Total points over the season are definitely not normally distributed

They're positively skewed with many teams getting the bulk of points below the mean and a few getting a lot more points (above the mean) with the median being significantly lower than the mean (because of title winners/European qualifiers etc skewing the distribution)

I think the model SCC referred to only seems to do the aggregating well (based on what ifs, so what if we won all 7 games or lose them all or all the various different scenarios in between) I can't see where the parameterisation and estimated for the likelihood of each of these scenarios is detailed anywhere.

The model input distributions are key for me. Otherwise like you said, the (aggregated) output is pretty meaningless

Oh well. Never mind.

Posted By: essexcanaryOTBC on March 23rd 2016 at 15:40:33

Message Thread

FAO: USAC (General Chat) - essexcanaryOTBC, Mar 23, 13:26:26
- reply (General Chat) - usacanary, Mar 23, 15:01:31
  - It should definitely be a credibility weighted prediction (General Chat) - essexcanaryOTBC, Mar 23, 16:12:56
  - Total points over the season are definitely not normally distributed (General Chat) - essexcanaryOTBC, Mar 23, 15:40:33
    - GIGO (General Chat) - essexcanaryOTBC, Mar 23, 15:41:59
    - *estimates (n/m) (General Chat) - essexcanaryOTBC, Mar 23, 15:41:21

Reply to Message

In order to add a post to the WotB Message Board you must be a registered WotB user.

If you are not yet registered then please visit the registration page. You should ensure that their browser is setup to accept cookies.