It’s the bye week, so it’s a time to talk about both what has happened before now and how we would like to see things happen from here on out. Often (especially when the bye week falls roughly in the midway point of the season) we use this time to discuss what we think the Chiefs should do differently.
I’ve already written about what the Chiefs did differently with the run defense against the Cowboys here (brought in KPL on nickel sets, more bear fronts, more Bennie Logan in nickel, more gap attacking) and I believe they should continue to go the same direction moving forward if they want to at least have an average run defense. I also wrote about a change I’d make on offense (give the ball to Tyreek Hill more) here. Now I want to write about another change I’d make moving forward on defense.
Terrance Mitchell needs to get his job back over Kenneth Acker. Like, now.
This is not likely going to be a controversial opinion. Look, I get why Mitchell was benched after the Raiders contest. He’d been getting targeted a TON over the first half of the season and had given up more receiving yards than anyone else in the league (of course, part of that is simply a function of being opposite Marcus Peters and not having as good of safety help as he had last year, but I digress). Andy Reid even implied that this might be temporary and was a way to motivate Mitchell. All that makes sense.
But Acker is a step backwards for the defense, and it shows.
Oof. pic.twitter.com/t60FjyvgBT— Seth Keysor (@RealMNchiefsfan) November 15, 2017
The eye test has told me since Acker came in that he’s not as good in coverage as Mitchell, but such things can be unreliable. So I went back and tracked every coverage snap for Acker on all-22 against Dallas. Then I did the same for what you all voted on as Mitchell’s worst game (vs Oakland) to see how Acker compared to Mitchell at his very worst (we already that Mitchell at his best is a solid corner, which we observed last season down the stretch).
If you need a primer on how I grade CBs, click here. Basically, I track every snap as a success/failure/neutral play in coverage, targets, catches/yards allowed, and passes defensed. By looking at every snap we get a MUCH clearer picture as to how well a corner is playing, rather than just looking at snaps he was targeted. or catches allowed (which are affected by the QB/WR play). Here’s Acker against Dallas (keep in mind that Steven Nelson reclaimed his spot as the No. 2 corner then moved inside on nickel snaps, so Acker’s snap count was a bit down this game).
Without commenting on whether those numbers are good or bad (hint: having more fails than successes as a corner is not good), let’s look at the game that got Terrance Mitchell benched (vs Oakland).
So what jumps out? Well, I gotta say in Acker’s defense, the first thing I think most people will notice is that Mitchell gave up 134 yards against the Raiders. That’s an absurdly high number, make no mistake. A lot of that was on a pair of deep passes to Amari Cooper (38 and 39 yards), one of which was a touchdown and the other of which was on the final drive of the game (which gave Oakland a chance at the comeback). That’s obviously not good, and again, I can see why Mitchell was given a game to sit out and regroup (as he’d had some rough games previously as well).
However, I’m a big picture guy. And in the big picture, the fact that Mitchell, despite being targeted a ton and struggling giving up catches (it was mostly Amari Cooper who gave him trouble. Crabtree had to commit OPI on every single catch he had against Mitchell. It’s remarkable what refs allow him to do), was MUCH more successful overall in coverage than Acker.
If we’re going by percentages, Mitchell failed in coverage on 26.8% of his snaps. That’s not ideal at all, to be sure, and definitely reflects that he didn’t have a good game against Oakland. However, when you compare that to Acker (55.6% failure rate in coverage), it suddenly looks like an All-Pro effort.
Acker appears to be a high-effort guy, but he guess wrong on a lot of routes, seems to lack quickness, takes a lot of false steps (and even hops), and doesn’t press particularly well. He does well when he anticipates the route correctly, but doesn’t seem to be able to adapt when he does not.
You know how people talk about CB's having smooth hips? This is the opposite of that. pic.twitter.com/fYRJRc8qEj— Seth Keysor (@RealMNchiefsfan) November 15, 2017
One thing Acker DOES do well is cover go routes, because he has good overall speed and uses the sideline well (note that on this particular snap he flips his hips to run with the receiver better).
It wasn't all bad for Acker. he does a nice job on go routes by and large, uses the sideline well and has decent speed. pic.twitter.com/QnUCp1mmMB— Seth Keysor (@RealMNchiefsfan) November 15, 2017
Beyond that, I can’t think of a single thing I saw in coverage that I believe Acker has the edge over Mitchell.
And to be clear, this isn’t to say that Mitchell is some kind of star. He put together a string of exceptional games last season that made me believe he could be a very good corner. However, this year he’s struggled with penalties a great deal, and has given up a very high number of catches and yards.
However, I think it’s a mistake to just stop at targets/catches/yards when it comes to Mitchell.
Mitchell has flaws (I'll write about those), but he's good at harassing receivers. Physical, quick, contests the ball... he's a feisty #3 CB in an ideal world. pic.twitter.com/Dn0SJviD7q— Seth Keysor (@RealMNchiefsfan) November 16, 2017
Mitchell’s catches/yards given up is in part a reflection of the fact that he’s been targeted a TON opposite Marcus Peters. Unless the Chiefs find another All-Pro caliber corner to place opposite Peters, whoever is in that spot is going to give up an unusually high number of catches and yards. That’s the nature of the beast. When you get targeted 12 times in a game where you’re constantly defending Michael Crabtree and Amari Cooper, some bad things are going to happen. And that’s the type of situation Mitchell has been in all year.
As far as film goes, Mitchell is more physical in coverage than Acker, he’s much quicker to mirror routes the majority of the time, has superior footwork, and possesses smoother hips. He also is much better about staying “square” with the receiver he’s covering (facing forward ready to break in any direction), as opposed to Acker, who consistently seems to be facing left or right (and thus limiting his ability to change direction).
One area Mitchell is exceptional in coverage is defending short comebacks/curls and routes of that nature.
Mitchell is far from perfect, but he looks WAY more comfortable in coverage than Acker. Much quicker feet, smoother hips, movement doesn't seem so tortured. pic.twitter.com/w3LO4NKhd1— Seth Keysor (@RealMNchiefsfan) November 16, 2017
There’s real value in being good against those routes. And he doesn’t necessarily give up his ability to defend go routes to do so, as he defends them about as well as Acker does (he’s had much worse luck with receivers making great plays, which is a huge difference in their “targets/catches” stat).
The differences between Acker and Mitchell are that Mitchell is in position more often, makes receivers work harder to get open with superior physicality and quickness, and is better at contesting the ball. He’s also a more physical tackler/hitter than Acker.
Again, Mitchell isn’t perfect. He has a tendency to get caught guessing wrong, which is where a lot of his failures in coverage come from.
As far as I can tell, Mitchell gets in the most trouble by guessing wrong or just failing to read/react to comebacks/outs/ins. Gets run out of plays that way. pic.twitter.com/spzu1pLJCP— Seth Keysor (@RealMNchiefsfan) November 16, 2017
But at the end of the day, he’s just BETTER than Acker. After reviewing the film, I don’t think it’s particularly close.
Now, there are some people who are going to stay stuck on targets/catches/yards allowed by Mitchell and insist that by that metric alone Acker should stay on the field. But frankly, using only snaps when receivers are targeted is a bad way to figure out how they did, as is just counting catches allowed. It doesn’t take into account a QB missing a wide open receiver, a receiver dropping the ball, OR the receiver making a great play. Give me a CB giving up a contested catch where he’s in position all day over a snap like this one:
This snap is a good example of why simply tracking targets/receptions doesn't tell the whole story. Acker gets beat here, but throw is way late and allows him to recover. Good on him for contesting though. pic.twitter.com/ZGBiuDSOsk— Seth Keysor (@RealMNchiefsfan) November 15, 2017
If you treat targets/catches/yards as your gospel, that snap was a win for Acker. But the reality is he was beaten and had bad technique on the play, but was bailed out by a late throw. It goes on the stats sheet as a good play though.
For me, what’s much more important is snap-by-snap consistency. And while Mitchell is definitely not a great corner, he offers a great deal more consistency in coverage than Acker. If I’m Bob Sutton, I’m going back to the Peters/Nelson/Mitchell combination that saw so much success late last season. Mitchell has had a few weeks to clear his head and the coaches have had a few weeks to examine their scheme. Now is the time to change things back. It won’t be perfect (or maybe even good), but it’ll be better.