Study Topics?

[sbrylski]

I have some time over the next few days, and I'd like to do a mini-study if I can come up with a decent topic. Any thoughts?

 

(Another idea might be to do a group study, if anyone was interested in helping out. Two part type thing. Smaller study I'd want to do solo - don't want to cramp my style...)

[logan82]

Managing styles of managers. Specifically bulpen/closer usage. Gaining split advantage with pinch hitters. Sorry to be so brief. Typing on Wii is difficult.

[TooLiveBrew]

seconded. Hopefully from more than just one season, but I know that's a bunch of work.

[sbrylski]

I could try something like that...

[logan82]

The main problems as I see it would be determining who the closer is and when for each team. It would be easy for a team like the Brewers who had Cordero as the closer all year. Much more difficult for teams with no set closer or teams that changed closer midyear. Also who was the manager throughout the year. Makes a team like the Astros a little tricky.

[logan82]

How about looking at every save for every team and splitting them into 4 categories

1) 3+ inning saves

2) <3innings but >3 out saves

3) 3 out saves

4) <3 out saves

 

Edit: Keep track of which reliever got each type of save so you can look at how many players each team relied on to close games. That way you can also go back and rank the tops 7-8 relievers on each team and see if the top reliever was the guy with the most saves or highest leverage innings. Just a suggestion of course. I might get around to doing it myself if nobody else does.

[rluzinski]

Order baseballreference.com's Play Index tool is free until January 11th. I bet looking at all the available splits will inspire you:

 

http://www.baseball-reference.com/pi/

[TerraceReserved]

There's been a study I thought of a bit ago that I've been wanting to do, and instead of making a new thread I thought I'd post it here.

 

Nate Silver did a study on clutch hitting, where he took the clutch scores of players on odd and even years, and if clutch existed the graph he made should be linear, and good clutch hitters will be good in both odd and even years. He found a small relationship, but it got me thinking.

 

What if the type of year (odd, even) that they had a worse clutch score, they also performed worse overall? For example, if I averaged the clutch scores for every player and in the type of year that they had a better clutch score they averaged 50 EQR. And in the type of year that they had a worse clutch score, they only averaged 40EQR. That would mean that when you are hitting better, you also hit better in the clutch. That backs up common sense. I could then adjust the years so that the players accumulated 45 EQR in both years (odd and even), changing there clutch scores slightly, and I beleive that would strengthen the case of clutch hitting. Hopefully I would find that a stronger "R" relationship when I graph the odd and even years after the adjustment.

 

Of course, I may find absolutely nothing, and my theory might be proven wrong, but I'll work on it and post it on here. Any thoughts?

 

Edit: clutch scores would be the ones from fangraphs.

[logan82]

I did the NL Central last night and think that think that I am going to have to include blown saves and somehow incorporate more stuff(not sure what yet) to get anything useful.

[logan82]

Saves and how they were acquired last year. If there are any mistakes let me know.

http://spreadsheets.google.com/ccc?key=pKN8HdBGjC9LMx1xERoBXBQ&hl=enhttp://spreadsheets.google.com/pub?key=pKN8HdBGjC9LbPpfvDTv3eA

[Fatter than Joey]

If you are still looking for ideas..

 

Here is a study that looked at managers in one-run games.... It only goes to 2004, perhaps stretching it out to 2007 would be interesting.

 

Link