Suggestions?

by ZBW on Apr 1, 2009 10:10 AM EDT

So I have a project for a statistics class that is completely open ended. The only requirement is that it involve multiple linear regression. I am definitely going to do something relating to baseball and/or the Rays, but I wanted to get some suggestions from the DRaysBay crowd because I plan on posting my analysis after I'm done.

I know there is a lot of statistical analysis around here, but what is something that you would be interested in seeing that hasn't been done yet (or, at least, hasn't been done for the Rays)? Suggest away...

This post was written by a member of the DRaysBay community and does not necessarily express the views or opinions of DRaysBay staff.

Tweet Comment 11 comments | 0 recs |

Do you like this story?

Comments

My project was to see if there was a correlation between ERA and/or WHIP to salaries

So I would say don’t do that.

Check out my blog on web development at kericr.wordpress.com

by kericr on Apr 1, 2009 10:23 AM EDT reply actions

Oh and also because it only netted me a B+

Check out my blog on web development at kericr.wordpress.com

by kericr on Apr 1, 2009 10:23 AM EDT up reply actions

If you can tolerate PitchFX

I’d love to see someone do something showing what exactly makes a fastball an out pitch (i.e. what type of velocity and movement on the x and y axis lead to more swinging and looking strikes).

Vogt early, Vogt often.

by Brickhaus on Apr 1, 2009 10:57 AM EDT reply actions

This is very interesting...

I’m gonna look into this. Might be a hell of a lot of data to deal with though…

by ZBW on Apr 2, 2009 10:16 AM EDT up reply actions

I did my Research Paper on a Multi-lin regression using Team wins as the Dependent variable

and Team ERA, DEF, SB, and OPS as my independent variables. You really cannot go wrong doing baseball since there is so much data. Research other papers that have been written and try to find areas where they did not go far enough or where you can take it in another direction. It is double good because you are going to need some references and you can use their methodology for the most part and expand on what they did. Just remember to give credit where it is due.

Do what you love to do and give it your very best. Whether it's business or baseball, or the theater, or any field. If you don't love what you're doing and you can't give it your best, get out of it. Life is too short. You'll be an old man before you know it.

-Al Lopez

by Sandy Kazmir on Apr 1, 2009 11:13 AM EDT reply actions

Am I the only one who hasn't taken a statistics class?

I could be wrong though

by staplemaniac on Apr 2, 2009 7:48 AM EDT up reply actions

I took a bs one in high school.

Swav or Die.

by SRQman on Apr 2, 2009 11:50 AM EDT up reply actions

Probably not

But there are a lot of nerds here. I majored in psychology and almost went into experimental psychology, so I had 4 or 5 statistics classes in college. Not that I remember how to perform a MANOVA.

Vogt early, Vogt often.

by Brickhaus on Apr 2, 2009 2:09 PM EDT up reply actions

Thanks for the suggestion man

I did something like that for my undergrad Junior Paper. It was pretty interesting and turned out that OPS was the best predictor out of the variables that I chose for hitters.

by ZBW on Apr 2, 2009 10:15 AM EDT up reply actions