Just like that Rodgers guy, eh???
Last season 15 players drafted after the top 12 made the All-Pro first team. Maybe they're harder to find but there are for sure a lot of talented players available late in the first round.
The 2014 AP All Pro team included 12 offensive players, 12 defensive players, and 3 specialists (place kicker, punter, kick returner) for 27 players in all.
So, if 12 of those guys were top 12 picks, that 44% of the players out 5% of the 224 annual picks (excluding compensatory picks and picks lost to penalties which I'm not going to bother to count).
Some further adjustments are needed.
The only place kicker to be drafted in the top 12 in NFL history was Charlie Gogolak in 1966.
No punter has ever been drafted in the top 12 in NFL history.
No fullback or center has been drafted in the top 12 over the last 10 years, which is as far back as I looked for those positions.
Only 2 OGs have been drafted in the top 12 over the last 10 years: Warmack and Cooper in 2011.
Only 6 RBs have been drafted in the top 12 over the last 10 years: Richardson, Spiller, Matthews, Moreno, McFadden and Lynch. Among those, only Richardson, Spiller and Matthews were in the last 6 drafts, and only Richardson in the last 5, with RB shelf lives being notoriously short.
I did not consider kick returners since it conceivable a top 12 pick would make All Pro as a side job, such as Pat Peterson in 2011.
So, of the 27 positions, PK, P, FB and C need to be thrown out because there were no top 12 picks at those positions over the last 10 years. Expecting a top 12 pick at OG or RB would be a big stretch given how few there have been. It should be no surprise that none of the 2014 All Pros at those positions were top 12 picks.
That leaves 21 positions where the top 12 picks are highly concentrated, or at least not very thinly concentrated. On that basis 57% of last season's All Pros came out of the top 5% of the picks.
That 43% would come out of the other 95% of the picks should not be surprising, but clearly when looking at any one pick, the likelihood one way or the other is highly skewed.