Illinois 77, Minnesota 67 Postgame
- Status
skyIdub
"I can't tell you. Secret."
Career night for random dude!
Haven’t given up 70 since Nebraska!
GrayGhost77
- Centennial, CO
And statbroadcast has 58 possessions.
Question on the possessions formula…
Why do we have to do FT x .44?
Are we really not capable of deciphering which type of FT it was? In my view, an And1 or technical/flagrant FT should not count as a possession and any other FT pair or Bonus should, including a missed front end. Of course if fouled on the shot then the free throws count but not the shot attempt.
Maybe it doesn’t really matter enough, but seems like the scorer could easily tag the type of free throw.
Why do we have to do FT x .44?
Are we really not capable of deciphering which type of FT it was? In my view, an And1 or technical/flagrant FT should not count as a possession and any other FT pair or Bonus should, including a missed front end. Of course if fouled on the shot then the free throws count but not the shot attempt.
Maybe it doesn’t really matter enough, but seems like the scorer could easily tag the type of free throw.
I don't think you "have to" use 0.44... some are using 0.475 (really just depends on the analytics research that's used for the multiplier)
All boils down to the feasibility of tracking as you say. It is technically possible for official scorers to tag the type of free throw, but currently that granular data is not available in the standardized box scores or play-by-play data used by most contemporary sports analytics sources. The multiplier is used as a compromise, and across a large sample size, it has been shown to be very accurate.
Here's what I think is happening here.
We scored 77 points.
The formula for possessions is: (FGA - OREB) + TO + (y × FTA).
Y can be anything between 0 and 1, but is generally agreed to be 0.44. But for this site, it does not appear that y is 0.44.
Since they have our ppp at 1.228, to figure out what y was for our game, we need to solve this problem:
77 / {(59 - 17) + 10 + (y × 19)} = 1.228
y rounds out to approximately 0.563, which I'd imagine is the factor that they are using.
Last edited:
^^^ Using the agreed upon formula, this is our points per possession against Minnesota.
While not exact, I’d say that’s a pretty accurate finding.
Using 0.563 as the constant (instead of 0.44), Purdue scored 0.994ppp against USC. The site has it listed at 0.992.
Using 0.563 as the constant, Vandy scored 1.2469ppp against FLA. The site has it listed at 1.256.
Using 0.563 as the constant, Arizona scored 1.0969ppp against UCF. The site has it listed at 1.110.
….
So in conclusion, 0.563 is significantly closer to the website’s constant than 0.44.
Using 0.44 as the constant, Illinois scored 1.275ppg against Minnesota. The site had it listed at 1.228.
Last edited:
Someone write them a strongly worded letter and demand the formula be publicized!! lol
delete
Last edited:
Here's a blog from Torvik about it - wouldn't be surprised if there is updated info available
Kenpom esoterica
Adam Chorlton and Bart Torvik discuss Wisconsin sports, mostly the Badgers.
adamcwisports.blogspot.com
tl;dr
Field Goal Attempts + Turnovers + (Free Throw Attempts * .475) - Offensive Rebounds
Team rebounds need to be accounted for properly
Using that exact formula, our ppp would be 1.261 against Minnesota. On the website it's listed at 1.228.Here's a blog from Torvik about it - wouldn't be surprised if there is updated info available
Kenpom esoterica
Adam Chorlton and Bart Torvik discuss Wisconsin sports, mostly the Badgers.
tl;dr
Field Goal Attempts + Turnovers + (Free Throw Attempts * .475) - Offensive Rebounds
Team rebounds need to be accounted for properly
Last edited:
Don't think the line above is true. "The formula that both T-Rank and Kenpom use..."Here's a blog from Torvik about it - wouldn't be surprised if there is updated info available
Kenpom esoterica
Adam Chorlton and Bart Torvik discuss Wisconsin sports, mostly the Badgers.
tl;dr
Field Goal Attempts + Turnovers + (Free Throw Attempts * .475) - Offensive Rebounds
Team rebounds need to be accounted for properly
Kenpom himself has stated he uses this formula for possessions: (FGA - OR) + TO + (0.44 * FTA)
RabidDawgClassic
- Los Angeles, CA
Um… I was told there would be no math here. 
I was math major and this whole formula thing disturbs me LOLUm… I was told there would be no math here.
This is a great post, EXCEPT the bolded. Underrated? How so? They keep losing... They are 10-8. 55.5%. If we were that record... Anyway, eventually you ARE your record.
- Status