfortable calling people out on the internet – the worst?
I am a HUGE fan of the Musburger-Knight combo. Brent is hilarious and makes both obvious and clever reference to the spread basically every game, and I don’t think you can beat Knight’s insights on basketball strategy. I also like Gus Johnson for the same reasons everyone else does.
Vitale is obviously unbearable, and I think Steve Lavin is pretty bad as well, which is really surprising when you consider how effective he was as UCLA’s coach.
-PECOTA went from being reasonably reliable to having a terrible year in 2009, and as you said there are, “questions about its methodology.” Is this because of Nate Silver leaving, bad luck, or a fundamental flaw going forward? You used to rely on it pretty heavily and I am curious as to whether that will change for you.
With all the changes they’ve made over the past year I have a hard time trusting PECOTA’s results at this point, particularly for projected standings. Nate Silver is a very, very bright individual, and he used to do all of this stuff himself — I think you probably lose a lot when going from that to having to automate it. There was almost certainly some bad luck involved last year, although the variance on their projections was so high (projecting good teams to do unreasonably well, and the opposite for bad teams) that they had some of that coming.
-What do you look for when analyzing pre-season Over/Under win totals for MLB?
The main thing I’ll be looking at is CHONE, although I think those O/Us have gotten a lot more efficient than when TB’s total was 72 or so in 2008. When I run a simulation of the season I’ll likely be relying on the O/Us themselves more heavily than I have in the past.
-Your Weekly NFL Survivor feature proved to be extremely popular. Correct me if I’m wrong but generally you operated off of Vegas lines, popularity of picks, and looking at future schedules. Did you find that you underrated or overrated the importance of any of these components? What did you learn from this exercise?
Yeah, those are the three main factors. I’m still not entirely sure how to factor in future schedule early in the season, when there’s so much uncertainty. I think I may overrate that at times. It’s also important to remember that the consensus numbers at Yahoo! and OFP and such don’t necessarily apply to your specific pool.
From this exercise I learned that getting lucky can be a lot of fun.
-Your bookshelf is dominated by sports literature. Are there any books in particular that you recommend?
The one that immediately comes to mind is “The Soul of Baseball” by Joe Posnanski — he is an incredible writer, and that’s a great story.
-What are your future plans for VegasWatch?
I would like to eventually make enough money writing about sports — either on the blog, freelancing, for some other outlet, or a combination — that I can do it full time. I have no idea if this will ever happen (I suspect that it will require a lucky break or two), but even as a hobby it’s a lot of fun.
Thank you, VegasWatch for allowing us to bask in your dignity, if only temporarily.