Right. I think there’s a decent chance your model might end up “chasing” improvement and results. Not a knock on your model in the least, but more a comment on the strength of this staff, their ability to get the most out of guys, and the roster as constructed.
I'm not a gambler and have no idea of the specifics of Manhattan's model other than what he has posted here but when it comes to people who are gambling as a business the model doesn't have to be perfect.
No gambler, as long as the game is fairly clean and he/she is gambling relatively frequently, is going to win on every bet. It just doesn't happen. That is why one of the first rules of gambling is never bet money that you can't afford to lose because no matter how good at it you are you will sometimes lose a bet.
For those who are doing it as a business they don't need a model to hit 100% of the time. What they need is a model that gives them enough of an edge that over a series of bets they will win often enough to cover their losses, cover their associated cost, and make enough on top to make being in the game worthwhile.
Manhattan could give a better number and that number will vary depending on the individual doing the gambling but if a model were to be correct say 65% of the time for someone betting five figures a week or above the return would solid. Others may be willing to go with say 60% or may not be comfortable below 70%.
There are always variables that can't be accounted for. Things like injuries, the bounce of the ball, etc. but if the model gives a consistent advantage over time then it does it's job.
I am not a large investor but the stock market is similar. I currently hold positions in 19 companies. Not all of those companies are or will make me money, some I will take a loss on. What is important though is that I have enough stocks making me money to make up for those losses plus enough more that I come out consistently ahead. There are some years I lose money but over time my average return is better than 11.5% so my system is working well for me.
The gambler looks at things from a similar vantage point with two differences. One is that the risks of losing money in gambling is much higher but so is the potential return. Second difference is that in stocks a significant portion of the gains are based on inputs from the companies represented by the stocks. Company value rises drawing more money to the stock, dividends are paid attracting investors, many of whom re-invest the dividends as I do. In gambling the book takes out a rake rather than adding in. The gain in value mostly comes from those gamblers who are doing it as recreation rather than as a business, the majority of whom lose money over time in the net.
Those recreational gamblers include some who think they are doing it as a business but don't have the knowledge and direction to win consistently. Many more are the one who lay down bets on "their" teams to "add a little fun" to the game.
My hope is that a lot of Nebraska supporters are contributing money that makes other gamblers happier this week.