Hooligans Sportsbook

Matty Rain's 2024 MLB Betting Thread

No no, I actually just added home and away yesterday. It doesn't have a huge effect outside of Rockies games. I'm still fucking around with the Rockies but I think I'm happy with the value I decided on.

It's just a big-picture model based on WAR and adjusted every game based on active players, starting pitchers and bullpens. It spits out implied odds and then you go line shopping. The rest is just staking and discipline.

Example: the model says Zac Gallen and the Diamondbacks have a 76.9% chance of beating the Rockies, which translates to a -333 moneyline. The actual price of -235 is deemed +EV so we bet that. That's it.
 
Last edited:
No no it's ok. Like I said this is an experiment and I expect to keep making slight tweaks to the model as the season progresses and my learning evolves. All the above are at worst neutral EV right now. Won't matter over the long run.
 
No no, I actually just added home and away yesterday. It doesn't have a huge effect outside of Rockies games. I'm still fucking around with the Rockies but I think I'm happy with the value I decided on.

It's just a big-picture model based on WAR and adjusted every game based on active players, starting pitchers and bullpens. It spits out implied odds and then you go line shopping. The rest is just staking and discipline.

Example: the model says Zac Gallen and the Diamondbacks have a 76.9% chance of beating the Rockies, which translates to a -333 moneyline. The actual price of -235 is deemed +EV so we bet that. That's it.
Perfect analysis of Colorado. They suck less at home.
 
Matty, not one to offend anyone here, just wondering if you're okay with outdated Wally trainwreck smilie being replaced or updated to 2024 GL status? Hoping against it but might come in handy towards the end of summer? Just sayin! :dunno:
:baltimorebridge:
 
I believe in you Matty

Listen to this words from MrX

MrX
I think the efficiency of betting markets is widely overstated. On the other hand, the betting markets do move towards efficiency. There are many approaches that would have worked wonderfully 10 years ago that are useless now. But, as new information and greater processing power becomes available, there are new approaches that wouldn't have even been possible ten years ago. I think we're still pretty far from anything approaching perfect efficiency, where closing lines would be unbeatable.

Maybe it's time to use the new GPTs processing power
 
I remember MrX telling me that he wasn't that focused on prices.

I had GPT write me a Python script to scrape some data off baseball-reference.com, until I realized that the data could be exported directly from Fangraphs.

I wonder if and how bookies/linesmakers are using AI.
 
Last edited: