That sounds plausible.
Hi! Stats newbie here so be nice!
I'm currently working on a projection model for fantasy basketball. I am trying to adjust a player's expected points scored in a game based upon the strength of his opponent. Here's a specific hypothetical question. Say I have Kevin Durant...
I know Kevin Durant typically scores 50 fantasy points per game. Say I also know his standard deviation as well. I also know the z score for Kevin Durant's average fantasy points per game relative to other players as well as all relevant factors about the normal distribution of all NBA player's fantasy points scored per game.
Say Kevin is playing Indiana tonight. Say I know Indiana allows 10% fewer fantasy points per game than the average NBA team and that this equates to a Z-score of -1.26
How do I adjust the normal distribution of Kevin Durant's expected points scored? Does the new mean value of his fantasy points scored become the point value of a -1.26 z score on his average points scored distribution? Or is there some other way of calculating this that I don't understand? Thanks for the help
That sounds plausible.
All things are known because we want to believe in them.
I thought about it a little more and it doesn't quite seem to make sense when I apply it. If Durant were to average -1.26 z score points on his probability distribution that would mean he if we shifted his curve to that mean value he would be averaging about 36 points a game now vs that team, which seems really low to me. Also it would mean there is about a 16% chance he scores under 24 points which also seems kind of unfathomable to me.
I'm so confused
Tweet |