Hokepoints Seeks HSPs in Tackles, Uselessly

Submitted by Seth on July 22nd, 2014 at 4:06 PM


Last year we predicted Dymonte would seize the nickel job. It's still open. [Fuller]

While doing Draftageddon this year Brian told each of us to draft an extra nickelback, or hybrid space player, because these defenses face 3-wide sets (and beyond) about as often as 2-wide ones. Modern offenses were made to take advantage of the run-stopping linebackers teams put on the strong side of the formation, forcing them to cover a jitterbug in space in addition to the running game. Defenses have countered with linebacker-safety hybrids of various forms. We've seen it in practice, and for many schools there's now an official hybrid/nickel position: STAR at Illinois and Ohio State, F-linebacker at Wisconsin, nickel at Michigan, etc. The HSP is the defense's answer to the slot receiver. It's the position we've ticketed Peppers for this season, at least to start.

However as I keep trying to find evidence of this in the stats, they keep eluding me. See: the division of Michigan's tackles (counting assists as 0.5 tackles) between the levels since 1995:

Michigan def stats to 1995

(click does the big thing)

Seen together:

Michigan def stats to 1995

Outliers. Doing this with just one team means we don't get much of a sample; unfortunately cfbstats just got bought out by two dudes who want $7500/year from each blog to use the stats he used to put online for free (i.e. under creative commons). If you downloaded the old spreadsheets from Marty (who does deserve to get paid for the work he did curating them) he says it's fine to use them. If you visit the old site you'll get the most salesman guy in the world who acts all cagey before telling you the $7500 price tag. Such is life.

We do still have the Bentley Library's stats, with the positions input manually. I put them on a Google Doc if you want them.

You'll note some years Michigan went to a 3-4 defense in 2004-'05 there's an uptick in linebacker tackles—that's Woodley being counted as one (for much of 1999 and 2000 they were a 3-4 but as often as not James Hall/Shantee Orr had their hands down, i.e. 4-3 under). And in 2009 and 2010 when Michigan went to a 3-3-5 (effectively three safeties) there's a safety hump. However the year with the largest % of tackles by DBs was 2011, the year Kovacs and T.Gordon were #s 2 and 3 on the tackle charts. Then it went down.

I think there's a couple things going on here. One, I think the transition to nickel happened longer ago than we gave it credit for. And two: Jake Ryan. Remember for the start of 2011 T.Gordon was playing nickel while Woolfolk was free safety. The typical configuration of Michigan's defense wasn't the 4-3 under we'd been told was coming; it was the same base nickel Michigan had before Rich Rod. The 2011 season's formations from the UFR:

Opponent Nickel   4-3  Heavy Okie 3-3-5 Other
Western Michigan 69% 5% 7% 11% 8% 0%
Notre Dame 52% 26% 6% 11% 4% 1%
Eastern Michigan 17% 54% 21% 4% 0% 4%
San Diego State 43% 45% 6% 6% 0% 0%
Minnesota 42% 50% 6% 3% 0% 0%
Northwestern 80% 15% 2% 0% 3% 0%
Michigan State 32% 55% 10% 2% 2% 0%
Purdue 35% 59% 4% 0% 2% 0%
Iowa 16% 63% 16% 5% 0% 0%
Illinois 51% 25% 6% 14% 1% 3%
Nebraska 35% 38% 8% 15% 0% 5%
Ohio State 23% 57% 8% 12% 0% 0%
Season 43% 39% 8% 7% 2% 1%

It was highly opponent-dependent. You'll note the trajectory of the Okie as they debuted it, shelved it, then brought it back against Illinois. But you'll also see Mattison deploying 4-3 alignments more often against spread outfits. Against Kain Colter and Northwestern's spread-option offense they were 80% nickel; against Braxton Miller and Ohio State's they were 23%.

I think what they discovered was they could get away with Jake Ryan as the HSP. Come 2012 and 2013 that was the base.

When did it get Nickel-y?

From personal recollection Lloyd used a lot of 3-3-5 nickel against spread teams after 2000, when his 4-2-5 nickel got shredded by Randy Walker's Rodriguezian offense. The tackling stats don't say. Even when I went through to identify who played "SAM" (Spur in 3-3-5, not the WDE in a 3-4) and nickel the tackle totals told no story:

Michigan def stats to 1995-sams

I'm giving up on this route. Eventually someone will find something useful to do with tackling stats but this isn't that day. If someone has an idea for how to find the rise of the nickelback in statistics, I'm all ears. In the meantime let's watch defense porn.



July 22nd, 2014 at 4:51 PM ^

It seems like you've done this some, but calling out who is the "HSP" and what they're categorized as each year might show a bit more of what you're looking for. Some years it's a safety, some a CB, some a small LB. I think your main point might hold though, the rise of the HSP happened earlier than noted.


July 22nd, 2014 at 4:59 PM ^

more often cause the opposing offense to run more plays to avoid the strength of that particular defense, thus resulting in more tackles my LB's, and DL's.

Essentially washing out any gained tackles by the DB's due to fewer opportunities?

Ron Utah

July 22nd, 2014 at 5:49 PM ^

One of the things I love most about Seth is that when the data doesn't support his hypothesis, he's willing to admit it.  Most people just change the data or make-up stats to prove their point.

This is a great piece, and I really enjoyed seeing the data, which I must admit I thought would be different as well.

Now the coach in me wonders if, upon seeing that extra DB, more teams audible to inside runs or plays away from the nickel, effectively giving more tackles to the LBs because the offense is attacking the space where there is no DB.  Even so, you'd think having the extra player on the field more would have more of an impact.

I do think JMFR was used as a nickel quite a bit, but not last year, and while the CBs did pick-up a few more tackles, they mostly took them from the safeties.  I don't know, man, but it was cool to see the data and think about it.  Makes me want to go back and watch games.


July 22nd, 2014 at 6:05 PM ^

I wonder if there would be any correlation if you added average yards per play to the mix.  There should be a difference in yardage with a hybrid player playing closer to the line and theoretically blitzing more versus a Safety or DB playing a deep third for example. 


I don't know...maybe too difficult to add another variable.


July 22nd, 2014 at 6:15 PM ^

You point out how 3-3-5 was used by Lloyd against spread option/power spread teams and a 4-2-5 against passing teams so wouldn't the schedule and opponent's offensive strategy make that different from season to season?


July 22nd, 2014 at 10:07 PM ^

Man all that pressure from the front 4 and the perfectly read plays from LB and S in that video are making me drool.

Funny seeing Mattison coaching for ND too. lol


July 22nd, 2014 at 10:31 PM ^

I was surprised that it was the 1997 team that won the NC and not some of the others. Boy do i miss that defense though. Looking at the video just makes me nostalgic.I keep losing the feed though and i am bummed.


July 23rd, 2014 at 11:30 AM ^

I remember those defenses were characterized by multiple guys flying to the ball and hard hitting, something we've only seen here and there over the past 5-6 years. I have to see them get back to that style this season or I think my head will explode.


July 22nd, 2014 at 11:15 PM ^

The amount of work put into some of the articles is what really makes this place what it is. Not just a bunch of homers spouting off about their school being the best. I've learned more about football (and I played in high school) in the last two years reading here than I have the rest of my life. Not that it's a lot...but, just sayin'.... The core group of writers do a fantastic job on this blog. Keep up the great work.


July 23rd, 2014 at 11:44 AM ^

for the defense in 1997, but what made him especially effective was the fact that our defensive line had three guys who were talented enough to play in the NFL—Josh Williams, James Hall, and Glen Steele—and a fourth—Rob Renes—who was All-American his senior year. They were regularly crushing the pocket up the middle, which made bltizes from the outside particularly devastating. If they weren't sacking the QB outright, they were hurrying the throw, and Woodson (and the other DBs and LBs) were there to pick off or knock down anything that wasn't perfectly placed.