Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 0 down vote favorite

I'm looking to generate as much interesting data as I can from some 2on2 basketball games.

The only data I have recorded is the players on each team, and the final score.

What interesting things can I learn from this data? I'm interested in what I can learn about the individual players, as well as the teams and player combinations.

Can anyone provide a good starting point.

share|improve this question
This will depend heavily on the amount of data you have available. To get anything significant, you're going to need lots of games, and preferably with just about any combination of players in the two teams. Do you have that? Can you give us more details on which data you have? – Nick Sabbe Sep 8 '11 at 6:57
3  
@Bill - Suppose you did some brainstorming to come up with interesting questions to ask of your data. Then you could use this group as a resource to help you answer those that seem technically challenging. – rolando2 Sep 8 '11 at 14:13
@NickSabbe We are keeping track now, and will eventually have hundreds/thousndands of games. – Bill Zenrick Sep 9 '11 at 2:32
@rolando2 This is kind of what I was asking. Is this enough data to do something interesting with? Can you think of other data I could collect that could be used to find out interesting patterns or statistics about the individual players? – Bill Zenrick Sep 9 '11 at 2:33
3  
@Bill - Let's put it this way: to ask for help with a technical question that's got you stuck is a fairer use of people's time than asking them to start from scratch and think up your research questions for you. The latter is really not what this site is designed for, or what many people are going to be willing to give you. Sorry if that sounds cold. – rolando2 Sep 9 '11 at 18:53

3 Answers

up vote 1 down vote

You could try to compute an Elo rating for each player, and for each pair of players playing as a team, and see whether the ratings for the teams agrees with the average of the ratings for the players.

Typically, in team games, the rating for the team is assumed to be the average of the ratings of the players for the purpose of handicapping matches to see how much you should adjust each player's rating on a win or loss. You might have enough data to test whether the assumption is generally accurate for your teams.

This would only use the wins or losses, not the scores.

share|improve this answer
up vote 0 down vote

Check out this blog. It may give you some ideas what you can do with basketball data. You probably want to record more than just players and the final score. Keep track of individual player's stats as well. If you are a basketball geek like me, check this website too.

share|improve this answer
up vote 0 down vote

I wrote an advanced stats primer for the NBA. It has numerous metrics that you might apply, although you will have to adapt them to 2-on-2 games.

http://thecity2.com/advanced-stats-primer/

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.