Hokepoints: Triangle of Doom

Submitted by Seth on October 29th, 2013 at 9:48 AM


Brian mentioned a "triangle" route during the offensive UFR last week. I wanted to highlight that play, not just because I don't remember if we've covered what that means, but because it illustrates how the emergence of Funchess as a complementary deep threat to Gallon makes defenses do unsound (read: relatively easy to exploit) things.


A triangle is another passing concept for picking on zone, and works pretty well against different coverages. We've covered smash/high-low (which attacks the top and bottom of a zone simultaneously) and 4-verts (which gets guys to either side of a zone). A triangle is actually putting dudes to three sides of a zone. It's a slightly more complicated version of high-low, the triple-option version of it if you will.

Here's a play from the Indiana UFR. I want you to watch the bottom of the screen and how the Funchess, Gallon, and Toussaint routes work together:

Ln Dn Ds O Form RB TE WR D Form Type Play Player Yards
O24 2 8 Shotgun 4-wide tight 1 1 3 4-3 over Pass TE out Butt Inc
IME Gardner blows this read presnap by going to the short side of the field. He's got three guys on the wide side and Toussaint's flare is going to be open as M runs a triangle there. Instead he's looking at the wheel/out combo M is running a lot this year. Both are covered; Gardner tries to fit a ball into Butt in a tiny window, LB breaks it up. (BR, 1, protection 1/1)

The left side is a high-low. The right side of the play, even though Gardner didn't go there, is the triangle we're looking at.


I don't know if Gallon's route was a deep curl or an out or flag or whatever; it doesn't matter. I drew curl just to make the picture smaller.

In this case Gallon and Toussaint are smashing the corner's Cover 2 zone, while Funchess's curl route keeps the SAM linebacker from interfering with that. The cornerback went with Gallon (from alignment it was obvious he was more worried about the deep part of his zone than anything underneath), and that left the SAM to deal with Funchess and Toussaint. He chose Funchess, and Toussaint's flare route was open.


This is the odd triangle that doesn't use a snag route. Here's Indiana using the same thing against us in 2010:


Evidence to the "Borges runs the other team's playbook" theory

Funchess's route did what the outside receiver's snag route above was meant to accomplish: keep that SAM away from the smash read.

The cool thing about a triangle route is it can work no matter which coverage is thrown out there. Smart Football:

The weakness of the triangle stretch is that it’s typically only possible to only get a two-man horizontal or vertical stretch, whereas with a true “flood” you can place three (or more) receivers across the field on a given plane to truly defeat a defense. This limitation means that a triangle can be throttled by certain coverages that rotate to the triangle side.

But all this is counterbalanced by the triangle’s versatility: the route concept should result in a completion against almost any coverage, and, as will be shown further below, triangle stretches are also usually conducive to having a man-beating concept within them. And if the defense does roll to the triangle side, some kind of backside combination of passes or runs can be used to keep the defense honest.

What you complicate in making a double-read you gain back by not necessarily needing to know what the defense intends to throw at you:


Gallons of Fun Chess™

What I like about this is it takes advantage of the Funchess and Gallon scare factors. The middle linebacker see a giant terror beast and a guy en route to a record-breaking receiving day both going straight upfield and have to respect that this could become a verts situation.

If, for example, it was Jeremy Jackson and Jake Butt releasing to that side, the coverage to that side should be more confident they can stay with anything that goes over the top, and thus can be more aggressive. The big thing making anything to the left side tough is the defense has numbers (to that in a second), but another is that the coverage to that side can play more aggressively without fear of getting burned by Michigan's danger men. The corner can make sure Butt doesn't do anything in his zone, then shuffles leisurely sideways to get into Jackson's hip pocket. Reading this, Gardner tried to get it to Butt on the abandoned inside of that cornerback's zone, however the WILL could be confident enough in safety help over the top to set his feet in the path of a seam, then ride Butt's hip and watch the quarterback's eyes. This got him a great read on the pass in time to step in front of Butt—if this pass was too far inside or the WILL was a little bit faster it's a pick.

Compare the alignment of the free safety and boundary corner to that of the strong safety and field corner:


Left side: Corner is five yards off the LOS and safety is sitting at the first down marker (eight yards off). Their feet are planted as they read the play. No fear; result is little room for the offense to operate.

Right side: Corner is eight yards off and safety is ten; at the snap they're both backing up. Much fear; result is a lot of space for the offense to operate.

It's not just the running game that you open up when the defensive backs stay away from the box! On the side with Gallon and Funchess the coverage is giving themselves a big cushion to help keep those dudes in front of them. This isn't sound. That safety has to know the middle linebacker isn't going to be able to help much since he has to stay in to respect a Gardner run. That means once Gallon clears out the corner there's only a linebacker to deal with any short routes to that side.

Indiana was probably hoping to entice Michigan into quick hitches to Gallon, hoping Gardner's throw would be inaccurate and the corner could run down and stop it for minimal gain—put Michigan in 3rd and 2 and maybe they'll try to power run into a stacked box or something. Borges exploited this mercilessly with triangles and floods.

This Was a Bad Pre-Snap Read

The smash on the backside of the play was there in case the defense rolled their coverage to the frontside, for example if the free safety was centered over Gardner. He wasn't:


It's not just the majority of Michigan's talent advantage hanging out to the field side; because Toussaint was lined up over there, Gardner should have recognized he had numbers to the right.

Will It Work Against Michigan State?

Well MSU will tend to play their cornerbacks tight and try to prevent the route from ever developing. But yeah, the curl route is a check against their beloved Double-A gap blitz. When not sending the house the Spartans' favorite coverage is quarters, the thing that's all the rage in NFL defenses. Triangles are the counter du jour for that stuff.


Depending on things. The cornerback can latch onto Gallon and keep him from getting his route deep, shrinking the triangle. If the safety sees Funchess stop, that's his cue to attack, and by the time the ball arrives, is caught, and the back has time to turn upfield, the safety may have closed much of that distance. Finally the SAM may see Toussaint leaking out of the backfield and decide to take away the flat, which opens up Funchess's curl but with a good MLB in the next zone over and the safety coming down on the curl, it'll be a tight window. So that's the downside of triangling: you're unlikely to get much more than a short pass unless you can wait long enough for the Z receiver's route to develop (not likely given our OL situation), so even executed well it's just a shot gain machine.


Space Coyote

October 29th, 2013 at 10:25 AM ^

You show two different ways of forming the triangle (Indiana offensive play and Michigan's offensive play). What you'll come to find is that there are a ton of ways to form the triangle, including two short and one deeper or two deeper and one short. The goal is the same though, stress the defense in and out, typically to take advantage of single high looks, and short and long to take advantage of two high looks. At the end of the day, it's a fairly simple read for the QB because all the receivers are in a fairly tight target window, but getting through the coverage progression can be somewhat difficult.

Now, as for the play Michigan ran against Indiana. This is the Y-stick concept that Michigan has been running all year. It's a little different look because they are using a WR on the swing route, but it's the same exact thing. They likely tried to do that because using the RB to a single TE side has been a tip for defenses the past few weeks, so they switched up the look (the RB and TE aligned alone on a single side also helped Michigan pick up yards on the inverted veer against PSU, as the backside defenders bugged out to cover a route concept that wasn't forthcoming). This is one of Michigan's base plays (Hoke talked about 4-base runs and 4-base pass plays, this is one of Michigan's base pass plays). Now, typically one side of the field is for two high coverage and the other side is for single high coverage. You also typically have a hot read. But to date, I've only seen Gardner throw to the backside (the triangle side) once this year, and that was against ND. The point is, that this is a bad read in the sense that I don't think Gardner is making a read on which side of the field to throw to here. They're running routes in case DG comes off the Y-stick concept, and for the sake of occupying the entire defense, but when I've watched this play DG rarely gets a real safety read and goes straight to his key defender.

Now, there are a couple reasons this is probably happening. For one, they are trying to simplify DG's read. This is a simple read on throws that Gardner is very comfortable making. It's trying to get him in a rhythm and feeling comfortable making a read and throwing the football. The second reason is that it helps him get the ball out quickly. This is the same play that had the strip sack against PSU because Gardner didn't make the read quickly enough. Asking him to read the safety, to pick a side of the field, will cause more of that. At the time, I said the strip sack was on the OL and Gardner, because Gardner knows with this play he has to make a quick decision and then run or throw it away. He didn't there and wasn't helped by his backside OL, but there is a reason they've been making a single read here and getting the ball out: to help Gardner and the OL.

Anyway, I'm a big fan of triangle concepts. I know Borges has quite a bit of them in the playbook, but I haven't seen as many this year because I think they're still working on DG's progressions and, combined with that, the ball needs to come out fairly quickly and on time, especially with this OL. If Dileo plays against MSU though, expect to see quite a bit of them with Dileo taking the inside and low portion of the triangle more often than not, which will allow him to work against LBs in that short area of the field.

Space Coyote

October 29th, 2013 at 10:53 AM ^

Typically, the SAM here will pass off the #2 and head to the RB. This has, at times, been a weakness for MSU with the LB trying to play the RB in the flat, and has been a source of good gains for teams this year relative to other things that MSU has faced. It also likely puts Dileo (if healthy) on a MIKE, which is a good match-up. If Michigan goes to it several times, expect MSU to move down their safety a bit more into coverage (similar to what they did against ND). This opens up the outside route, and allows for the possibility of the outside and high receiver to run not just streaks/fades, but also posts/digs to get open, making the coverage on him much more difficult.


October 29th, 2013 at 11:43 AM ^

Given that MSU likes to blitz their linebackers and scout Michigan mercilessly, I totally expect this.  They'll have a pre-snap read and if they read run they'll blitz the gaps; if they read pass, the SLB will head for Toussaint.  It's not just the triangle; this also covers the throwback screen Michigan typically gets 20 yards out of when defenses don't expect it.  This is perhaps why Dileo is more effective against MSU than in other games?  They give the MLB too much to do, and when you can force the reads to mean the MLB has to be in two places at once, either there's a gap for the RB or Dileo's open -- usually the latter, as they're dead set on shutting down the run and seem to blitz when in doubt.  (Hell, I would -- they make plays by being aggressive.)

The key thing for MSU will be to exploit Borges' tendency to tip his runs.  Even with a bye week for Michigan to roll out something new, MSU WILL find their pre-snap read.  They'll chain an army of interns to computers and make them watch footage sweatshop-style if that's what it takes.  If Borges is clever he'll basically use his pre-MSU games as a base to anticipate MSU's read and exploit it, but I won't hold my breath.  So this will come down to whether or not Gardner can make his quick read and get rid of the ball before MSU's front seven collapses the O-line.

Space Coyote

October 29th, 2013 at 11:53 AM ^

Because MSU's two primary defenses: cover 4 and cover 3 (which is how they typically cover behind the blitz or in and long situations) both match up the #2 receiver against LBs in short coverage.

MSU will still blitz the interior because of Michigan's youth there. If Gardner can step into throws though and make the correct reads, they will quickly back off that and favor more of a CB blitz package to get off the edge and force Gardner to stay in the pocket. They wouldn't have done this with Denard because of Denard's weakness as a passer and vision.

So the key to backing them off is for DG to make his quick reads and step into his throws enough to make the passes. He's been fundamentally better at this the past few games (people that say he's regressed are way off base, DG's QB fundamentals have come a long way in-season, which is fairly impressive), it's important that those improvements show themself against MSU and that he doesn't revert to previous mechanics under the heavy pressure and after getting hit a few times. He will get hit.


October 29th, 2013 at 12:25 PM ^

These are almost exactly my thoughts. Gardner has progressed quite nicely in his technique, and probably in his reads as well. He was great against Indiana and very good against PSU (maddening pick to the DE notwithstanding). We'll need him to continue his good form in the face of a defense that will get pressure to him.

I think a great play for us could be a TE stick to a 2 TE side. Butt on the stick, Funchess on a flag, Fitz to the flats. We've run plays from this personnel all season, so it wouldn't be a tip off. This would theoretically force them to keep their backers in coverage and could get you into a deepish Funchess vs SAM matchup.

Ron Utah

October 29th, 2013 at 3:35 PM ^

Because of the exotic nature of MSU's blitzes--usually coming from the A-gap or the CBs--I'd rather have my extra help in the backfield.  If DG starts in the shotgun and has a back on either side of him (Butt and Fitz) then those guys can read the inside blitz or the CB blitz.  An in-line TE might not be able to help with either one.

Of course, the downside of this is that backfield blocking is not typically as effective as LOS blocking--those backs probably aren't going to hold their blocks very long and the pocket will be constricted.  But that risk is tempered by the fact that now we've got three guys running routes--Gallon, Funchess, and Chesson/Dileo/Jackson/Reynolds--that can make the defense pay.  If the blitz doesn't come, the backs can leak out to force the LB's to respect them.

I'm not sure if this would work, but it's a concept that I think makes sense against MSU and would allow our backs to keep their LBs off balance with blocking and route-running.  I'll take Fitz vs. any of their LBs, and I'll certainly take Gallon and Funchess one-on-one against any of their DBs.  We just have to keep DG on his feet!


October 29th, 2013 at 11:31 AM ^

On the play pictured, Gardner releases the ball in no more than 2 seconds post-snap.  In that brief time he doesn't look to his right at all.  I think that's what he was instructed to do, pick a side and stay with it, lest he die in a pocket protected only by the OL, with five players in routes.

Contrast that to a play against PSU when he was sacked.  Lewan and Schofield are right, with AJ Williams in the LT position.  AJ is a statue as a DE rips past him and reaches Gardner in about 2 seconds.  Gardner holds the ball too long and is blindsided.

Are there not "designed", mandated quick-release situations?  Shouldn't this latter play have been one of them?


EDit: sorry, I didn't see that Space Coyote actually answered this.


October 29th, 2013 at 11:50 AM ^

I think the SAM is going after Fitz.  He comes back and holds up after he sees Gardner going to the other side of the field.  The way he ends up turning back makes it seem like he was on Funchess. 

But if Gardner had opened to the field, I think it would be Funchess that comes free.


October 29th, 2013 at 12:33 PM ^

Actually seth, the first one is the TROJAN PLAY (using Rich Rods terms of the play) that Borges still runs 4-5 times a game. Its usually ran out of a form of trips but in this case the rb becones the 3rd wr. #1 playside WR runs a go, #2 runs an out and #3runs a choice route (5 yard stick or Out cut)... the Backside Wr runs a slant/settle route while the rb runs a wheel out of the backfield.

If there is a two-high safety look, the qb is reading the stong side combo, taking the go route vs cover 2 if he can, if not he chooses between the out or the stick routes.

If there is one-high safety the qb is going to look backside automatically. If the corner follows the slant/settle he will hit the wheel route to the RB

This is the wheel route we had success with from the backfield earlier this year. One time they even lined Funchess up back there to run the wheel