APBRmetrics

Crow wrote: ↑Tue Apr 29, 2025 7:28 pm It was a good performance.

Pre-season adjustments by player and / or team?

Preseason adjustments are done at the team level, but weighted by the number of minutes played by players, so if you underperform without playing any of your starters, there's less weight given.

KMedved, what ws the final error for Darko? As linked before, the projections are here: https://docs.google.com/spreadsheets/d/1UIU0Wse3qKbcXRQSufA1B4qd7Tt2H_tgg1o1ewMw1ZM/edit?gid=235611189#gid=235611189, and the final results were: 1) 0.533 R^2 2) 7.24 MAE 3) 9.15 RMSE I generally think R^2 is th...

It's the same system along the way, with some incremental improvements here and there. The big thing I've done the last few years is incorporating more preseason data, but there's no other major changes recently. I forget when exactly I launched DARKO. Before DARKO, I was just using RAPM with an agi...

Planning to score your projections when final results are in? Post here? Sure, I'll post them when finished. The spreadsheet has live scoring (pulling from BBR for now) as well, although to reiterate, I agree with the concept that projections that weren't entered here shouldn't "count" (f...

What’s the difference between the LEBRON and EPM projections and the advanced model and Vegas advanced model ones? Sorry about that - those aren't mine. My projections are on the "DARKO wins" tab. @Knarsu3 asked me to add his projections to the spreadsheet as well for scoring purposes, bu...

I posted the DARKO projections on Twitter, and they're available here: https://docs.google.com/spreadsheets/d/1UIU0Wse3qKbcXRQSufA1B4qd7Tt2H_tgg1o1ewMw1ZM/edit?gid=235611189#gid=235611189), but I don't actually mind a standard that they need to posted/linked here to be included. It's fairly low cost...

1) Yes - DARKO is always a projection. The historic trajectory is a snapshot of where DARKO projected a guy to be as of that game. I don't "go backwards" and update with "well, in retrospect, this uptick in performance was more sustainable than DARKO thought." The LOESS smoothing...

re: Updates - I run them daily, but sometimes the linkage to the the Shiny App breaks (I am not gifted with front end web dev stuff, and my partner on the app is often busy). The Google Sheet link here will update daily and is more reliable - just without the fun visuals: https://docs.google.com/spr...

I am now more awake to full Darko's large lead over the 2 blends being almost entirely based on preseason. Darko without preseason is just on par with the 2 blends. I don't know off hand how much other metrics use preseason. If they cared about losing to DARKO they might either monitor DARKO with p...

Generally, while I'm not gonna pretend I'm not thrilled with how well DARKO is performing here, there is just a ton of noise in any single-year projection. The preseason model is doing incredibly well because it was super down on Portland for instance, which was largely plagued by a Damian Lillard i...

I am very surprised to see EPM so far down that list. Were all of the stats-based predictions using similar approaches to minutes projections? That's where Retrodiction (using actual minutes played but preseason player ratings) can really help level the playing field. DARKO uses preseason minutes p...

I am also tracking some public win projections here:

https://docs.google.com/spreadsheets/d/ ... 1209160265

The preseason-only model (which use just preseason results) is dominating.

Our of curiosity Ryan, what's the difference between the version of RAPM you posted here, and the version on NBAShotCharts? I see slight differences throughout. Is this a playoffs inclusion thing?

This is a very cool idea. Do you have any retrodictions available so we can go "ooh, ahh" about the sleeper picks this system nailed?

I am interested in approaches for how to weight a sample of increasing size against some set of priors to predict future outcomes. This is somewhat python focused, but I am interested in general modeling thoughts here as this is a common issue to sports stats. I have created a dataset where I have e...