OP, I think you might benefit from reading Justin's book:
He has a section on props; start with props and work your way up from there.
He has a section on props; start with props and work your way up from there.
-
Comment -
There is no objective way to determine if a given line was value. What really tells the story is the overall balance after more than 1,000 bets though. If you're in profit then you must have picked value mostly.Is not your "edge" a 50/50 shot in the dark? Many believe the Cardinals @+142 had an edge. I would also say many people thought the Phillies @-152 had and edge too. Where do they come up with that decision? It took me 3 seconds of common sense to see the Cardinals were the "right" side.
If I used the same logic and ran it through whatever filters I have in my mind for every game everyday, I should grind out a profit? Very unlikely because I really can't be sure what the right side is. How do you translate relative strength of the teams and many other relative factors into an estimated moneyline? It is like speculating on a speculation using estimates....and the bankroll is your result.
Other than that heed the advice that has been given to you in this thread. It's invaluable insight.Comment -
well for me the goal is not just profitable but huge profits. i need very high winning %-- like 90%.Comment -
degenerate gamblers always lose in the end, i have a friend who will bet on will there be a hit in the inning just for the rush. These guys spend 3 minutes of research before they place there bet
Comment -
[quote=That Foreign Guy;10537043]That's what she said
only hookers gabble while f*kking
Comment -
Would love to know how you qualify this as a B.S. post.
1. You need an edge, and if you can't calculate your edge, how do you know you have one? How do you know what size bet you should make? Guess you don't know about Gambler's Ruin.
2. As to know if a model you made has an edge and that that edge is true (and not just victim of a small sample size, even of say, 40 bets, where people would start to think it can't be chance that it went 25-15, but they take no time to find out their Z score.) you need to back test the model, or forward test it without betting it.
3. If you can't / won't go for a period of time making no bets because you can't calculate your edge / you don't even know if you have an edge, then you're not trying to win... you're just trying to gamble. Which is fine if that's what you want to accomplish, but that's not what the OP asked.Comment -
Kind of an interesting thread.....
I find it amusing how some people want to actually help while others become demeaning.
Anyway, as for a model, I'm not sure how others "model", but without some sort of factual method how would one define a positive edge? Doesn't it seem logical that if one can determine an outcome with a degree (percentage) of certainty that this would then remove the contest from "gambling" to investing?
My model contains just over 30 pages of data.....all mined from the internet and updated daily using Excel. I'm not a programmer and cannot write code although I am progressing towards that goal. This data is then compiled into basically three sets of information. One set is TEAM STATS such as record, H/A record, record vs L & R handers, and other such info. Another set of data is PITCHER data. This set has all the information about todays starters and how they fair against the competition. It also has pitcher data based upon yearly data, monthly and last 3 game stats as well as bullpen stats. The last set of data is TEAM BATTING stats compiled into columns with headings that include H/A avg's, vs L/R handed pitchers, current and 7 day avg's.
There's more, but I think you get the point. Now, using sabermetrics, log5 and pythagorean calculations, I can compile this data into meaningful stats that has a factual base and then determine a value or line for each game.
Lets say I determine by use of this model, that team "A" has a 55% probability of winning today. However the line is -180. If I bet this scenario over and over, I'll eventulally lose all my money. I simply cannot pay more for a game than its worth, but without a workable model, I'm not sure how I would be able to determine this.
Since May 18 there have been 81 games in which this has occured: One team's pitcher has a FIP rating above his opponent by 1 or more. His overall ranking exceeded his opponent and his ERA in the last 3 games was better than his opponent. The team's batting avg exceed their opponent and in the last 7 games their avg was better than their season avg. In addition, my computed game score showed that team to win by an amount exceeding .51 runs. This scenario wins at a rate of 67%, so whenever this occurs and the line permits, I bet it.
A lot can be learned by compiling a database. DO IT! I know I'm getting a little long-winded, but finally let me say that compiling data and a workable formula for forcasting is benefical and worth the time spent if you have any inclination to make money at this.
Good luck on your action...Comment
