Tag: Robert Saleh

The 49ers were 7:13 away from being crowned world champions. Holding the Chiefs’ powerful offense to a measly 10 points through the first 53 minutes of Super Bowl 54, San Francisco was a stop on 3rd-and-forever away from putting the game out of reach for good. The NFL’s best pass defense, with a pass rush that had harassed Patrick Mahomes all night and a secondary that had picked him off twice, surely wasn’t going to be beaten in that situation.

Well, you know what happened next… Tyreek Hill found himself wide open for a 44-yard reception that sparked a 21-0 run for the Chiefs — and gave Kansas City its first Super Bowl title in five decades.

It wasn’t surprising that Mahomes, the NFL’s most talented quarterback, was able to connect with Hill, the league’s most explosive receiver, for a big play. It was, however, shocking that the 49ers secondary had allowed the Chiefs’ biggest weapon to get this open that far downfield.

How does that happen? Quite simply, Andy Reid called the perfect play for the coverage 49ers defensive coordinator Robert Saleh had called.

Perfect play-calls don’t happen by accident. They’re the result of intense film study, pattern recognition and, in this particular case, a naive defensive play-caller who thought he could get away with showing one of the NFL’s sharpest offensive minds the same coverages over and over again. That one play didn’t cost the 49ers a championship. But it was just one of many failures by Saleh to change the picture for Reid and Mahomes, which helped them lead the furious comeback.

San Francisco head coach Kyle Shanahan was always going to receive the bulk of the blame for the blown lead. His conservative decisions at the end of the first half and the beginning of the third quarter drew the ire of the nerds. The 49ers’ run-pass ratio in the fourth quarter had football guys steamed. Both groups missed the real culprit. Shanahan (and his quarterback) certainly deserves some blame, but most of it should be placed on his defensive coordinator.

How much blame does Shanahan deserve?

Before we get into Saleh’s missteps, let’s take a look at Shanahan’s.

It’s clear the 40-year-old coach handled the end of the first half poorly with his decision to (1) not use a timeout after the Chiefs were held on third down with 1:59 to go in the second quarter and; (2) start the ensuing drive out with two runs, essentially killing any chance the 49ers had of adding points before the half. But how much did those two decisions really cost his team? We can use the “Expected points” model to get an idea.

Let’s start with the timeout. Let’s just assume the result of the punt (a touchback) would have remained the same. After the third-down stop, the 49ers could have called timeout at the 1:47 mark. The punt took nine seconds off the clock, so we’ll give them the ball at the Chiefs’ 20-yard-line with 1:38 remaining and two timeouts to work with. According to the Expected points model, the 49ers would be expected to score about 0.8 points in that scenario. With 0:59 seconds on the clock and three timeouts, that number drops down to about 0.6, so the failure to call timeout cost San Francisco about 0.2 expected points. The 49ers eventually throwing downfield and going for points complicates matters, but Shanahan ultimately was content to go into the half with a 10-10 score, which cost him 0.8 expected points and a percentage point of win probability. It wasn’t the optimal decision, but it wasn’t a game-changing one either.

That wasn’t Shanahan’s only conservative decision, though. After the halftime break, the 49ers drove down to the Chiefs’ 24-yard-line, where they faced a fourth-and-2. Most analytics devotees were advocating for San Francisco’s offense to stay out on the field, but Shanahan opted for the three points, much to the chagrin of Analytics Twitter. It actually turns out that kicking the field goal was the best decision in that case, at least according to ESPN’s model…

Also in postmortem, FWIW, our model supported SF's FG on 4th and 2 from the 24. I was personally surprised. Can't know exactly the 'reason', but suspect it liked a 2nd-half lead, and the 24 is not as juicy a 4th dn region as longer/marginal FG ranges or nearer the goal line.

— Brian Burke (@bburkeESPN) February 3, 2020

That makes sense. On fourth-and-2 in that area of the field, NFL teams have a conversion rate of 57.7% since 2010. Teams average about five yards per play on those attempts. For the sake of argument, let’s just give the 49ers those five yards and put them on the Kansas City 19-yard-line. The Expected points in that situation is about 4.5. So, at most, Shanahan’s decision cost San Francisco 1.5 expected points, but that’s also assuming a 100% chance of conversion on fourth down.

Even when taking the most cynical view — at least a cynical view backed up by numbers — Shanahan only really cost his team about three points with those two heavily scrutinized decisions. That’s obviously not ideal, and game management has absolutely been an issue three years into his head coaching, but when factoring how good of a game he called, Shanahan did more good than harm. Thanks to his designs, Garoppolo finished the night with an expected completion percentage — which is based on a variety of factors including receiver separation and depth of target — of 69.8%, per Next Gen Stats. For context, Drew Brees led the NFL during the regular season at 68.0%. The running game averaged 0.23 Expected Points Added per attempt, which nearly doubles the Ravens’ league-leading mark of 0.12. That’s next level play-calling.

Of course, there are those people who believe that the loss falls on Shanahan not because of his suboptimal decision-making but because his decision to abandon the run after the 49ers took a 10-point lead in the fourth quarter. San Francisco ran 17 snaps in that final frame. Only four of them were runs.

That sounds bad, but when you add situational context, that ratio makes a lot more sense. Let’s start by pointing out that seven of the 49ers’ 13 dropbacks came after the two-minute warning with the team trailing. You cannot blame Shanahan for any of those calls as his team was in a must-pass situation.

That leaves six pass calls to four runs.

Before the two-minute warning, the 49ers ran on every one of their first-down plays in the fourth quarter. So, no problems there. The reverse was true on second down. Garoppolo dropped back on all three second-down snaps in the fourth quarter before the two-minute warning. The first of those dropbacks resulted in a 12-yard catch for George Kittle. The second play was actually changed at the line by Garoppolo, so you can’t pin that decision on Shanahan. So that just leaves the second-and-5 play call, which got Kittle. Unfortunately, the pass was batted down by Chris Jones.

The 49ers (wisely) called pass plays on their two third-down plays. The first was a third-and-14 where the protection broke down and Garoppolo was forced to scramble. On the next third-down play, Shanahan dialed up his “Arches” concept and got the look he wanted. But Garoppolo didn’t take the wide-open throw to Kittle and a miscommunication with his receiver led to a punt on fourth down.

So, really, you can only question one of Shanahan’s run-pass decisions, and even that play should have worked. The 49ers’ play-calling was good throughout the game. At least the offensive play-calling was good…

The 49ers’ defensive game plan

I’m an idiot. I just wanted to point that out before I get into the 49ers’ plan for stopping the Chiefs. Because you need to realize this: if an idiot like me recognized some of this stuff, I know a play-calling savant like Andy Reid certainly did.

So what was the 49ers’ plan? It won’t take long to explain, actually. On first down and second-and-short (under 7 yards to go), the 49ers essentially played two coverages: Quarters and Cover 3. And it wasn’t difficult to figure out when they’d play which coverage. When Mahomes was in shotgun, they played Quarters; when he went under center, they played Cover 3.

Every time.

We’re talking a 100% tendency for the entire game.

Here are all of Kansas City’s under-center snaps in the game…

It’s all Cover 3.

Now, the Quarters calls did change based on the pre-snap distribution of the receivers. Against formations with three receivers to one side and an isolated pass catcher to the other, the 49ers played a variation of Quarters some coaches refer to as “Solo.” In that coverage, the corner to the single-receiver side plays man coverage on that receiver and the linebacker to that side takes the running back. This allows the defense to flood its zone coverage to the three-receiver side to avoid being outnumbered.

 

Here’s an example of the 49ers playing “Solo” against the Chiefs…

The key man in “Solo” coverage is the backside safety, who is responsible for the No. 3 receiver (the receiver lined up furthest inside) if he goes vertical. The Chiefs had dominated defenses all season with deep crossing routes from three-by-one alignments, so Saleh playing this as a base coverage against three-by-one sets made a lot of sense.

But playing it every time the Chiefs got into one particular formation did not.

Especially down the stretch when Reid would have picked up on the tendency. Again, I’m an idiot and it didn’t take long for me to pick up on it, but Saleh decided it was a good idea to show Reid and Mahomes the same picture down after down. The Chiefs ran nine first-down plays from a three-by-one gun formation during the game. San Francisco played “Solo” on every single one of them.

When the Chiefs lined up in a two-by-two shotgun formation, the 49ers played Quarters 93% of the time. On third-and-3+, they played Cover 1 man (sometimes with a safety lurking over the middle, sometimes with an extra pass rusher) 80% of the time. On third-and-extra-long (more than 10 yards to go) Saleh would call Cover 3 Buzz, which is sorta, kinda similar to “Solo” coverage in that you have the backside safety looking to take the No. 3 receiver if he goes vertical…

Here’s an example from the game.

The Chiefs faced third-and-extra-long three times outside of the red zone. The 49ers called Cover 3 Buzz every single time. Converting on third-and-long is difficult, but it gets a lot easier when the offense knows what coverage it’s getting, as San Francisco would find out later in the game.

With all of these easily diagnosed tendencies, Saleh might as well have been calling plays into Patrick Mahomes’ headset.

Andy Reid adjusted … Robert Saleh did not

The obvious rebuttal is “It was working!”

The mighty Chiefs offense had scored only 10 points through 53 minutes, so why would Saleh change things up? But was it actually working? Consider this: in the first half, the Chiefs averaged 2.5 points per drive. They averaged a league-leading 2.7 points per drive during the regular season, so not much of a difference there. On their drives before the 21-point run, the Chiefs averaged 45.7 yards per possession. During the regular season, the Ravens led the league at 41.8 yards per drive.

Kansas City’s offense was moving the ball just fine, but two uncharacteristic interceptions thwarted promising drives in the third quarter. Mahomes wasn’t going to keep throwing it to the defense. Saleh had to at least throw some changeups at Reid and his young quarterback. He didn’t, and Reid took advantage.

Knowing he’d get Cover 3 if the Chiefs lined up under center (100% tendency), Reid dialed up “Y-Leak” and created an explosive play downfield.

Knowing the Chiefs were playing man on third-and-long (80% tendency), he called this pick play to get an easy first down completion.

Knowing he’d get man coverage on third-and-long within 25 yards of the end zone (100% tendency), he called slot fades for both Travis Kelce and Hill and let Mahomes pick a target based on the movement of the free safety.

Knowing he’d get either Solo or a Cover 3 Blitz on second-and-long (100% tendency) — and that either way Sherman would be locked onto Sammy Watkins with no safety help — Reid called for a fade route.

And that brings us back to the play that changed the game: Hill’s 44-yard catch on third-and-15. Thanks to NFL Films, we know the play Reid called, at the behest of Mahomes: “3 Jet Wasp Y-Funnel.”

And thanks to Saleh’s schematic rigidness, Reid knew the 49ers’ play-call: Cover 3 Buzz. So Reid used Kelce’s over route to occupy the strong safety, while Watkins’ dig route would draw the attention of CB Emmanuel Moseley. That left Jimmie Ward, playing the deep middle, to cover Hill all by himself. With Hill looking as if he were running a post route, Ward opens up his hips to run with it, only for Hill to break back toward the sideline, leaving him wide open.

There was no read there. Mahomes didn’t have to go through his progressions. He simply had to buy enough time for Hill to get open, because, based on everything he had seen that game, Mahomes knew Hill would get open. He said as much after the game:

I asked Patrick Mahomes about "2-3 Jet Chip Wasp” shortly after he left the field following Super Bowl LIV. pic.twitter.com/itaxQ3fbxR

— Pete Sweeney (@pgsween) February 6, 2020

“They were playing this kind of robber coverage all game long where the safety was coming down and kind of robbing all our deep cross routes, and we had a good play call on it where we had (Travis) Kelce do a little stutter deep cross. We had Tyreek getting one-on-one with that safety, but the biggest thing was we needed really good protection.”

There’s no shame in losing to Patrick Mahomes and Andry Reid. Robert Saleh fell victim to a duo that has left  many defenses in its wake. But the 49ers defense deserved more from its coach. Saleh had the league’s deepest and most talented defense at his disposal. He had two weeks to put together a game plan. And this is what he came up with? Mahomes and Reid do not need any extra help to make an opposing defense look silly, but Saleh gave it to them anyway.

Saleh’s simple approach to play-calling had served the 49ers well all season. With a stacked group of pass rushers, an athletic linebacker corps and a smart secondary, he had more than enough talent to just line up and beat opposing offenses even if they knew what was coming. Against Andy Reid and this offense, which was just as talented, that was a mistake. A far bigger mistake than any Shanahan made that night in Miami.

[jwplayer UO045H78-q2aasYxh]