To provide examples for the undergrad stats class I teach (and for my own interest), I made up a spreadsheet in SPSS detailing each Gonzaga game. There is a lot of neat stuff there, but I thought some of the most interesting data relates to turnovers. Statistics have their shortcomings, but the numbers really point to the importance of turnovers and turnover margin. For example:
There is a strong correlation between Gonzaga turnover numbers and opponent point total (but not GU point total) as well as GU turnover numbers and margin of victory
Even after controlling for opponent strength (via RPI), Gonzaga’s turnover and rebounding numbers both explain a significant proportion of the variance in margin of victory or defeat.
In another way of looking at things, after controlling for RPI (which explains about 40% of the variance in margin of victory/defeat, turnover differential explains an additional 44% of the variance.
In looking at Gonzaga’s wins versus losses, looking at all of the major team statistics - turnovers are the only statistically significant individual team variable. GU averages just over 12 turnovers in wins and 16.5 turnovers in losses. Turnover margin is also statistically significant (+1.29 in wins, -5.50 in losses – as is rebounding margin (+7.48 in wins, -.5 in losses).
--
If anybody is as big of a nerd and would like the SPSS database, please PM me.
There is a strong correlation between Gonzaga turnover numbers and opponent point total (but not GU point total) as well as GU turnover numbers and margin of victory
Even after controlling for opponent strength (via RPI), Gonzaga’s turnover and rebounding numbers both explain a significant proportion of the variance in margin of victory or defeat.
In another way of looking at things, after controlling for RPI (which explains about 40% of the variance in margin of victory/defeat, turnover differential explains an additional 44% of the variance.
In looking at Gonzaga’s wins versus losses, looking at all of the major team statistics - turnovers are the only statistically significant individual team variable. GU averages just over 12 turnovers in wins and 16.5 turnovers in losses. Turnover margin is also statistically significant (+1.29 in wins, -5.50 in losses – as is rebounding margin (+7.48 in wins, -.5 in losses).
--
If anybody is as big of a nerd and would like the SPSS database, please PM me.
Comment