In Defense of Sean Considine - Kinda

In yesterday's post, it turns out I made a bit more of a leap than some were willing to grant when I said that besides Brian Dawkins "there's not a single other safety on the roster who's proven he's a legitimate NFL starter."

I got kind of called out on it in the comments, which I always appreciate.  That's a lot better than just thinking everyone agrees -- when they don't -- and it also indicates areas where I should spend more time doing some research.  So while I did cover some of this ground before, there's certainly a lot more we can discuss here.

Here's the thing about Quintin Mikell.  I like him just as much as you do.  Everyone likes rooting for the guys who come into the league as unrestricted free agents, bust their tails on special teams for a few years, and then earn their spots on the field.  Much better than the top ten picks who get handed a bunch of money right out of college but then fall short of the hype. 

So there's nothing personal about my concerns about him.  Hell, I really hope he proves me wrong, since that would be a great sign for the defense in 2008.  I'm just not sure, based on what I've seen so far, how Mikell is going to handle a full-time starting role once teams start going after him.

- - - - - -

Let's begin with the stats.  To remove the special teams effect as much as possible, these are only the numbers for the games during which these four guys started last year:

Those numbers aren't that easy to compare, however, so here's the same table with every guy's stats extrapolated over a 16 game season:

The obvious first takeaway is that Dawk clearly wasn't himself last year.  No sacks and one interception just isn't his thing.

But we knew that already, so the second takeaway should be that we should avoid reading too much into small sample sizes, because J.R. looks like an absolute monster in those numbers.  Especially because we don't know how many of those 15 tackles might have been picked up on special teams (although Mikell and Considine both played on STs as well).

Randy Moss certainly knows his name, though.

And the final takeaway is that Mikell had a nice season, but it's a big old stretch to claim it was anything like Pro Bowl caliber. 

In fact, if you compare Mikell's season to a couple other recent samples, you can get a better sense of context:

The 2004 season was the year the Eagles sent three guys from the secondary to the Pro Bowl, including Dawkins.  I picked it randomly from his five Pro Bowl seasons mostly because I wanted to look at Michael Lewis' stats as well.  That guy sure could tackle. 

Here's the 16-game extrapolation:

Now I think we really need to take Considine's tackle numbers with a grain of salt, because he tallied a fair number of special teams tackles that year according to the Eagles' stats (although, again, these tables are based only on the games each player started).  Also, it's generally not a good sign when your safeties are making too many tackles ... especially on running backs ... 10 yards downfield.

But Dawkins' sample season was clearly much better than either of the other guys.  Beyond that, I'm not seeing a whole lot of obvious separation there between Mikell and Considine.  And while Considine's injury history is certainly something to consider, we know now that the first surgery he had to fix the problem didn't do the job.  Maybe this time the doctors have it figured out. 

- - - - - -

So those are the numbers, but let's talk about what we actually see on the field as well.  Generally speaking, fans only notice two things while the game is going on:  1) the really, really good plays and 2) the really, really bad plays.  We don't see everything in between, like when a safety correctly plays a coverage and takes away a deep receiver or when a defensive tackle holds up two blockers so a linebacker is free to make the tackle.

It's actually really hard to judge these guys properly in real time, so what we end up doing is a kind of evaluative shorthand where we tally up all the good plays (big hits, interceptions, passes knocked down), compare those to all the screw ups (missed tackles, balls caught on their coverage) and then see how things net out. 

The problem is that this really isn't a very good way to evaluate players.  It misses 98 percent of what they do on the field.  And it's prone to serious perception biases, like how Considine can miss a tackle and he's a bum, whereas Dawk misses the same tackle and "boy that guy made a great move."

Considine's problem is (mostly) that he's missed some very visible tackles in some very embarrassing ways.  We may not be able to explain any of Jim Johnson's zanier zone coverage schemes, but we're pretty damn certain that the whole point of being on defense is bringing down the guy carrying the football.

In Mikell's case, we don't see the same kind of thing, because tackling isn't his problem.  He's great at that.  But I know there were times when I was re-watching games last year when people caught balls on that secondary and I thought -- although you can never be sure -- "that sure seemed like Mikell's responsibility there." 

- - - - - -

And that right there is my concern about Mikell in a nutshell.  We saw just two years ago with Michael Lewis what can happen when a safety who is very good at playing near the line of scrimmage starts getting picked on by teams further down the field.  You need a whole lot of tackles for three-yard losses to make up for one blown coverage that leads to a 60-yard touchdown. 

Right now, Dallas is going through the exact same thing with Roy Williams, a multiple Pro Bowler who admits that sometimes he hopes the quarterback doesn't throw the ball his way.

Now maybe I'm wrong to worry about that.  Maybe Mikell will be even better this year than he was last.  But I wouldn't bet against there being some shuffling again this year at that very important position. 

Might even be time to dust off that old Considine jersey.

You don't? 

Comments

[Cornelius]

Why the hell do you keep bringing up the fact that Mikell was an undrafted player? That's irrelevant! His 40 time coming out of college is also irrelevant. He proved himself on the field last season. He was more than adequate. He played well according to Andy Reid who breaks down game film. There's no reason not to be confident in Mikell.

[Gabe from BountyBowl]

The actual problem we all face in evaluating safeties is that you can't see them on television. That is, the way the game is photographed for broadcast ignores most of what's actually happening on a passing play. In person/ at the stadium, you can obviously see more (but you can't DVR!).

When is the NFL going to get around to offering the coaches tape to the lunatic fringe of its fanbase? I'll take an online-only subscription. That'd be fine.

[Me]

I brought up the fact that he's an undrafted player to explain part of the reason I like him.

I mention his 40 time because it speaks to his ability to play in space. There are no questions about his ability to attack the line of scrimmage. There are still questions about his ability to defend the deep outsides.

Michael Lewis actually made the Pro Bowl in 2004, which is one heck of a proof point. And yet two years later he got benched for exactly this reason -- he was a target in space.

Maybe Mikell will be fine, maybe he won't. I hope in the end you're proven right, but the jury's still out.

[Andrew2]

So here's where we are:
1) Stats don't give us an adequate snapshot.
2) TV doesn't give us adequate coverage.
3) Andy Reid (understandably) never gives us an honest appraisal of players.

So maybe we can't figure this one out.

One side note: The secondary took a step backward last year, under the coaching of John Harbaugh. Now I read that McDermott is moving back to coach the secondary. Since the secondary played better when McDermott was in charge there previously, and since the linebackers played well under McDermott last year, maybe that's a sign of hope.

Or am I over-valuing the role of position coaches?

[Me]

Considering all the injuries last year, it's a little tough to put that on the coach.

We can figure it out, it just takes more than a handful of games. We'll know by the end of this year what the story is.