web analytics

Tightened Rotations

April 16, 2011

After a ton of math and fudging, I’ve come up with these estimates for how much a “tightened rotation” will help each team in the playoffs. I took close games between playoff teams in the past 1.5 months, figured out the top rotation (MPG/player) for each team when everyone’s healthy, and took the max of that and the average of all playoff teams’ average top rotation minutes split. Then, I compared that tight rotation with the average MPG for each player over the season. I truncated the list when each rotation reached 240 minutes total. Then, to see how much that helps each team, I used ASPM and RAPM for each player with that MPG and calculated the team sum both for the short rotation and for the idealized long rotation.

Yeah, it’s a bit clunky, but it sorta makes sense.

I still can’t get the Lakers to look dominant, though…

I did it for every team above -1.25 Efficiency differential. I forgot that the Pacers were below that (!) so I guess I’ll use the average for them!

Team Tightened Benefit
HOU 2.20
MEM 1.80
PHO 1.66
NOH 1.61
SAS 1.52
PHI 1.18
ORL 1.14
DAL 1.09
LAL 0.92
DEN 0.88
MIA 0.80
BOS 0.64
ATL 0.63
OKC 0.59
NYK 0.29
POR 0.22
CHI 0.16
Average 1.02

Yeah, it doesn’t help the really deep teams to tighten their rotations. Some of those numbers look a bit odd. I don’t have time at this point to check them, though!

Tags: , , ,

3 Responses to Tightened Rotations

  1. Nathan Walker on April 16, 2011 at 8:54 am

    any chance you could share the minutes% for each player? I’d like to use it for the RAPM ratings :)

    • DanielM on April 18, 2011 at 9:50 am

      I don’t think they’re valid for each player. I do have a spreadsheet that goes through and scrapes all the box scores off of BBref that you specify–that’s how I came up with this. Scrape the close games between playoff teams and try to rough in minutes distributions based on those close games compared to the season as a whole.

      It’s really clunky, though, and I’m not sure it’s totally valid for individual players. More of a ballpark figure.

      BTW: I do not like trying to predict minutes distributions. It’s a headache!

  2. Ben on April 16, 2011 at 9:39 am

    Somehow, I don’t think it will really help Houston much…

Leave a Reply

Your email address will not be published. Required fields are marked *

Current day month ye@r *

DSMok1 on Twitter

To-Do List

  1. Salary and contract value discussions and charts
  2. Multi-year APM/RAPM with aging incorporated
  3. Revise ASPM based on multi-year RAPM with aging
  4. ASPM within-year stability/cross validation
  5. Historical ASPM Tableau visualizations
  6. Create Excel VBA recursive web scraping tutorial
  7. Comparison of residual exponents for rankings
  8. Comparison of various "value metrics" ability to "explain" wins
  9. Publication of spreadsheets used
  10. Work on using Bayesian priors in Adjusted +/-
  11. Work on K-Means clustering for player categorization
  12. Learn ridge regression
  13. Temporally locally-weighted rankings
  14. WOWY as validation of replacement level
  15. Revise ASPM with latest RAPM data
  16. Conversion of ASPM to" wins"
  17. Lineup Bayesian APM
  18. Lineup RAPM
  19. Learn SQL