Sunday, September 21, 2014


Written By: LM

This post is a little late but I thought it would be a pretty cool to capture the Texans predictions from the local sports media.

It's all guess work and fun.  However, I did want to analyze some basic statistical elements of the data.

Basic Statistics
  • Average/Mean - Add your group of data.  Divide the sum by the list of numbers
  • Mode -  Value that occurs the most often
  • Media - The value that occurs in the middle when data is organized lowest to highest (useless)
  • Range - Add the lowest and highest value, divide by 2
  • Variance - How far your set of numbers are spread apart
  • Standard Deviation - Measures the variation from the average.  A low Standard Deviation means the data is very close to the average. a high Standard Deviation means the data is far apart which likely means there is an outlier variable which is skewing the data. In addition to, the Empirical Rule (68, 95, 99.7) is used in conjunction with Standard Deviation.  Empirical Rule states nearly all values will lie within 3 Standard Deviations of the average in a normal distribution (bell curve).  In other words, 68% of our predictions will lie within 1 SD, 95% will lie within in 2 SDs and 99.7% will lie within 3 SDs.

It appears, on average the local market believes the Texans will win 7 games with a standard deviation of nearly 11 games on the high and 5 games on the low.

It's only a little fun with numbers but very interesting to see predictions so close.


No comments: