This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
We've recently made an accessibility improvement to the community and therefore posts without any content are no longer allowed. Please use the spoiler feature or add a short message in the message body in order to submit your weekly challenge.
There is evidence supporting the fact the fact that the Superbowl is less predictable. I used a linear regression model, but I wouldn't mind knowing which models I should use under which circumstances. Solution attached,
Standard predictive analysis here. While I'd love to be snarky and claim that I only used offensive stats because the Pats have no defense, the real reason is in the p-values. They were statistically significant.
I explored whether removing some "outlier" games had any impact - and they resulted in worse p-values, so I went with all of the data excepting the superbowl games for modelling.
Oh, and FLY EAGLES FLY. Though I am actually a Giants fan ;)
I used the Association analysis tool to select 3 predictor variables. Based in the difference between : - The predicted scored points VS actual points in regular season matches - The predicted scored points VS actual points in Super Bowl matches
I found that the predictions were pretty accurate for regular matches but not for Super Bowl ones, I guess it is because of the whole motivation and adrenaline levels behind those special games !