LikeLike

]]>LikeLike

]]>LikeLike

]]>LikeLike

]]>LikeLike

]]>Boston gets first pick: Phoenix has a 26.53% chance of getting the second pick (199 / 750)

Knicks get first pick: Phoenix has a 21.01% chance of getting the second pick (199 / 947)

LikeLike

]]>The example you give with Boston is captured using the “p” variable in the Python code I quoted in the post. ALlow me to walk through it to show how that part works!

Suppose the variable “team” is NYK. Then we walk through all other 13 teams as “team2.” The first iteration is Boston. Then, it’s exactly as you say:

At 25% chance, Boston gets that pick. That’s “balls[team2] / 1000”. OK… so what’s NYK’s chances of getting the second pick?

balls[team] / (1000 – balls[team2])

So in this case, NYK has a 53 / 750 chance of securing the second pick **given Boston has the first pick** That latter half is important! In this case, and in this case alone… the Knicks improved their odds from getting selected in the top 3 from 5.3% to 7.07%. Similarly, with Boston having the first pick, the Suns improve from 19.9% to 26.53%.

Thanks again for the comment!

LikeLike

]]>Another thought, angle, is the dependence upon who gets #1. For instance, if BOS (thru BRK) gets the #1 pick, then their 250 balls drop out, thus each team’s likelihood would jump dramatically of getting the #2 pick by comparison to say, if the NYK jumped up and got the #1 pick. Is that correct?

e.g. the use case: If BOS gets #1 pick, Phx is more likely to get #2 than they would if NYK got #1 pick.

LikeLike

]]>LikeLike

]]>LikeLike

]]>