Pinned Tweet
Still working on the UI but if you are curious of my epm models results they are linked here. Please ask any metholodogy questions or anything. nba-epm-projections.streamli…
I added a backtest feature if anyone is curious what old results looked like
Replying to @bballstrategy @DannyAnalytics5
Always thought there was lots of value in reviewing any database, analyzing it hard and creatively. By the creator and / or users. At time permits and interest exists.
Fiddling with a spread picks model using my player projections. Have lost two picks this season so far (Orlando and Bucks spread). Don’t expect to keep up this rate but damn
Danny retweeted
Ausar Thompson:
• 98th% turnover generation
• 94th% defensive shot selection
• 89th% offensive rebounds
+3.82 ARC per 100 possessions
Curry: 31st pct in those decomps, but 100th in offensive conversion
ARC +5.79: 50% more impact per possession
Offensive Conversion is law.
Danny retweeted
Contemporary all-in-one impact metrics often regress box score actions onto RAPM to find their weights.
A fitted model can give a steal about 3 times the value of a point scored leaving us with an opaque number that can't be decomposed into how or the degree they affect the game
why not just count em up?
Introducing ARC v3
An Expected Box Value Prior RAPM (EBV RAPM)
That aims to quantify impact using possession by possession accounting 🧵














