RCL data

Place to discuss all things to do with the RCLs. League updates, league trash talk and RCL trade talk.
OaktownSteve
Posts: 317
Joined: Mon Mar 28, 2011 9:55 am

RCL data

Vin do you have RCL data you are willing to share? I would love to work with a spreadsheet that basically has each teams final standings position in their league and the category point total for each category (not necessarily the category value ie number of home runs, just the category rank). If you have anything easy I can use, I will be happy to share my findings with you.

VinWins
Posts: 3199
Joined: Sat Feb 12, 2011 3:33 am

Re: RCL data

Definitely willing to share anything I've got. Would you want each league on a separate sheet, or all 456 teams on 1 page?

OaktownSteve
Posts: 317
Joined: Mon Mar 28, 2011 9:55 am

Re: RCL data

all on one page. I'm just going to do some simple regressions. It's awesome to have roto meta data to play with.

VinWins
Posts: 3199
Joined: Sat Feb 12, 2011 3:33 am

Re: RCL data

I think this is what you are looking for.

I had it mixed in with a bunch of other stuff which I cleaned out. Had been thinking of doing something with these numbers, but hadn't figured out what.

OaktownSteve
Posts: 317
Joined: Mon Mar 28, 2011 9:55 am

Re: RCL data

perfect. Really appreciate it. Will definitely share results with Razzball community.

VinWins
Posts: 3199
Joined: Sat Feb 12, 2011 3:33 am

Re: RCL data

Sheet 3 now has both the score for the category and the stats, if that's of any interest.

OaktownSteve
Posts: 317
Joined: Mon Mar 28, 2011 9:55 am

Re: RCL data

Vin thanks again. All I did was calculate R-Squared for final standing versus the category rank and the category totals. Basically what that does is say how closely does x correlate to y or x predict y if you prefer. In this case x is the final standing and y is the category. For R-Squared, 1.0 means it perfectly predicts. The higher the score the closer the correlation.

One thing that was not surprising is that the rate categories had relatively low correlation (especially average). I expected HRs would be the best predictor, but that turned out to be wrong. It's actually R, RBI, K, W. I have some theories about this but I'll think about it some more.

Here are all the R Square values, first against category rank and second against category total

R .61,.539
HR .318,.293
RBI .48,.451
SB .248,.22
AVG .11,.111
K .59,.52
W .482,.45
S .31,.309
ERA .24, .256
WHIP .301,.2896

VinWins
Posts: 3199
Joined: Sat Feb 12, 2011 3:33 am

Re: RCL data

You're welcome. And, thanks to you, too.

I guess this validates Grey's hate of average and love of strikeouts. No one who lead their league in Ks finished worse than 4th.