A good question.

Posted in a league forum, but I thought it more likely to get noticed here.

peterg wrote:

Has anyone ever done a big bad correllation of ppg against winning percentage, using lots of KFBA data?

I never felt the need...  There should be a direct correlation between Kentucky Fried Points per Game and Wins...  It should be a strong correlation, but not perfect...  I would bet my left nut on it.

colonel wrote:

I never felt the need...  There should be a direct correlation between Kentucky Fried Points per Game and Wins...  It should be a strong correlation, but not perfect...  I would bet my left nut on it.

That still looks like a risky bet.   

Ryan looked this up for us and put together some pretty graphs.  Unfortunately we can't figure out how to copy paste them on here...

broham wrote:

Ryan looked this up for us and put together some pretty graphs.  Unfortunately we can't figure out how to copy paste them on here...

Can he just do a print screen and paste the entire screenshot?

 

Thats the screenshot... hehe.

Attached is the 4 Mb sheet...

How do I get the attachment? I don't see a link.

Been trying keep getting an error on that as well.  This sucks.

.

Nice graph.  It seems to support the fact that to be a winning team you need to have scores averaging in the high 190's.  200+ averages are a pretty good team.  As you get over 210, you are a great team.

For those who don't have excel.

Yeah, that's a nice graph. Good job, Ryan and Broham! There's a pretty good correlation, apparently. The Colonel is spared his left nut.

There's something I've been wondering about winning percentages. I know there's probably a correlation between winning percentage and number of trades proposed, but I'm not sure if it's a strong correlation... For example, JCPeery, zzRay, and myself propose lots of trades. We end up winning more than we lose (~60%) and our teams average pretty high weekly scores (around/above 200 ppg). On the other hand, the Colonel, Broham, and Doug are more passive when it comes to trading... but somehow they have higher winning percentages than us and similar weekly averages. Is this because they've been playing longer/have more knowledge of this game or is it because trade volume explains very little of the variation in winning % / weekly avg?

Other random, pointless questions:

Is there a correlation between number of waiver wire transactions and winning % / weekly avg? If so, is this only because these managers also happen to be managers who propose lots of trades?

Can we break the avg. PPG data into percentiles? Like, if my team averages 210 ppg for the season, what percentile would that place my team in (if we include all teams over all KFBA seasons)? Some owners would like to know where their teams rank.

I think we have trade data cuz I remember a thread that we had about % of trades accepted for each owner, the one that showed that the Colonel had the highest acceptance %. I also remember a thread that listed the top waiver wire users, so I know we have waiver wire data.

jcpeery wrote:

Nice graph.  It seems to support the fact that to be a winning team you need to have scores averaging in the high 190's.  200+ averages are a pretty good team.  As you get over 210, you are a great team.

Actually it looks like if you get over 210, you are either a perfect team or a terrible winless team. Check out all those zero win percentage dots on the graph.

Anyway, many thanks for putting this together that was very kind/useful. Just wondering also if you could flip the chart axes, I think it would be more informative that way, but I guess that depends on whether you're a maths person or an economics person.