Phantom Island/Parker Flemming (of cfb-graphs.com) shared some new trends their seeing in roster building, viewing it through the lens of the two teams in the national championship. They talk through numbers in this episode and how this applied specifically to IU/Miami, and how those teams' roster building strategies are indicative of the future of the sport.
Paradigm Shift
Parker's criteria for evaluating the talent on a roster suggest a pretty big paradigm shift from what college football fans and media have grown accustom to:
- Instead of analyzing total recruiting talent on the team over a 4(ish) year window, it makes more sense to judge just the talent in the 2-deep that has entered the program over the previous 2 years.
- He points out that only about 40 players on the 105 person roster really impact your team's performance this year.
Why is this the case?
Pretty obvious - It's more difficult to project how good a high school recruit will be in two to three years than it is to project how good a transfer will be this year - there's just more randomness and unknown in scouting (watching tape and projecting how good that player will be) high school recruits than transfers.
What should coaches do about it?
- Regrade players every year to see where you can improve and where you should allocate money
- Do everything to keep your core players. If someone is on the margin, you can let them walk
- Aim to cut/improving the bottom 25% of your 2-deep ever year
- Get QBs out of the portal - High school QBs are rarely well enough developed to contribute immediately
- Prioritize Identity and Staff Continuity - given that there is significantly less randomness and tighter talent development timelines, it's really important to understand how can one player fit into what you already have. If you're unclear on your identity or you have constant staff turnover, it can be harder to answer that question
Other Interesting tidbits:
- Miami really developed most of their OL from high school. I think it's interesting given that this is the one homegrown position group. I think it makes sense given that (a) OL takes time to develop and (b) Mario is a former OL.
- A good chunk of talk about the QB position specifically: "QB is a necessary but not sufficient condition" - coaches should build the rest of the team then worry about finding QB.
- Historically, teams could get good, then stay good. That's no longer the case because you can't stack/retain talent for 4 year periods. We're about to see way more parody at the top of the sport. We might start to see intentional down cycles (like you see in pro sports) where teams are tossing the towel in on a season to save money for a future roster/season.
My thoughts:
- I don't think anything in the episode was groundbreaking, but I haven't heard anyone frame it so straightforward yet; Every fan knows high school recruiting rankings don't matter as much, but I haven't seen anyone straight up say 'maximize talent in the 2 deep in a 2 year window'
- A lot of fans/pundits have been saying 'if you can, you want to get most of your talent from high school' (including some here). I'm starting to second guess this. All four semi-finalist this year had >40% of snaps played by transfers, with three of the four being above 50% and two above 60%. Going to be interesting to see if this is a trend or just a single data point.
- It does make me wonder if we made the right hire... while there were many issues with Pry, it does feel like we hired a guy who recruited on relationships right as the pay-for-play era started. I do have some concern that James Franklin will be focused on 4-year recruiting instead of 2-year recruiting
Comments
Thanks for posting.
I think Franklin's roster moves have shown a couple of themes:
Clean up messes and patch holes. (We haven't had the consistent and high end talent to compete.)
Focus on the trenches. (He has upgraded both depth and talent on both sides.)
Try to show improvement, but really peak in 3 years. (This may be the 'gamble', but most fans/schools are willing to give a new coach grace for a few years. Seats become hot starting 3 years into a new hire if there's no obvious improvement)
Kneecap psu (I honestly think this is playing a part, and im ok to watch it.)
The successful teams will have the stack of their 2 deep in the last 2 years of eligibility.
I am not worried about Franklin's method. #25 in HS recruit #3 transfer. Gotta get the HS kids looking for the start of the 2 deep for 2 years from now. Then get the transfers in to the gaps that are exposed and keep the ones that will be starters as Juniors/seniors
This goes inline with what I said in the other thread, take a transfer QB even if they have 1 year because they are that important and that relied on.
Olines need built still because they dont enter the transfer portal like everyone else.
I think taking an actuarial approach is what teams are going to start, if they are smart. What are the premier positions, in the NFL its easy to see this based on the salaries, and I would assume that QB, DE, LT would be most important in college but I could see some being switched up based on play styles. I know some teams like a 3 man front because finding 2 DTs is harder than 2 MLBs, which the NFL its easier to find 2 DTs than 1 that can play a 3 man front.
In addition, with out a salary cap you can mess up the market, but the data is still there and probably reveals a lot.
I would like more parody but I think what we'll see is more parity...but I like that, too
I think this goes in line with the changing money landscape too. I think that Rev Share will be heavily focused on about the top 60, NIL will focus on the two deep with a few high end outliers. The other 45 will essentially be scholarship level.
Agreed.
The median FBS football player makes ~$3k in NIL. So it's already super top heavy.
like the NFL, its like 65% make within 5% of league minimum