The first team ratings are released for Men’s Basketball. I wanted to give you a brief breakdown of what these ratings are, and how you should use them. Here are the ratings in their entirety (Click Here), and check out the explanations below:
How to use the ratings:
These ratings are first and foremost predictive. They use the data available, and make predictions on what the end of the season will look like. This is unlike other ratings, because most ratings tell you what the ranking are to date… this is not exactly helpful in my opinion. I want to know… given the current data, what will these teams look like at the end of the year if their current performance plays out throughout the season.
Winning percentage and Pythagorean Expectation:
These ratings are based on Pythagorean expectation. This is a theory created by Bill James that uses points scored and points against to create a winning percentage. I prefer this over a basic win/loss winning percentage because “pythag” has been proven to be more accurate predicting future games.
Offense Adjusted / Defense Adjusted
When you watch college basketball (and any sport for that matter), you know that the most important element to any sport is how efficient a team is. Efficiency is the single most important factor for long-term success in basketball. Obviously in a one game scenario… anything can happen, but in a long-term prediction, efficiency is incredibly important.
For these ratings, efficiency is determined by how many points a team would score and/or allow if given 100 possessions. The only other factor is strength of schedule. This is simply determined by an average of the opponents (and the opponents, opponents) Pythagorean expectation.
Luck is actually an interesting statistic. This is basically the comparison of a teams winning percentage and their Pythagorean winning percentage. The idea is that the pythag is more accurate, and therefore you can see if a team is over or underperforming based on their point differential.
These numbers will change drastically because there is such little data these ratings are based on, but I do believe that it is still interesting to see how things are working out so far.