Every Pass Tanner Morgan Threw in 2019

The Gophers' 2019 season was its best in more than half a century, and the biggest reason for that was their dangerous passing attack. Minnesota jumped from 40th in FBS in passing yards per attempt the year prior to 6th. Tyler Johnson and Rashod Bateman were both named First Team All-Big Ten receivers, with Johnson breaking a bundle of the program's single-season and career records and Bateman appearing on multiple All-America teams. Tanner Morgan had the most productive season ever for a Gophers quarterback, setting single-season program records for passing yards, touchdowns, and completion rate.

With a new season beginning in just weeks, it is a good time to reexamine and try to learn something from the passes Morgan threw in his breakout sophomore year.

Here, then, are all 318 of those passes.

Click on any image in this post to enlarge it.

The above scatter plot, inspired by those on the NFL's Next Gen Stats site, is a bit messy — not totally useless at a glance, but far too indigestible to be the basis for any analysis. But by bucketing and quantifying the points on the plot, we can examine Morgan's strengths and weaknesses and the different roles his receivers had in the offense, and hypothesize at how the passing attack might change in 2020.

First, some important points on the charts and data in this post:

• I plotted these throws myself; they weren't placed automatically with some tracking technology. Therefore, though I tried to be as precise as possible, the points' exact placement are not perfect. Someone else may view the location of a given pass slightly differently.
• At times, I disagreed with the spot of the ball given by officials or with what the official play-by-play read. In such cases, for the sake of consistency with the statistical record, I deferred to the official sources and placed points a yard or two from I would have spotted the ball. Those instances were few, and their effect on the overall data is negligible.
• In this post, I bucket passes based on how far beyond the line of scrimmage they are and use "short," "intermediate," and "deep" to define passes within specific ranges. A short pass is between 0 and 9 yards downfield; an intermediate pass is between 10 and 19 yards downfield, and a deep pass is at least 20 yards downfield.
• I charted touchdowns caught in the end zone based on where the ball was approximately caught rather than on their air yards. For example: A 9-yard touchdown caught roughly 6 yards into the end zone would appear on the chart as a 15-yard touchdown and would count as a pass between 10 and 19 yards, but Morgan would only be credited for 9 air yards. This is to accurately reflect the distance the ball traveled downfield.
• I placed deflected passes differently based on the effect of the deflection on the pass. In general, I attempted to chart the pass in the vicinity of its intended receiver if it still reached or almost reached him; if it did not, I placed the pass at the point where it was deflected. Multiple passes broken up by pass rushers count as incompletions at or behind the line of scrimmage. Additionally, if I could tell who the intended receiver was based on the initial direction of the throw, I counted the throw as a target of that receiver.
• Drops are not indicated differently from other incompletions.
• I did not filter out garbage time.

Because of many of the above points, it is important to not focus too much on the particulars of the data, but rather consider the lessons gained from larger trends. To that end, I have created charts to help glean those lessons from the data.

To begin, here are Morgan's 318 throws bucketed into 10 zones based on depth and the third of the field to which they were thrown (except on throws behind the line of scrimmage, which are counted as one zone).

As indicated by the scale to the left, the zones are colored according to how their completion rates compare to Morgan's total rate of 66.0 percent. The highest rates are the brightest green, the lowest rates are the brightest red, and the rates closest to Morgan's average are dark grey. Three notes on the methodology:

• The extremes of the scale are the points 15.2 points or more away from Morgan's average. (50.9 percent is not a typo but the result of rounding.) 15.2 points is the standard deviation of the 10 zones from their average completion rates. You can see each zone's z-scores here.
  
  
  
  
  
  Though the mean of the 10 zones' completion rates is slightly lower than 66.0 percent, when calculating z-scores, I prefer to use the total completion rate instead of the mean. The total completion rate, unlike the mean of the zones' completion rates, accounts for the number of passes thrown to each zone. I don't want to weigh deep passes as much as intermediate passes when Morgan threw far more intermediate passes.

• I would like to compare Morgan's passes to each zone against national averages. However, that is not doable without the data. While it is not ideal to compare Morgan's performance in a given zone to his performance in other zones, I believe it remains instructive to see to what areas of the field he is a better or worse passer.
  
• As mentioned above, some passes broken up at the line of scrimmage are charted as incompletions beyond the line of scrimmage. This is likely a major reason why Morgan's completion rate on short passes to the middle of the field is so much lower than his completion rate at the same depth in the outer thirds.

Back to the charts. One of the trends apparent from both the scatter plot and the grid is that Morgan was far more successful when throwing the ball to his right than he was to his left. This was true at all three levels beyond the line of scrimmage. Dividing the field into thirds, these are Morgan's numbers:

Removing throws less than 10 yards downfield paints a starker picture: Morgan completed 45.3 percent of intermediate or deep passes to his left, just two-thirds his rate on other passes of at least 10 yards.

Interestingly, however, Morgan threw the fewest touchdowns to the right side of the field and the most to his left. Another glance at the scatter plot above shows that Morgan's leftward throws were usually along the boundary. It seems that on deeper passes to this direction he almost exclusively threw to the sideline, as modeled on this touchdown pass to Johnson.

This partially explains how Morgan's completion rate on intermediate-to-deep passes to his left was so low while throwing just one interception that direction. It is likely that Morgan aimed to put the ball where either his receiver could catch it or nobody could. With the Gophers' talented receivers, the all-or-nothing strategy made a little more sense.

It's worth nothing that Morgan's sticking to the sideline is somewhat mirrored on the right side as well. When Morgan went deep, he seldom threw to the middle of the field. Completing seven of his eight attempts might signal that it could be worth trying that zone more often in 2020, but it's more likely that Morgan was being smartly selective. It's not common for nobody to be home in center field. All three of Morgan's touchdowns to that zone came off of play action.

The Gophers did work the middle of the field plenty in the short and intermediate game, however. A look at the below heatmap of Morgan's tendencies shows what anyone who watched him last year knew already: Minnesota's slant game was a major weapon.

The big red dot between roughly 8 and 16 yards downfield is almost entirely the footprint of the Gophers' slant game. It was often attached to play action or run-pass option, such as on this throw:

Often, to get more space for the receiver, since-departed offensive coordinator Kirk Ciarrocca would pair the slant with a bubble screen. The bubble is meant to attract the attention of the defender who would be in the space the slant is entering. When combined with play action, this concept can leave the middle of the field wide open, something Minnesota took advantage of to great effect at Purdue.

Catches on routes like the above slant were relatively uncommon for Chris Autman-Bell, however. Johnson and Bateman were Morgan's preferred targets over the middle between 10 and 19 yards, combining for 43 targets. (Autman-Bell had 3.) The two star wideouts were the team's most targeted receivers overall as well, but their specific usages differed. Below are the target rates of Minnesota's top four receivers last season, broken down by depth.

As you can see, Johnson was the most trusted receiver at the short and intermediate levels, as well as behind the line of scrimmage. Within 20 yards of the line of scrimmage, 40.7 percent of Morgan's throws were to him. Note, however, that this includes fly sweeps as passes behind the line of scrimmage. They were a small part of the offense and are practically run plays. See an example below.

Autman-Bell, a regular target on tunnel screens, was the second-most-used receiver in the backfield. He was the only Gopher to score as the receiver of a pass behind the line of scrimmage, doing so against Nebraska and Penn State.

Though Autman-Bell saw a handful of targets on deep throws, Minnesota clearly favored Johnson and Bateman on passes beyond 20 yards. Four-fifths of Morgan's deep passes were to those two, with Bateman leading the team with 27 such targets. The same was not true on shallower routes; only around a quarter of his targets were shorter than 10 yards downfield.

However, Johnson was the more efficient deep threat of the two. Though Bateman was the recipient of five deep touchdown passes to Johnson's four, Johnson's 76.5 percent catch rate on deep throws was considerably higher than Bateman's 51.9 percent. That said, they were both serious deep threats. Their effectiveness was a major reason why Morgan's completion rate on deep throws was an impressive 58.2 percent.

Further examination of target data can give us an idea of whether certain Gophers receivers were more or less reliable than should have been expected given the location of their targets. Using Morgan's completion rates per zone and each player's number of targets per zone, we can calculate an expected number of receptions per player. After dividing that number by total targets, if we subtract it from that player's actual catch rate, that will give us what in the table below is marked REC % oEXP — catch rate over expectation. A positive value means the player caught more passes than expected; a negative value means he caught fewer.

For most of the receivers, their rate statistics are rather meaningless due to small sample sizes. That said, those near the top of the table received enough targets for their stats to be at least somewhat informative. Minnesota's top three receivers all performed at roughly the level that was expected.

But if anyone stood out by this measure, it was Demetrius Douglas, whose catch rate was 5.9 points higher than should have been expected from a receiver getting Douglas' targets. Though that is not a dramatic difference, it is enough to suggest that Douglas perhaps was more reliable a pass-catcher than his reputation and workload suggest. Alternatively, it could reflect that defenses allowed him easier catches because they were more worried about the stars on the outside than they were the Gophers' slot receiver. (Half of Douglas' targets beyond the line of scrimmage were between the hashes, a mark no other top-four receiver matches.)

Though his catch rate seems impressive, Douglas was also the least productive of the four when he caught the ball, averaging 3.2 yards after catch. Meanwhile, Bateman led Morgan's most frequent targets with 6.2 yards after catch — despite that being the source of less than a third of his yardage.

The players most dependent on yards after catch, as is to be expected, typically got the ball close to the line of scrimmage or behind it. The teams' running backs, who received one pass beyond 10 yards, accumulated 82.8 percent of their yards after the catch. Among wideouts, Autman-Bell's yardage was the most dependent on what he got after the catch.

This was especially true of scoring plays specifically: Even after removing his outlier 71-yard score against Purdue, more of his yardage on touchdowns (40.3 percent) came after the catch than Johnson's or Bateman's did. (That said, the sample size is quite small.)

TD (YAC=0) % is the percentage of a player's total touchdowns that he caught
without gaining any yards after the catch — i.e., those he caught in the end zone.

Johnson caught seven of his touchdowns in the end zone, and another two he caught within 5 yards of the goal line. He also had the fewest yards after the catch per touchdown. Though Morgan occasionally went to Bateman when within striking distance of the goal line, Johnson was clearly his favorite red zone target. It will sting to be without Johnson's unique ability in the air this season.

Bateman also showed the ability to outleap defenders when needed, but he often succeeded by outmaneuvering and outrunning them. He frequently created separation between himself and cornerbacks before getting the ball, and once making the catch, he could keep them off his heels until reaching the end zone.

With Bateman granted eligibility for 2020, the Gophers know they have a No. 1 receiver and a top-notch deep threat. The two biggest questions are whether they can find both a second option to keep defenses from focusing too much on Bateman and whether someone can fill the roles Johnson had in the short-to-intermediate game and in the red zone. Though Minnesota didn't often spread the ball around, there are many candidates to replace at least pieces of Johnson's game — especially if new coordinators Mike Sanford Jr. and Matt Simon decide to split opportunities between more players.

Autman-Bell may not have Johnson's leap, but he has shown the speed to make plays once the ball is in his hands, and he flashed some potential as a deep option his freshman year.

His target profile suggests that with more opportunities, he could be the kind of (if not the same level of) well-rounded receiver Johnson was for Minnesota. He is a popular breakout candidate this season, and there is plenty of reason to believe he could live up to those expectations.

Tight ends were a bigger part of Sanford's offense at other stops. With his arrival to Minnesota, Jake Paulson, Brevyn Spann-Ford, and Bryce Witham may play a greater role, which could be useful near the goal line.

At wideout, there are a couple more jump-ball candidates in redshirt freshman Nnamdi Adim-Madumere and true freshman Jonathan Mann, listed at 6-foot-4-inches and 6-foot-3-inches respectively.

After Douglas' decision to step away from football, the options to take Johnson's slot snaps are inexperienced. Two new arrivals may feature prominently. Top signing Daniel Jackson and late addition Douglas Emilien could each see plenty of the field and be dangerous with the ball in their hands. Though Emilien's high school highlights show him primarily as an outside receiver, the size difference between him and the returning starters might mean he spends some time on the inside, perhaps moving between spots as needed.

Then, technically, there's Seth Green, who made one catch on one target in 2019. I wrote after the Wisconsin game about how Minnesota should have used Green more often, if for nothing else than to force defenses to not ignore him when he lined up as a receiver. Alternatively, they could have restricted his deployment to the "Green Line" package, rather than signaling that the Gophers were about to run the ball because he and Douglas were on the field instead of Bateman and Johnson. Either way, Green needs to be used differently.

Regardless of the specific receiver combination and how Sanford and Simon differ from Ciarrocca, it's unlikely Minnesota's offense is dramatically worse. SP+ currently projects the Gophers' offense to be the 11th-best in the country. Massey sees it as the 16th-best. With the turnover on defense, if the Gophers have a successful 2020, Morgan and company are likely to once again drive that success.

Thanks to Daniel House for assistance finding game footage and to George Roberson for feedback.

*    *    *

A postscript for the curious: Below are Morgan's game-by-game passing charts for 2019. The points on the full-season scatter plot came from these. A smattering of observations:

• Unsurprisingly, most of the games in which Morgan attempted the most passes and accumulated the most air yards were the games in which Minnesota spent a significant amount of time behind or within a score of the opponent. This included losses to Iowa and Wisconsin, the two games in which Morgan attempted the most passes and threw for the most air yards. The Gophers loved to kill off games with long, run-heavy possessions once they had the lead (even when that lead wasn't that safe), so this trend is no surprise.
• Morgan's 19 throws to the left against Wisconsin were by far his most in a game in 2019, accounting for just over half his total throws for the game. On those leftward throws, he completed just seven (36.8 percent) and threw an interception.
• Against Penn State, Morgan attempted six passes beyond 20 yards. That accounted for 30 percent of his throws in that game, a mark he did not match in any other game.
  In the same game, he attempted six passes behind the line of scrimmage, which again accounted for 30 percent of his throws. He did not match that mark in any other game.
• Minnesota's win over Fresno State was the game in which Morgan attempted the fewest deep passes: one, Autman-Bell's 20-yard touchdown to tie the game.
• Minnesota's win over Northwestern was the only game in which Morgan did not attempt a pass behind the line of scrimmage.
• In the Nebraska game, Johnson was the only player to finish with more than one catch or more than 1 air yard. (He had six catches for 65 air yards.)
• The Morgan completion that traveled the farthest in the air went 43 yards to Johnson, whose late touchdown against Rutgers did nothing to increase his team's odds of winning the game. But it padded his résumé for awards season and perhaps the NFL Draft.
• The Morgan incompletion that traveled the farthest in the air went approximately 46 yards to Bateman against Iowa. (His five throws beyond 30 yards in that game were a season high.)
• Gophers receivers gained just 18 yards after the catch against Northwestern. This both was the smallest number of yards after the catch they gained and made for the smallest proportion of their yardage in a game to come after the catch (8.5 percent).
• In the Georgia Southern game, Gophers receivers gained 53.3 percent of their yards after the catch, their greatest such percentage of the season. A single Johnson touchdown was responsible for 61 of the team's 154 yards after the catch.