I am exploring the relationship between time of possession and scoring in rugby. Each match has two teams that appear on subsequent rows in the data set:

       Match Event  Team             Points Possession
0      1     Dubai  SOUTH AFRICA      36    209
1      1     Dubai      PORTUGAL       0    113
2      2     Dubai         WALES      14    231
3      2     Dubai        CANADA      19    197
4      3     Dubai          FIJI      54    218
5      3     Dubai        FRANCE       7    141
6      4     Dubai     ARGENTINA      17    238
7      4     Dubai        BRAZIL       5    146

I subtract the second team's points and possession from the first team's to find the point and possession differential for each match:


I plot the possession differential against the point differential in a scatterplot to visualize the relationship, add a best fit line, then use pearsonc to characterize the relationship with a single number.

What are the drawbacks of this approach?

Is there a better way to characterize the relationship?

Would it be better to normalize the data to the match winners so that all point values are positive?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.