How about a method combining randomness with elo ranking for simulating scorelines? One example: Let's say I have a teams A (Home) and B (Away) with Elo scores EloA and EloB. Then I am taking n random matches from a sample, and I calculate the average Elo rankings and goals scored, which I then adjust to the EloA and EloB, to predict the A Vs. B scoreline. It would be then ScoreA = EloA * AvgHomeScore / AvgHomeElo and ScoreB = EloB * AvgAwayScore / AvgAwayElo. Just an example, of how we could introduce randomness in an Elo-based model.
Thank you very much, i know that there is not millions of views, but for the ones that do see this videos help us a lot, sorry for my english, im from Mexico
away_score_rate=poisson_model.predict(pd.DataFrame(data={'team': away_team, 'opponent': home_team, 'home':1},index=[1])) home in this case should be 0, is it ?
Why does the lambda HAVE to be the average goals scored per game? I can think of a much better metric to measure a team's scoring ability. IMO the average is fundamentally flawed if only considering a single team's data from a single season.
You all probably dont give a shit but does anyone know a way to log back into an Instagram account..? I stupidly lost the login password. I love any assistance you can give me!
@Landyn Mauricio I really appreciate your reply. I found the site thru google and I'm waiting for the hacking stuff now. Seems to take a while so I will reply here later when my account password hopefully is recovered.
