APBRmetrics

I was wondering if this is a reasonable way to calculate rebounds for each player.

In a game there is roughly 45.5 missed shots, and with missed 2nd free throws, let's estimate a game has 47 missed shots.

Offensive rebounding % for each player on the Cavs and the defensive rebounding % for each player on GSW is:

Kyrie Irving 2.3%; J.R. Smith 1.1%; LeBron James 6.1%; Kevin Love 6.5%; Tristan Thompson 16.3%
Steph Curry 14.9%; Klay Thompson 9.0%; Harrison Barnes 12.0%, Draymond Green 22.2%, Andrew Bogut 20.3%

The Cavs = 32.3%, GSW = 78.4%. Together they = 110.7%.

Whenever a player is on the floor, my thought is that they would get their percentage divide by 110.7% of the rebounds. So for Tristan Thompson, when he is offensive rebounding, he would get 16.3%/110.7% = 14.7% of the rebounds while he is on the floor.

To compare with average, I would then multiple that by all missed shots. 14.7% * 47 = 6.9 offensive rebounds for TT if he played the whole game. Adjusting for his minutes, (29.6/48) * 6.9 = 4.25. When compared to his average of 4.1, it is pretty close.

Comparing the other players to their average:

Kyrie Irving = .76, actual average .7
JR Smith = .34, actual .3
LeBron James = 2.12, actual 2.0
Kevin Love = 1.77, actual 1.7
Tristan Thompson = 4.25, actual 4.1

Steph Curry = 4.52, actual 4.7
Klay Thompson = 2.84, actual 3.0
Harrison Barnes = 3.31, actual 3.5
Draymond Green = 7.55, actual 8.3
Andrew Bogut = 3.00, actual 3.5

Pretty similar, not perfect of course, as each player's rebound percentage will change based on who's out there. Does suggest Cleveland is a better offensive rebounding team than GSW is defensive, and that seems to check out. Could this be good enough to accurately project final scores? This method at least adjusts for different types of players. Perhaps it could be weighted a little based on lineup.

It's interesting that all 5 Cavs were lower than their prediction, while all 5 Warriors were higher.
This wouldn't be explained by the Cavs shooting better than expected: While that would leave fewer OReb opps for them, it would also predict fewer DReb for the Dubs.

I'll suggest it's because the Cavs were more fatigued -- their top 5 guys were playing more minutes than usual.Love's minutes were down for various reasons; these others got barely (or not even) half as much rest as in the RS. And most of that came in blowout times.

It is hard to say. The prediction is the league average though. In the playoffs Cavs are 4/16 in offensive rebounding %, and GSW is 12/16 in defensive rebounding. Given that, it would make sense that when faced against each other, Cavs would have better offensive rebounding and GSW would have worse defensive rebounding when you compare to each of their averages.