I'm Mark, been dabbling in sports betting for a few years now. I don't gamble for/with any amount of money that will make a difference to my lifestyle...what I am is a guy who really likes challenging games, and sports betting is one of the toughest of em' all. Billion-dollar Vegas versus us, the little guys. It's like sitting across the chessboard from Bobby Fischer and still really wanting to win.
I make my picks in a number of ways. I'm a big believer in numbers and statistics...so many things on this planet that can appear to be somewhat random you realize after observing for awhile that it can all be quantified...and so it is true with sports.
But that's only one angle, and a weak one at that. Vegas can afford much better statisticians than I'll ever be. So predicting that Team A has a 30% chance of winning a game is one thing, but more importantly is figuring out whether that 30% chance is more likely today than other days. Not easy to do.
Another important aspect of picking to me is to disregard a team's overall record; if a #1 ranked team is playing a #2 ranked team, who cares how either fared against the #15 through the #30 ranked teams? I almost entirely disregard that useless data, and I feel a lot of people do too, to some degree, but not enough.
Finally, I make my picks by reading, by borrowing from those who know better than I do. I don't watch every NHL game...but some people do. A big part of my gambling day is finding out who the hot cappers are and seeing what they think. Often if I have a play I kind of like, and the best cappers are all leaning for it, I'll play it, and the inverse is true too.
This year is my first year dedicating so much time to the game...until now I had just casually wagered a little here and there, with much less success. This year I hope to develop programs to help compute teams' probabilities of winning, the final score and the margin of victory. Already for the NBA I have a program that across 89 games has predicted the SU winner correctly 71% of the time where Vegas ran 50% across those same games. My average team total score was only off by 1.1 points, and game total score only off by 2.2 points. Vegas did better on scores, going 0.6 points off on the team totals and 1 points off on the game total (some numbers are rounded, but you get the idea.) I'm still working on margin of victory (point spread) calculations, but am testing one right now that so far is 100% (1-0!) haha. Alas though, the NBA season is winding down, motivation is less predictable, and so this program will have to go on the shelf until next year in just a few more days.
What I take from that though is knowledge on how to play the game, the game of wagering on sports. Predicting SU winners was my specialty using the program in the NBA, and with the MLB season kicking off I think I will have a much easier time constructing a program to pick winners there. I'm going to give it a few weeks for the samples to build up, but once we get rolling I feel pretty confident about edging out a profit. Will do something like similar pitchers versus similar batters, the average innings pitched by the starter versus the bullpen, etc. etc. Coupling that with a few sharps I've found I really like my chances.
Anyhow, I won't be using this blog all that much; I just figured it would be a good way to give an introduction. I'm attaching a photo of my NBA program so you can see what all goes into my predictions...the weighting is what's important, and those numbers aren't included, as that's what can really truly be classified as unique intellectual property. What you may notice is that I use a Pythag calculation (if you don't know what that is you'll want to search the internet); the Pythag is a fine example of statistics that can hurt you if you take them simply on their own...if you know a team has a 41.2% chance of winning (like team #1 in the photo) the trick will be knowing when that 41.2% is most likely to occur. The photo shows Vegas has team 1 at 25.4%...what does Vegas know that the Pythag doesn't? A lot. And that's where the rest of the program comes in. After considering many factors, such as how a team performs on home/away versus similar teams, days rest, strength of wins, injuries, the Pythag gets washed down to 36.4%, showing value in the dawg in this case, at least on the ML.
Then finally, knowing this value is here I take to the internet and see what the sharps are saying. I track my sharps in another program and know who's hot and who's cold, and try to ride the appropriate waves. When my lean is favored by the sharps, we've got a play.
And the result of all this? In the NBA it wouldn't be right to say, as the program has been being tested/re-vamped simultaneously, and so the sample doesn't represent the ability of the current program. But weaker versions ran 71% to Vegas' 50%, and I can only see it getting better from there. With all this I think pulling off a 5% profit over 100 samples is possible...so for the NBA we'll have to wait and see until next year. And for the MLB, what with all the skewed lines thanks to a naive public, again I expect to be able to yield a nice 5%.
I guess that's all there is to say about my method and motivation. As for short-falls, I think my biggest threat is temptation...I see many games that seem okay, or even good, but playing these weaker plays is really a fine way to wash away profits. Discipline is a major thing I'll be working on, trying to limit myself to playing only the cream of the crop, isolating the profitable plays and shedding the rest.
Doubt anybody ever reads this, but maybe if I make it big-time and beat the game some will poke around someday and find it of some interest. Just felt wrong having a space with no real information about the better. Thanks for reading!