tallguyindc's Blog

Interesting meaningless trends....

By tallguyindc | View all Posts
Posted Tuesday, October 19, 2010 07:00 PM   2 comments

OK.  I was updating my playoff odds spreadsheet and I noticed an interesting coincidence.  The spreadsheet has the teams listed in alphabetical order  and none of the top 12 teams by city won this weekend (4 byes; 8 losses).

Obviously, this is purely random and means absolutely nothing.  Still I thought I'd mention it anyway.  Its a very good example that trends that appear meaningful can often be created by random luck and that 0 for 12 on things that should be coin flips doesn't necessarily mean anything other than overextrapolating a small sample size.

 

 

 

2 comments
comment Post A Comment
tallguyindc says:
10/20/10 01:06AM

Disagree with what....that randomness will provide some things that look significant but aren't...

Do you think the alphabetical thing was significant?

Its easy to say its not when its so obviously spurious but quite a few things appear more significant than they are.

Since there were 14 wins to go around this week, the odds of none of the top 12 alphabetical teams getting them was .001.   Looks really significant doesn't it???

 

The bottom line is random chance does provide random results.

 

 

MJ2345 says:
10/20/10 01:18AM
This is why i only bet trends where you can explain why its a trend and advantage.

For example the favorite off a bye week is 72% ATS since 2005 this year 2-1 so far and a road favorite off  bye has won the last 10 in a row ATS.This is the 3rd year I have bet this trend but this trend makes sense.First most of the time a team favored is going to be a good football team obviously there are exceptions and a team favored on the road is normally a very good football team.Well give a good to very good football an extra week to heel and prepare well its an advantage.Just my opinion.
Add your response:
Please note that the blogger has the option of reviewing all comments before they are displayed to the public. Your comment may not be published immediately.
 
Advertisement