Zach
@zachnielsen.bsky.social
Analytics, sports, trying to figure out Bayesian math one step at a time
I've been looking at ways to solve, especially with a focus on over/unders. Basing the calculation on negative binominal didn't work, a grid search for the best parameters is still ongoing. And we do get some interesting probabilities too.
Really fun stuff!
Really fun stuff!
January 7, 2025 at 12:50 AM
I've been looking at ways to solve, especially with a focus on over/unders. Basing the calculation on negative binominal didn't work, a grid search for the best parameters is still ongoing. And we do get some interesting probabilities too.
Really fun stuff!
Really fun stuff!
An example of the type of analysis I'm now producing. I'm not sure about that win probability with an upset rate of 40% or so. But it looks good! More to come around testing and experimentation.
January 5, 2025 at 6:19 AM
An example of the type of analysis I'm now producing. I'm not sure about that win probability with an upset rate of 40% or so. But it looks good! More to come around testing and experimentation.
I think there's an interesting factoid - building a project this complicated. I learned a good lesson about the value of automated testing, which would have saved my bacon and an hour or so.
January 5, 2025 at 6:19 AM
I think there's an interesting factoid - building a project this complicated. I learned a good lesson about the value of automated testing, which would have saved my bacon and an hour or so.
Game Predictions for tomorrow (using the new complicated implementation), and a happy new year!
Moneyline Predictions:
LAK (41.4%) vs NJD (58.6%)
Total Goals Predictions:
LAK vs NJD
Predicted total: 5.7
Under 6.5: 64.5%
Most likely score: 2-3
Moneyline Predictions:
LAK (41.4%) vs NJD (58.6%)
Total Goals Predictions:
LAK vs NJD
Predicted total: 5.7
Under 6.5: 64.5%
Most likely score: 2-3
January 1, 2025 at 6:00 AM
Game Predictions for tomorrow (using the new complicated implementation), and a happy new year!
Moneyline Predictions:
LAK (41.4%) vs NJD (58.6%)
Total Goals Predictions:
LAK vs NJD
Predicted total: 5.7
Under 6.5: 64.5%
Most likely score: 2-3
Moneyline Predictions:
LAK (41.4%) vs NJD (58.6%)
Total Goals Predictions:
LAK vs NJD
Predicted total: 5.7
Under 6.5: 64.5%
Most likely score: 2-3
For today's focus, I wanted to continue pushing on the implementation from the rest of the article - Claude made that pretty quick. I thought it was interesting to look at prediction accuracy with bias of future data (top part) versus only using historical data (bottom section)
January 1, 2025 at 6:00 AM
For today's focus, I wanted to continue pushing on the implementation from the rest of the article - Claude made that pretty quick. I thought it was interesting to look at prediction accuracy with bias of future data (top part) versus only using historical data (bottom section)
So for a first draft, got the following! 3 main differences, will be interesting to see how those shake out.
Odds from Will Hill (for ease of use and speed, will hopefully find a market number programmatically soon).
Happy to take any and all feedback! More to certainly come
Odds from Will Hill (for ease of use and speed, will hopefully find a market number programmatically soon).
Happy to take any and all feedback! More to certainly come
December 31, 2024 at 4:56 AM
So for a first draft, got the following! 3 main differences, will be interesting to see how those shake out.
Odds from Will Hill (for ease of use and speed, will hopefully find a market number programmatically soon).
Happy to take any and all feedback! More to certainly come
Odds from Will Hill (for ease of use and speed, will hopefully find a market number programmatically soon).
Happy to take any and all feedback! More to certainly come
That being said, I've always wanted to work with an Elo model (versus some of the bayesian stuff I've been doing for NFL) so I built one up using the great NHL API, Claude for some of the programming help, and used the metrics from Fivethirtyeight's old NHL model (RIP)
December 31, 2024 at 4:56 AM
That being said, I've always wanted to work with an Elo model (versus some of the bayesian stuff I've been doing for NFL) so I built one up using the great NHL API, Claude for some of the programming help, and used the metrics from Fivethirtyeight's old NHL model (RIP)
Big qualifiers - will try to post everything publicly, none of these models will be profitable for AWHILE (if at all), and this is mostly to scratch my own itch.
And have fun!
And have fun!
December 31, 2024 at 4:56 AM
Big qualifiers - will try to post everything publicly, none of these models will be profitable for AWHILE (if at all), and this is mostly to scratch my own itch.
And have fun!
And have fun!
Coach and OC both left already, glad to see the wide receivers coach (I think??) having fun with it
December 19, 2024 at 4:02 AM
Coach and OC both left already, glad to see the wide receivers coach (I think??) having fun with it
It's been nice to not have to "Select Top 10 * " from each table and paste that into a chat. More to come!
December 11, 2024 at 5:49 AM
It's been nice to not have to "Select Top 10 * " from each table and paste that into a chat. More to come!
As part of that, I've been working on new tools for context management - including a SQL helper that analyzes tables and lets you pick which ones to send to the LLM
December 11, 2024 at 5:49 AM
As part of that, I've been working on new tools for context management - including a SQL helper that analyzes tables and lets you pick which ones to send to the LLM