I'm trying to predict tennis match outcomes in R using only match scores from previous opponents to predict winners/losers for future matches. The match scores, however, are only as granular as the set score and game score (there is no knowledge of point scores or other statistics). The resultid is largely irrelevant, I believe.

If this sounds confusing, please have a look at my

training: https://docs.google.com/spreadsheets/d/1s28NQ-vUlSeNAZIjCsOnoH5LGSp1z3WkaIw4KENB-24/edit?usp=sharing

and test: https://docs.google.com/spreadsheets/d/1rTKje2W0BnkhOWniDWAsyBcKYhdI5q_2JHDt9BOneA0/edit?usp=sharing

data sets.

I've hypothesized that I might need to use Markov chains to do this with different game scores (1-1, 1-2, 1-3, etc) as different states, but I'm not sure if this is possible, due to how limited the data is.

Please advise. Any guidance would be greatly appreciated.

  • $\begingroup$ I think it is easier to reason about your problem if you write down a mathematical model describing the phenomenon you model. Then it is easier to figure out if Markov chains are applicable to the specific model. $\endgroup$ – hakanc Dec 3 '20 at 10:47

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.