Patches pal nailed it: I don't like this scheme.
Seattle isn't the first team to try and resurrect Seifert's elephant. Mike Nolan attempted the trick in 2006. The problem then, as I would argue the problem is now, was that San Francisco lacked the talent. It could move a defensive end or linebacker around attempting to find mismatches, but it couldn't make that player better.
The assumption is that because Chris Clemons had a career season, Carroll's elephant succeeded. The fact that the Seahawks defense was among the worst in the league, and that the pass defense specifically was among the worst in the league, contradicts that. Before any one of Brandon Mebane, Colin Cole or Red Bryant was injured, Seattle ranked 20th in pass defense efficiency. Before Cole or Bryant went down in Oakland, the moment often cited as the beginning of the collapse of Seattle's defense, Seattle's pass defense ranked 20th. By the end of the year, whether because of Bryant, scheme adjustments, or just an expanded sample, Seattle ranked 25th.
The problem is two-fold: Team stats are a very rough way of judging individual players. Even if Clemons were an elite defensive end, it wouldn't ensure that Seattle's overall pass defense would work.
The other problem is the bigger problem, in my opinion. That is, apart from sacks totals, what evidence do we have that Clemons was helping Seattle's defense?
A "sack" in some ways is like an RBI. It depends on the execution of other players, and so though it's framed as an individual stat, no sack is accomplished individually. It is also context independent, and that means not all sacks are equal. We can think of the sacrifice fly that produces an RBI but does not improve a team's chances of scoring. A sack itself can never be damaging, but the method with which a team accrues sacks can be. Most notably, a team is much more likely to sack a quarterback if it blitzes, but it also risks more completions, more first down conversions, more yards and more touchdowns. It's a trade off, and one reason some defenses eschew blitzes and all defenses seek to create pressure from a four man pass rush.
Which leads me to something I have wondered but can not confirm: Did Clemons's high sack totals belie his actual ability as a pass rusher?
The missing component is Seattle's performance when blitzing. And I say that is the missing component because we know this: Seven of Clemons 11 sacks were accomplished when Seattle blitzed. I didn't know if that was uncommon or not, so I looked at all 14 players that finished in the top ten for total sacks. STATS Inc doesn't use fractions of a sack, so I didn't either.
Here is how it looks:
That isn't conclusive, mind you, but it does give some context for Clemons' high sack total. As does the fact that Raheem Brock had nine sacks, but only one sack when Seattle blitzed. The missing component is knowing just how good Seattle was when blitzing, and when blitzing compared to its overall defense. That still wouldn't be conclusive, but it would get us closer. It could tell us, for instance, if Clemons needs a blitz to succeed as a pass rusher, and Seattle performed poorly when it blitzed, that though Clemons notched a bunch of sacks, he might not have been an effective pass rusher.
Right now though, we are on the trail of what looks like a sensible explanation for how Seattle's pass rush improved its sack totals, but Seattle's pass defense barely improved. Does that invalidate the Leo position? No. Does it suggest that Clemons was not helping the pass defense as much as his raw sack totals would suggest? It does. A team could rush ten defenders on every snap and lead the league in sacks, and also field the worst defense in the history of the NFL. Conversely, a team could never once land a sack, but create consistent, fast arriving pressure, force premature check downs and mistake throws, and be the best defense in NFL history.
The take away is, a sack isn't a sack, and high sack totals are not a sure sign of a successful defense.