Four Factors Importance Using Win Probability Added

[SkyJuke]

I did a quick test of this(wanted to see if there were any changes from the regular season to the playoffs) and got these results.
Regular Season — No Intercept (R² ≈ 0.99)
Factor Oliver Full Past (≤2009) Recent (≥2015)
Shooting 40% 42.5% 42.0% 43.5%
Turnovers 25% 40.0% 40.8% 37.6%
Rebounding 20% 8.5% 8.2% 9.5%
Free Throws 15% 9.1% 9.0% 9.4%

Playoffs — No Intercept (R² ≈ 0.78)

Factor Full Past (≤2009) Recent (≥2015)
Shooting 43.9% 43.7% 41.0%
Turnovers 37.0% 36.9% 40.2%
Rebounding 8.7% 8.6% 12.0%
Free Throws 10.4% 10.8% 6.8%

Hacked this with an LLM so might not be accurate but it seems there's little difference between the regular season and playoffs.
And rebounding importance has grown over the years.

[Crow]

Another of many cases of other analysts finding factor weights significantly different than Oliver's original. I would go with the average of these followup studies.

[Crow]

Thunder strongest on the 4 biggest factors.

None of the other contenders more than 2. Rockets just one.