Page 1
Page 1
Started By
Message

How exactly does on3 grade transfer classes?

Posted on 12/18/24 at 6:42 pm
Posted by NorthTxLSU
Dallas to Austin
Member since Nov 2018
12114 posts
Posted on 12/18/24 at 6:42 pm
Was just checking their site for how we are looking, and they have us at #22? Feels like we’re doing a lot better than that.

I know rankings are meaningless but just genuinely curious.

LINK
Posted by 1609tiger
Member since Feb 2011
3437 posts
Posted on 12/18/24 at 6:52 pm to
It looks like it’s attempt to give a net value after accounting for transfers out. Looks like it kind of overestimates the value of those transfers out. At this point there is more of them and many were role players.
Posted by 1ranter1
Louisiana
Member since Dec 2008
10550 posts
Posted on 12/18/24 at 6:58 pm to
Pitt being #25 with 0 in and 10 out is literally the dumbest ranking shite I’ve ever seen and whoever came up with the formula should never be allowed to touch a computer again
Posted by Reeaholic
Moss Bluff
Member since Jun 2019
1182 posts
Posted on 12/18/24 at 7:03 pm to
It's looks they tried to go too sophisticated and now it looks like arse.

Michigan lost 17 transfers out and only gained 4 transfers in but yet are ranked 5th. They netted 2 4 stars but net lost 14 3 stars, LSU netted 1 4 star and only had a net loss of 8 3 stars and is 17 spots lower than them.

Its not even by if the avg player rating gained vs avg player rating lost.

In have no idea what they're doing as I'd imagine most fans don't either.
Posted by lostinbr
Baton Rouge, LA
Member since Oct 2017
11687 posts
Posted on 12/18/24 at 7:07 pm to
quote:

It looks like it’s attempt to give a net value after accounting for transfers out.

I think you are correct. From the explainer at the top of OP’s link:
quote:

On3's Team Transfer Portal Index utilizes the On3 (P)erformance score to measure a team’s production during the transfer process, compared relative against its roster and not a comparison against other schools. This proprietary algorithm determines if a school has improved its overall team talent, stayed the same, or declined in talent during the transfer window.

So they are just looking at how the transfers out + transfers in affected each team’s overall talent relative to their current roster.

One big question to me is how they weigh the differences between quantity of incoming and outgoing transfers. For example LSU was always going to have a net loss in the transfer portal because we had more incoming signees than we had available scholarships. My first thought was that maybe they cap the negative weighting for outgoing transfers at the number of incoming.

But then I look at Pitt, who is ranked #25 with 10 transfers out and 0 transfers in. That makes absolutely no sense.

Either they’re saying Pitt’s roster got better by losing 10 guys, or they’re saying every team ranked below Pitt also got worse. The latter might be possible, but seems like there are some flaws in their formula at least this early in the process.
This post was edited on 12/18/24 at 10:18 pm
Posted by Lapaz
Member since Dec 2018
709 posts
Posted on 12/18/24 at 8:03 pm to
The relative nature of the ranking is bizarre. I think the basis is that if a team's average player ranking on their roster is higher than the average rating their portal players signed, then the portal has not made the team better. For example, if LSU's roster is rated as 85, while another team has a rating of 75, then I think the same portal class would be ranked worse for LSU than the other team. It doesn't tell you how good the portal class is.

It seems they also try to account for the quality of the players that are being lost. We lost a couple of highly rated players, so the difference between the ratings of the players we signed to the players we lost is only about 9 points. Other teams have a much larger difference.

I think using the team's roster to determine the portal ranking is dumb, but I do think there is value in evaluating how much the portal has helped the team. I don't think their algorithm does a good job of assessing that. Accounting for losses makes sense to me, but LSU's losses were largely unproductive.

If we're strictly evaluating the quality of the portal class, I'll take LSU's portal signees over just about any other team's. LSU has signed a lot of high quality players. LSU has signed 9 players with an average rating of over 87. Ole Miss is the next best with 9 players rated at just over 84. Oregon, Miami, and Duke's are the only ones in the top 25 with higher average ratings, but they've only signed 4, 2 and 1 respectively. There are teams in the top 50 with 1 signee slightly higher than 87, but that's unlikely to impact your season unless it is a QB. In my opinion, the ON3 team ranking is garbage.
first pageprev pagePage 1 of 1Next pagelast page
refresh

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on X, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookXInstagram