I don't think we can use a percentage success rate like that with any value.
1-There are only 32 offenses in the NFL.
2-There are only so many RB openings in those 32 offenses.
3-So, success is tied to opportunity.
When you get to the later rounds you have a high number of players taken
1 Round - 0
2 Round - 2
3 Round - 4
4 Round - 7
5 Round - 4
6 Round -5
7 Round - 6
So in rounds 2-3 you have 6 picks and 4-7 22 picks . So to hit your 67% success rate the 4-7 selections would need 14.74 or 15 rbs to succeed in your criteria. That's never going to happen--not enough offenses to play rookie RBs.
We should probably look at the median/mean production of the backs chosen by round in last ten drafts. I think this would allow a more fair comparison. But we need data, time, etc.
In the meantime, what if we look at the best back chosen in each round and see what the quality difference is:
2nd Round - Cook
3rd Round - Kamara
4th Round -Cohen? They all kind of are meh
5th Round-Aaron Jones
6th Round-Elijah McGuire(going off production)
7th Round-Chris Carson
I will give you that the 2nd and 3rd round backs are definitely the best. But an Aaron Jones or Chris Carson in the 5th or 7th respectfully--yes please. Kamara/Cook is a hair better than an Aaron Jones. I'll use less premium picks to take more chances to hit on those guys than using my 2nd rounders on them. Third round is the gray area for me--I'd rather take a more valuable position but I'd take a back if the quality is there.
How many 5th rounders is a 2nd rounder worth?