Ryan Bates

46 posts

Ryan Bates

Ryan Bates

@rj_bates2

UIUC Stats ‘24 | Inquiry: [email protected]

Katılım Ekim 2025
137 Takip Edilen101 Takipçiler
Ryan Bates
Ryan Bates@rj_bates2·
@mikegrib8 Another way to look at it is that ratings will be more reactive to change in the first few days of a season because the new information will be weighed more heavily compared to say a few month time scale in-season.
English
1
0
0
48
Ryan Bates
Ryan Bates@rj_bates2·
@mikegrib8 Relative to the ratings of other players, his new information would be weighted more heavily because his older information (relative to other players) would play a smaller role.
English
1
0
0
65
Ryan Bates
Ryan Bates@rj_bates2·
I’ve added time-decayed three factor influence (PPS, REB%, and TOV%) to the RAPM page so you can better see how players achieve their impact. The PPS estimates are points added per 100 team true shooting attempts, so free throws are included. docs.google.com/spreadsheets/d…
Ryan Bates tweet media
English
2
5
60
7.6K
Ryan Bates
Ryan Bates@rj_bates2·
@mikegrib8 There are no age adjustments but it’s definitely something to look into due to how quickly some players can develop or how situations change. I know Jeremias Engelmann does a sort of age adjusted RAPM.
English
1
0
0
94
Ryan Bates
Ryan Bates@rj_bates2·
@mikegrib8 Yes, older possessions (or in some cases rebounds or true shot attempts) are exponentially weighted less relative to how old they are to a certain date. This is from the end of the season.
English
1
0
0
229
Ryan Bates
Ryan Bates@rj_bates2·
Blackwell is an excellent addition to Duke. Has a unique combination of high usage, high efficiency, solid off rebounding, low turnovers, and underrated defense. According to my time-decayed effect on team’s TOV% (coming soon), Blackwell rates as one of the best in the country.
John Blackwell@_Johnblackwell1

Committed! #TheBrotherhood

English
0
0
3
530
Ryan Bates
Ryan Bates@rj_bates2·
Sneaky good pickup for Ben McCollum. Coleman has a good combination of rim finishing, shooting and defense. He ended the year with a +4.1 time-decayed DRAPM, an elite mark for not only a freshman guard but a mid-major player as well.
Tyreek Coleman@TyreekColeman

Let’s do it!💛🖤

English
0
0
0
348
Ryan Bates
Ryan Bates@rj_bates2·
It’s hard to find a non-paywalled list of players in the transfer portal. Thankfully @CFB_Data has a great API at collegebasketballdata.com. Here are the top players in the portal (some pending eligibility) by my time-decayed RAPM. Jasper Floyd could be an underrated addition.
Ryan Bates tweet media
English
2
6
38
5.5K
Ryan Bates
Ryan Bates@rj_bates2·
The Bidunga 3P% projection discussion reminded me of a great article by @jessefischer33 of hierarchical modeling. Bidunga’s low 3PAr and FT% should probably be pulling his projection towards a much lower estimate than league average. tothemean.com/2020/09/19/hie…
Evan Miyakawa@EvanMiya

Flory Bidunga isn't just the best player in the portal this year, he's the 2nd best rated transfer ever at EvanMiya.com. The only player to receive a higher transfer rating was Yaxel Lendeborg last year. He turned out to be pretty good.

English
0
0
2
429
Ryan Bates
Ryan Bates@rj_bates2·
@bballstrategy The parameters were set to best predict the following day, so I would take a date and use all data previous to it to predict. I don’t have a concrete answer for your second question but generally these results are more precise for players with lots of career possessions.
English
0
0
1
27
basketballstrategy
basketballstrategy@bballstrategy·
@rj_bates2 How did you find, set, test, confirm or reset your time-decay factor? How many games do you need for time-decay RAPM to be superior to standard RAPM?
English
1
0
0
87
Ryan Bates
Ryan Bates@rj_bates2·
The most surprising player ranked highly in my time-decayed RAPM is Tulsa’s Tyler Behrend. The 6’10 Jr is only averaging 6 PPG in 20 MPG but is 12th in ORAPM. The reason? Tulsa has a 136(!) adjust ORTG and is 33 points per 100 possessions better with him on the court this year.
Ryan Bates tweet mediaRyan Bates tweet media
English
1
1
10
1.5K
basketballstrategy
basketballstrategy@bballstrategy·
@rj_bates2 Is the file at this link updated thru end of regular season? thru NCAA tournament to date?
English
1
0
0
73
Ryan Bates
Ryan Bates@rj_bates2·
Time Decay RAPM for NCAA. Will try to update this as frequently as possible. More meaningful for players with a lot of possessions played. docs.google.com/spreadsheets/d…
English
2
1
2
615
Ryan Bates
Ryan Bates@rj_bates2·
@AlbertEzem I have the hyper parameters optimized to predict the following day. Multi-year approach is definitely more predictive than single year. Thank you for your interest!
English
0
0
1
28
Albert Aranye
Albert Aranye@AlbertEzem·
@rj_bates2 Did you find time decay improved predictiveness? Super interested in your work btw!
English
1
0
0
76
Ryan Bates
Ryan Bates@rj_bates2·
To adjust for seasonality for my time decayed RAPM, I estimate league avg offensive rating for each day of the season. Apologies for the base R plot, but it’s clear how much offense has exploded in college basketball. The early season struggles match the peak of only 3 years ago
Ryan Bates tweet media
English
3
0
60
3.1K
Ryan Bates
Ryan Bates@rj_bates2·
JP Estrella could be an underrated player in the portal if he decides to enter. Career 64 TS% and 16 OREB% Currently second on Tennessee in my time decayed RAPM. Could be a very solid starter/rotational big for a high major team if he improves on defense over the offseason.
English
0
0
1
364
Ryan Bates
Ryan Bates@rj_bates2·
Flory Bidunga should probably be in preseason NPOY talks next season. The 6’9 sophomore has made the biggest jump among returning players in my time decayed RAPM since the beginning of the season. Shoutout Josh Schertz for his development as well.
Ryan Bates tweet media
English
1
1
12
1.6K