Joel, I don't think there is anything that I do that isn't really analytical. I analyze the games, evaluate the probabilities, gather information about the game, review the screen who is dealing the best numbers, etc. Making a bet involves all of these factors. Unless of course you see an off number then you just bet it and save yourself a bunch of time. Or you instantly recognize a factor isn't incorporated into a line and you can bet it.
It is a solely database system that isn't analytical. It provides information but limited business intelligence. First off, what is the information? The average goals of Columbus scored so far this year? The average goals scored by Columbus vs. Minnesota this year? An accurate hockey dynamic totals model would be a scope of a project that would involve simulations with specification of the players and who is in the lineup, penalties for power plays, ice time, referee and tendencies generally and with these teams if known, likely matchups of lines of one team vs. the other and likely strategies for the game on the date we're talking about. And this simulation would need alot of assumptions that if they're wrong of course would effect the results. A database yields pretty simple analysis. Certainly some interesting information but may not be intelligence.
Now guys who work a DB model on 1st period or 2nd period scoring obviously they might be able to get more intelligence out of it but it is likely the handicapper who can capitalize more by recognizing when the "averages" won't apply to the particular game you're talking about is going to be more successful.
if CB wants to share what his database is doing that would be great. I'd be happy to trash it like everything else and why it didn't apply. That will take me a minute.
This applies beyond CB. I do respect alot of Professor Meyer's work but I had no problem discounting some of his work and I've had some major laughs on this board with Dr. Bob's "perverse" NBA work when he drifts off into some totally inacurrate analysis.
Without any real evidence or discussion I'm assuming this is some simplified analysis of average goals scored through 5 games, etc. Whatever it is, it won't be more sophisticated then the numbers the book comes up to begin with is my guess. I've already clarified that if someone consistently wins, sure we can make an exception. That is not one play or 50 plays or 100 plays.
ANYONE who tells me their models are flagging huge value if you've gambled or invested about one minute in this world if you aren't way more skeptical than excited about that result, you're misguided.
I know I read something about modeling player receptions or completions with the Poisson distribution in one of CB's notes. What a laugh I got out of that no offense.
