ST Alpha - "Since all stats are treated equally, my regressions favor the well-rounded player over the specialist." (Reading a Post)

SlickerTalk
Search Archives
The Leaderboard
The FAQ

Down to Post

Backboards:

Posts: 151

In response to "I had extra points this year so was just looking for good candidates to get to 28 (I'll have 13 total) and I was like "that can't be right"" by spamlet

Since all stats are treated equally, my regressions favor the well-rounded player over the specialist.

Posted by Swilcan Baron (aka BaronWasteland) Jun 15 '22, 11:38

I've started using a unequally weighted linear combination of the regression scores predicted to determine a player's value for the overall rankings.

For example, a PG with E level hands and F rebounding will be preferred over an otherwise identical one with F level hands and E rebounding.

I've liked the results of the weighting so far. The defensively minded centers have definitely benefited from the change.

I've also downgraded the predicted Int scores in all my rankings since that portion of the predictive model is kind of sh*t. :D

Responses:

I'm taking 8 to signing day and I thought that was a lot -- nm - rollo Jun 15, 11:54
My rankings basically do what you're doing but not quite that granular and then do filters from there. - spamlet Jun 15, 11:43

Post a message top

Since all stats are treated equally, my regressions favor the well-rounded player over the specialist.

Replies are disabled on threads older than 7 days.