Hello all! Once again it's your friendly neighborhood Leprechaun back at it for another MLB season. Now that we're almost a third of the way through the season, I think there are enough stats to truly start a performance based system. During the offseason this year, I worked with a programmer to automate this system. It allowed me to add in some stats to my equations that weren't possible to calculate by hand everyday. I've been able to tweak my formulas and run entire season, game by game simulations that I hope prepare this system to do much better this year.
I ran this system here on this forum last year. You can find the thread here:
If you read the first couple posts, it explains a lot, however there are a few big changes to how the system will run this year so I'll give a "short" explanation taken from last years thread with some modifications.
This system is based on how teams perform in a 2, 3, or 4 game series at home or on the road. Based on a rather complicated formula, we chose a team to win 1, 2, or 3 games in a given series. That means, if our team wins game A, we sometimes go for a second or even a third win in that same series. Also, this system does not chase to win just one unit per series. We use stats to predict our probability and unit size for each play. The plays are decided based on the matchups. This system is dynamic so there are no set teams to play on. As the season progresses, we collect team data that will put matchups in or out of the system. That means we don't always start betting on the first game of a series and we don't always chase after a team in a series if we lose the first game. As each day is completed and the results are processed, the factors that decide the plays for each team changes. I think this is the most important part of the system and one of the major reasons it has such great potential. You are relying on up to the minute data. The team to play on and if you play the road or home team, is completely dependent on the numbers. That means teams will move in and out of the system depending on 1)how good or bad they do, 2) how good or bad their opponent does, and 3) how good or bad the league does AS THE SEASON PROGRESSES.
Here’s a summary of the big differences from last year:
1. We will not always be chasing for team to at least one game in a series. We will play teams that fit in the system but because there are many other factors in the equation this year a team might fit the system for the first game of a series but fall out of the system for the second game and vice versa.
2. Because of #1, we will not keep a series record, but only a straight up record. In addition, there will be no game A, B, or C listed with each play. There will be many times where we play a team that lost the previous day, but it won't be a straight chase by any means.
3. Strength of plays will take the line into account. We did not do this last year. We simply played whatever the system dictated and laid the chalk. This system picks a lot of favorites (although you'll see a lot more dogs this year), so this new rule is important to keep volatility down during losing streaks
4. We will very rarely play a team on the ML if the line is -180 or higher. Most often, we'll play the RL in that situation or skip the play all together. Last year's data was pretty solid in teaching us that playing the RL is worth it. The amount you save on the losses far outweighs the games that are won by one run.
If you're looking for extra info, you can always check my site (easy to find in my profile). Don't worry, the site is not for pay and I have no intentions of ever going that route...it's just where I centralize all my information. Centralized info works better as I will post this system on a few forums. I like having a place to put info for all that won't get lost in a jumble of threads over the years.You can see all of last years results and there is a lot more info and insight behind the plays sometimes. If you have a question, it's probably answered somewhere over there, but of course I will answer questions in this thread as well.
I don’t promise you’ll win a million units, but I do promise rock solid record keeping, a fun thread, and an interesting project to follow. Unlike many people here, I have no grandiose dreams of becoming a professional gambler and trying to make millions. This is more about my personal passion for sports and the challenge of trying to come up with a system that wins consistently. Hopefully we’ll make a few bucks along the way.
find my picks and systems there!