tywalters / nfl-prediction-model Goto Github PK

View Code? Open in Web Editor NEW

29.0 29.0 16.0 2.6 MB

Model created for FiveThirtyEight Forecast Competition

Python 100.00%

fivethirtyeight-dataset nfl pandas prediction-model python scikit-learn sports-betting sports-data

nfl-prediction-model's People

Contributors

Stargazers

Watchers

Forkers

num951 jprice4444 azieky krishanu-datta hoke98 sol-won-1100 jesse989 timbuckwho bkrippner ssavala ccdickerson clintondurney joeyminichielo demothedread cob05 borneonose

nfl-prediction-model's Issues

Usage

Hey Ty,

Thanks for all your work on this project! I apologize if this is a dumb question since I'm fairly new to using python. I have everything set up and running fine. However, how do I go about running projections for each upcoming week? Am I supposed to be adding the new data every week to the CSV files? Then which data frame would contain the projections for the upcoming week?

-Dan

Winner of each game every week

add a week by week probability of a team's chances to win in a graph format using the historical data. Unless I missed something in the code.

Adding more features

To add more features would one create multiple dataframes after the pts diff dataframe?

line 365

to get the probability of houston beating kansas city week 1 (lets say), should line 365 the predicted line by printed and not the df_results line? I am just stumped on why I cannot get the probability that houston will beat kansas city to show up

Recommend Projects