Basketball Transfer Portal '25

[GoBearcats31]

Take with a huge grain of salt, but ...

Current 247 transfer portal rankings

Outgoing:
- 54 Mitchell
- 217 Skillings
- 257 Betsey
- 307 Page
- 366 Reed
- NR: Hickman, Griffith

Incoming:
- 22 Thiam
- 162 Celestine
- 220 Kriisa
- 264 Harris
- NR: Haynes



On3 rankings

Outgoing:
- 56 Mitchell
- 132 Betsey
- 205 Skillings
- NR: Page, Reed, Hickman, Griifth

Incoming:
- 13 Thiam
- 171 Harris
- 187 Haynes
- NR: Celestine, Kriisa

[sedz]

Huge grain of salt here too. NetRtg is the production we need to replace. It's BPM weighted by minutes played.

OUTGOING	BPM	%min	NetRtg
Mitchell	6.8	70.2	4.8
Skillings	5.8	47.2	2.7
Reed	3.1	41.5	1.3
Page	1.2	20.0	0.2
Betsey	-1.2	27.2	-0.3
Hickman	-2.5	18.4	-0.5
Griffith	-3.5	7.7	-0.3

INCOMING	BPM	%min	NetRtg
Thiam	3.1	64.9	2.0
Harris	2.7	66.1	1.8
Haynes	2.8	62.8	1.8
Celestine	1.9	54.8	1.0
Kriisa	3.3	10.9	0.4

We lost a total of 8.0 to the transfer portal and we currently have 7.0 incoming. Of course this doesn't attempt to predict player progression, regression, or change in minutes. We lost another 4.9 to graduation and we return 4.7. Add those up and our total NetRtg last year was 17.6. Incoming plus returning is 11.7. Maybe we get 3.0 from freshmen, bringing the total to 14.7. We need to be at least 20 for an at large bid. That would mean about 1 BPM progression on average per position. Seems doable.

[sedz]

BearcatBunch added some rather speculative names to their tracker based on Kriisa following their social media (Ian Schieffelin, Ven-Allen Lubin, Rylan Griffin). I doubt those are serious connections.

Andre Morgan followed AK Okereke, 6'7 forward from Cornell. That would be a home run. Okereke posted a 7.2 BPM this year as a high usage point forward and slasher. I don't know what to make of any of these connections though. We're probably out of money.

[GoBearcats31]

We lost a total of 8.0 to the transfer portal and we currently have 7.0 incoming. Of course this doesn't attempt to predict player progression, regression, or change in minutes. We lost another 4.9 to graduation and we return 4.7. Add those up and our total NetRtg last year was 17.6. Incoming plus returning is 11.7. Maybe we get 3.0 from freshmen, bringing the total to 14.7. We need to be at least 20 for an at large bid. That would mean about 1 BPM progression on average per position. Seems doable.

Losing 8.0 and adding 7.0 but currently have 7 dudes out the door and 5 incoming -- do these numbers account for that?

Where do your numbers / advanced stats come from anyways?

[sedz]

Losing 8.0 and adding 7.0 but currently have 7 dudes out the door and 5 incoming -- do these numbers account for that?

Where do your numbers / advanced stats come from anyways?

Outgoing is accounted for. The bottom three guys had terrible BPMs but they barely played so the total impact is only about negative 1. Mitchell on the other hand played 70% available minutes with a 6.8 BPM. That leaves a gaping 4.8 hole.

Incoming is harder to say because we don't know how many minutes these guys will play. I just used last year. Kriisa will probably play more and Harris will play less than that. But overall I think it evens out.

I use Torvik for almost everything. https://www.barttorvik.com/team.php?...nati&year=2025

[DesertFog]

I am a huge stats fan, but what is missing is a stat that accounts for the importance and effectiveness of the coach.

[sedz]

I am a huge stats fan, but what is missing is a stat that accounts for the importance and effectiveness of the coach.

Yes, absolutely. That's the reason our roster was ranked #27 going into last season but we finished #44. It doesn't matter how good your players are if they're going to stand around and watch one guy chuck up 40% midrange jumpers. But I can't control that. All I can do is evaluate.

[sedz]

Andre Morgan followed AK Okereke, 6'7 forward from Cornell. That would be a home run. Okereke posted a 7.2 BPM this year as a high usage point forward and slasher. I don't know what to make of any of these connections though. We're probably out of money.

Committed to Vanderbilt. That didn't take long.

[leo from jersey]

I am a huge stats fan, but what is missing is a stat that accounts for the importance and effectiveness of the coach.

totally agree - very astute comment -- stats are not comparing like with like as a good coach can improve the stats. I will wait to see how these new guys perform once on the court.

[DesertFog]

Yes, absolutely. That's the reason our roster was ranked #27 going into last season but we finished #44. It doesn't matter how good your players are if they're going to stand around and watch one guy chuck up 40% midrange jumpers. But I can't control that. All I can do is evaluate.

I am not familiar with the way these statistical models work, but would it be possible to put in a number that represents the skills of the coach, and then see if the predictive ability of the model improves?

[sedz]

I am not familiar with the way these statistical models work, but would it be possible to put in a number that represents the skills of the coach, and then see if the predictive ability of the model improves?

Yes, for sure. Kenpom and Torvik have many inputs in their models, including coaches or at least prior seasons. In fact, before substitution data came along, models were almost entirely team based. Most models now (including Torvik) use the 16 step player based system laid out by Dan Hanner in Basketball Prospectus back in 2012: https://web.archive.org/web/20130530...articleid=2415

Hanner prescribed an adjustment giving "high weight to the most recent season for each coach, but also some weight to all historic seasons". Torvik says "each coach has their own ratings on offense and defense, which are based on their past performance compared to their school's overall program mean". Kenpom says "The components and weighting are based on a regression of the past nine seasons ... Projected offense is largely determined by the quality of a team’s offense over the previous three seasons."

I don't have a model myself. I'm just using BPM as a crude first building block. It's easy to see what effect adding a player or changing minutes distribution will have. And it's pretty close. My crude NetRtg calc using BPM only for this year's roster was 17.6. We ended at 16.3 in Torvik's model and 14.9 on Kenpom. So their models do think we are worse than what you'd expect from the roster alone.

Hanner closed that article with "Data can't tell us everything about the upcoming season. There will always be surprises, and that's what makes college basketball great. But the techniques for predicting the season are getting better, and through them we can gain insight into teams we may have overvalued or overlooked."

[GoBearcats31]

Sam Kayser says UC and many others have contacted FAU’s Baba Miller. 7 footer naturally.

[GoBearcats31]

OSU’s Sean Stewart hitting the portal. Considered UC last year. Not great but could help at the 4. Averaged almost 6 and 6.

[sedz]

Sam Kayser says UC and many others have contacted FAU’s Baba Miller. 7 footer naturally.

Miller played a lot of PF alongside 7 footer Matas Vokietaitis. Miller posted a 4.5 BPM playing 30 mpg. He made 35% from three on 112 attempts, and 56% effective field goals. Good passing for a big with solid rim protection. Led the Owls in defensive BPM. He would be perfect for our team (and many others). I'd make him the priority if we can afford it.

[sedz]

OSU’s Sean Stewart hitting the portal. Considered UC last year. Not great but could help at the 4. Averaged almost 6 and 6.

Stewart would give us a dependable defender and rebounder at PF, which may be important if McKinley struggles in those areas as a freshman. Stewart is not a perimeter threat, turns the ball over a lot, and shot just 50% at the stripe, leading to a negative 1.5 offensive BPM. But he posted a solid 3.3 defensive BPM with elite rebounding (10% OReb, 27% DReb) and shot blocking (6% of twos).