As a part of a paper I am working on, I found this interesting: Looking at bowl games played in the 2002-2007 season (n=96)...
- The team with the higher AP ranking going into the bowl game won 59.5% of the time.
- The team with the higher USA Today ranking going into the bowl game won 62.5% of the time.
- The correlation coefficient (Excel function CORREL) between these two polls giving the higher ranking to the team that ultimately won the bowl game is 0.542
The observation that the team with a higher ranking in the poll going into the game will win most of the time implies rationality in the polling. However, I expected the correlation between them to be much closer to one (i.e. they would error on the same games), and I am surprised at how close it is to .5. The inference of being less than one (but still positive) is that they agree most of the time on the ordinal ranking but that upsets occur for different games.
Data here.
Should I have expected this magnitude of disagreement? What would you have expected?
14 comments:
Maybe look at only observations where there is a clear favoring (i.e. not just ranked higher, but ranked 5 spots higher at a minimum or something like that). I am guessing that some of those bowls are between 10 thru 20 ranked teams and often at that level the ordering is not the same for the 2 polls (I suspect). This may apply to even lower ranked ones too, I would imagine. How low do your rankings go? Can you try further back? How about with the BCS and Harris polls as well?
All very good suggestions, and my eyeballs suggests you are right.
As per your question...in all the bowl games at least 1 team was ranked by both polls in the top 25. So teams without a ranking I cannot judge if they were seen as #26 or #50 by the pollsters.
I'm still working on going further back and getting more polls. What I have now is just what is relatively easy to obtain from the internet. Would love it if anyone has other sources of data to suggest!
Try this for AP.
http://www.appollarchive.com/football/ap/seasons.cfm?seasonid=2008
It is unclear why you use the coefficient of correlation here, which is designed to find the correlation between two numerical variables.
Why not, much more simply, look at the % of the time that they match or do not match? They do not match slightly less than 22% of the time (21/96 do not match). Is this significantly high? It would imply that if you knew one, that you would know the other more than 78% of the time (it is basically equivalent to an r-squared of .78 in an ols regression.)
Perhaps if you were looking at the difference in ranks, it would be appropriate to use the coefficient of correlation.
If this is heading down the probit path...there are plenty of things to control for, as well as one that I think is why you're looking at this in the first place. ;)
I used to be able to find good information from your articles.
Take a look at my web-site :: teens tight pucker is fucked and her throat penetrated
I like the helpful information you provide
in your articles. I'll bookmark your weblog and check again here frequently. I'm
quite sure I will learn many new stuff right here!
Best of luck for the next!
my web page visit
Hi there, I found your website by means of Google
whilst searching for a similar topic, your web site got here up, it seems great.
I've bookmarked it in my google bookmarks.
Hi there, just was aware of your weblog via Google, and located that it is really informative. I am gonna be careful for brussels. I'll appreciate in the event you proceed this in future.
Many other people might be benefited out of your writing.
Cheers!
Here is my page: nubiles net brie Turner cheerleader masturbates with a magic wand after
Hi there, I discovered your site by means of Google even as looking for a similar matter, your website came up, it looks great.
I've bookmarked it in my google bookmarks.
Hello there, simply became aware of your blog through Google, and located that it's really informative.
I'm going to watch out for brussels. I will appreciate in case you continue this in future. Many other people can be benefited out of your writing. Cheers!
Also visit my web site :: click here
Hmm it appears like your website ate my first comment (it was extremely long) so I guess I'll just sum it up what I wrote and say, I'm thoroughly enjoying your
blog. I as well am an aspiring blog blogger but I'm still new to the whole thing. Do you have any helpful hints for novice blog writers? I'd genuinely
appreciate it.
My website - http://pornharvest.com/index.php?q=nubiles+camille&f=a&p=s
Hey there would you mind letting me know which webhost you're using? I've loaded
your blog in 3 completely different web browsers and I must say this
blog loads a lot faster then most. Can you suggest a good web hosting
provider at a fair price? Thank you, I appreciate it!
Look at my page: http://facebooks.lv/
I am really impressed with your writing skills as
well as with the layout on your weblog. Is this a paid theme or did you modify it
yourself? Either way keep up the excellent quality writing,
it's rare to see a nice blog like this one these days.
Review my web page Additional Resources
We're a group of volunteers and opening a new scheme in our community. Your site provided us with valuable info to work on. You have done a formidable job and our whole community will be grateful to you.
Feel free to visit my blog post ... do you agree
electronic cigarette starter kit, best electronic cigarette, ecig, smokeless cigarette, e cig, electronic cigarettes
Post a Comment