This is a great question and I would love to know the answer. There are more and more touts these days, and I suspect they have fewer and fewer clients, but I can't say for sure. It's tough for me to understand how there could be a ton of people paying for picks, but on the other hand, never underestimate the power of…

There's always a big lag between information becoming available and people knowing what to do with it. The bigger opportunities (if you don't have any inside info) are right after info becomes public but before people know what the heck is going on. Think Baseball Pitch F/X data right now for example.

There will always be opportunities. People are still coming up with new stuff like figuring out how much influence a catcher has on getting ball and strike calls. Or interpreting Pitch FX data to figure out if a pitcher is to back to his "old self" post injury. That being said, you're absolutely right ... baseball's…

Of course, you'll never have time to built the perfect model and so you have to get practical about what your model is going to take into consideration, and what you won't bother with. When I was betting baseball, there was a group I heard about called the Pokers (I believe they were originally funded by poker…

Now basically almost anything that can happen in a baseball game can be estimated like this. It gets a even more complicated once you start looking at balls in play, because of the issues of fielders and baserunning, but the concept is the same. And once you look at it that way, it's almost like there is an optimal…

The thing is, if you work at it, you can actually do a really good job of estimating this. You'd look at Ian Kinsler's historical homerun rate, putting quite a bit more weight on recent seasons and do the same for Shields' rate of allowing homers. You'd regress these numbers to the mean, probably moreso for the…

Baseball is a nice game to model, because it's a series of discrete matchups, and there is a lot of relevant historical data that can help you estimate the probabilities of various outcomes in each matchup. Let's say, for example, that it's Opening Day this year and Ian Kinsler is stepping in against James Shields,…

Baseball is a nice game to model, because it's a series of discrete matchups, and there is a lot of relevant historical data that can help you estimate the probabilities of various outcomes in each matchup. Let's say, for example, that it's Opening Day this year and Ian Kinsler is stepping in against James Shields,…

Bill Benter. A guy who's made gazillions betting horses with a statistical model and seems open to discussing it.

Statistical analysis can help you identify profitable opportunities. I think of arbitrage as meaning being able to do trades (bets) where you lock in profit automatically. That was sometimes possible online 13-14 years ago when there was a ton of competition, but pretty rare to see those opportunities now.

That being said, I don't mean to deflect any of the blame from myself. I could have done a better job of prepping ... i.e. saying, hey let's talk a little bit about building a fundamental model for baseball because that's some I actually have experience with.

Yeah, I have. I think a great deal of that is the nature of the business. It's a gambling forum and information is money. It's going to be really, really tough to get people to open up and share great info. Sloan hit the jackpot last year when Bob Voulgaris felt like coming in to give crap to one of the Cantor…

Okay! Just got here ...