Challenge #18: Predicting Baseball Wins

[CaraI]

Fun predictive exercise!  Straight in line with what most people did.

Spoiler

Spoiler - 

[justynam]

My solution

[Jim7]

My projected wins were different than the solution since it made the most sense to me to pick the top 10 predictors based on the initial multiple regression, which takes into account unique variance. The ideal would have been to use a stepwise regression following the multiple regression, but sample size was too small.

[joshuaburkhow]

I used a forest model on this one ;)

[Kenda]

Spoiler

This one was a little tough for me. I had never used either of these predictive tools before. In fact, I didn't even have the category visible in my tool pallet until I started this challenge.

[StevenBetts]

Here's my solution :)

Spoiler

[dillon_bok]

Spoiler

[OldDogNewTricks]

Here is my solution.  Although I the linear regression doesn't appear to be the best model to use and limiting to the top 10 variables based on correlation alone is bad practices (if this was a REAL problem).

[aking]

First time using the predictive tools

Spoiler

[Mikkel_D_Karlsen]

Solution with Method 2