It’s stat projection day here at Broadway Sports. John Glennon posted an article this morning picking over/unders on the Titans team and individual projections based on Vegas odds.
For my exercise, I’m going to look at things from a mathematical point of view to project a range of what I consider likely outcomes for the Titans 2020 offense.
Often you’ll hear numbers thrown around as fans predict the stats for their favorite players, but without analyzing the full context of a team’s offense, it’s hard to know what’s actually realistic. After all, there’s only so many yards and touchdowns to go around.
Can Henry run for over 1,500 yards again? Will the Titans have two 1,000-yard receivers? Can Ryan Tannehill be the first Titans-era quarterback to hit 3,600 passing yards? Can Jonnu Smith break 1,000 yards himself?
By toying with pace of play, run/pass ratio, and yards per play, we can develop realistic projections for how the 2020 season might end up on the stat sheet.
We’re going to do this a bit like Goldilocks, and by that I mean, I’ve created three different scenarios: a Conservative Approach (too low), an Aggressive Approach (too high), and a Probable Projection that I believe is most likely (just right), using specific parameters based on the Titans scheme and players.
By applying this method, we can get a sense of the range of statistical outcomes we might expect for the 2020 season.
(If you’re just here for the stat projections themselves, click here to scroll to the final numbers.)
Assumptions
We have to make some assumptions for the sake of this exercise. There’s obviously no guarantee any of these will hold true (especially this year), but for our projections, we are assuming:
- Ryan Tannehill will take 100% of the quarterback reps
- Everyone will play 16 games
- The Titans will maintain the same offensive philosophy
Variables
Most of my final season-long stat projections are calculated based on a combination of “rate stats” that are multiplied together to get big-picture totals. For each of my three approaches, there are a number of variables I’ve altered to generate a range of statistical projections, as follows…
Pace of Play
It starts with total plays. Looking at the NFL’s pace of play stats from last season, we can estimate where the Titans might fall in 2020, which will ultimately determine the offense’s total opportunities. The various approaches assume a low, middle, and high-end pace of play, which naturally delivers vastly different outcomes.
Pass–Run Splits
Obviously, the amount of dropbacks compared to handoffs will be important. It’s not enough to say the Titans will run about 1,000 plays this season. How many of those plays will be rush attempts and how many will be pass attempts? We’ll end up using the run-pass split to determine total dropbacks and total carries, which allows us to calculate individual player stats.
The rest of the variables are adjusted at the player level:
Rushing Share and Yards Per Carry
By fiddling with a player’s rushing share, we can project their total carries. For example, if we think Henry will have about 65% of the rushing workload, and we think the Titans will run the ball about 55% of the time, we can project Henry’s rushing attempts using our plays-per-game estimate.
Yards per carry is similar. By multiplying our projected rush attempts number by this yards per carry estimate, we can see how many yards he might gain. Repeating this exercise for the other backs — and Tannehill, too — can give us a projected total rushing offense for the Titans this season.
Rushing TD Percentage
Rushing TD percentage measures the percent of a player’s carries on which he scores a touchdown. Last season, Henry scored touchdowns on 16 of his 303 carries, a little more than 5%. By plugging in various rushing TD % estimates, we can project a range of rushing touchdown totals for each player.
Target Share
Moving to the passing game, the biggest determinant of receiving stats is a player’s target share, or the percent of the overall pass attempts that come their way. Last year, A.J. Brown led the Titans with an 18.75% target share. The top receivers in the league usually see a target share somewhere between 20% and 25% (Michael Thomas led the NFL at 31.8% last year). The Titans offense is a bit different; no player hit 19% in 2019. Obviously, all receiver, running back, and tight end shares must add up to 100%.
Similar to the rushing share, multiplying the player’s target share by the total team pass attempts will tell us each player’s expected number of targets.
Catch Percentage and Yards Per Reception
To calculate the projections for quarterback completion percentage, passing yards, and yards per attempt, as well as each receiver’s catches and receiving yards, we plug in numbers for each receiver’s catch percentage and yards per reception.
Using career averages as a guide, we can pick realistic catch % goals and yards per reception averages to project out the team’s passing numbers.
Quarterback Touchdown Percentage and Receiver’s Touchdown Share
To determine how many passing touchdowns our quarterback will throw, as well as how many touchdowns our receivers might catch, we punch in numbers for the quarterback’s touchdown percentage and each receiver’s share of the total passing touchdowns. Again, looking at career numbers can give us a range of possible outcomes to play with.
Quarterback Interception Percentage
This one is simple. We will project interceptions using different ranges of interception percentages. In our case, using Tannehill’s past seasons, we can project interception totals for “worst-case,” “best-case,” and “most likely” scenarios.
Quarterback Sack Percentage / Yards Lost Per Sack
The final piece of the puzzle is quarterback sack percentage. By multiplying a projected sack percentage by total dropbacks, we get actual sacks and pass attempts. We can then multiply our sack estimate by the projected yards lost per sack to determine how many yards to subtract from our total numbers, giving us a final total yardage projection.
Those are the variables we are working with. Now, let’s plug in some numbers and see what we get…
The Conservative Approach
Note: The chart above includes the 2019 Titans stats, as well as the numbers for highest in the league, lowest in the league, and NFL average across the board, which we can use for reference (meaning, if my projection has the Titans yardage well below the NFL’s worst number last season, my variables are off).
For the Conservative Approach, here are my estimates from which we’ll calculate season-long stats:
Pace of Play: Our conservative estimate lands the Titans near the bottom of the league in terms of plays per game. Last season, the Titans were 30th in plays per game at 59.3. That seems like a good floor for the 2020 team, so I set them at 950 total plays, one more snap than last season.
Run-Pass Split: A more “conservative” offense to me means a more run-heavy approach. This projection has the Titans just a hair more run heavy than last season at 53% pass to 47% run.
Rushing Shares: In 2019, Derrick Henry handled 68% of the Titans running back workload, 5th-most in the league. Using that as a baseline, I’ve projected a 63.5% workload for my 2020 conservative approach.
Adding a more effective second back in Darrynton Evans might mean a slightly reduced workload for Henry to help keep him fresh down the stretch. Dion Lewis was at about 12% carry share last year, so handing 3% from Henry to Evans seems fair. I gave Tannehill slightly less than the total QB rushing share from last season and transferred Dalyn Dawkins share to Khari Blasingame. Finally, I projected very small rushing attempt numbers for A.J. Brown and Corey Davis on end-around type plays. Both have taken handoffs for 39+ yards at points in their NFL careers.
Yards per carry: I have Henry’s yards per carry at a very conservative 4.2 average. This seems mighty low given his career average of 4.8, but this is the conservative projection, so 4.2 it is. The other players have conservative projections here, too.
Rushing TD %: Heny has a high career rushing TD percentage at 4.7%, and last year scored on over 5% of his carries. Still, NFL players rarely score 16+ touchdowns in back to back seasons, so I’ve dropped Henry’s rushing TD % to a more modest 3% for this conservative approach.
Target Share: The target share estimates are based on an assumption that A) these receivers all stay healthy enough to play 16 games and B) A.J. Brown and Corey Davis emerge as the team’s true top receiving options, especially with Tajae Sharpe, Dion Lewis and Delanie Walker freeing up about 22% of the team’s targets from last season.
I gave most of Lewis’s targets to Evans and most of Walker’s targets to Jonnu Smith. Kalif Raymond and Adam Humphries got slight bumps as they are expected to play more snaps this year, especially after Humphries missed four games in 2019. The rest of Sharpe’s targets were split between Brown and Davis. Finally, Anthony Firkser and MyCole Pruitt maintained their share from 2019, while Cody Hollister absorbed Rashard Davis’s 1 target.
Catch Percentage: I created a weighted average using career numbers and last season’s stats to determine conservative catch percentage estimates for each player.
Yards Per Reception: Again, using last season’s numbers weighted against career averages, I created conservative yards per reception estimates for each player. Brown’s yards per catch comes down quite a bit in this conservative approach after he averaged 20.2 yards per catch last year.
Touchdown Percentage: Tannehill finished second in the league to Lamar Jackson with a 7.7% touchdown percentage last season. For our conservative approach, I brought that down to Tannehill’s career average of 4.5%. Receiving touchdown share was determined based on last year’s stats, again allowing Smith to absorb Walker’s numbers and splitting up Sharpe’s touchdowns amongst the other four receivers.
Interception Percentage: Tannehill threw interceptions on 2.1% of throws last season, which is slightly below his 2.5% career average. For our conservative approach, I bumped his INT% up to 3%.
Sack Percentage: Marcus Mariota and Tannehill both took too many sacks last year, but Tannehill’s 9.8% sack percentage was a huge improvement over Mariota’s 13.5%. Despite this improvement, Tannehill was actually 2nd-worst in the league to Dwayne Haskins in this stat (Mariota had too few dropbacks to qualify). So if we assume the sack percentage only improves by a very small bit for our conservative estimate, we can estimate a 9% sack rate. Yards lost per sack is based on last year’s team rate.
That was the logic behind the Conservative Approach. This gives us what I consider to be the “floor” for the Titans offense this season. If they aren’t scoring at nearly the same rate, if Henry’s yards per carry and Tannehill’s yards per attempt and touchdown % drop from last season, this is about the “worst case scenario” barring a major injury.
So what do the totals look like for the headliner players in this scenario?
Tannehill: 288/458 (62.9%) for 3,308 passing yards (7.2 YPA), 21 touchdowns, 14 interceptions
Derrick Henry: 284 carries for 1,191 yards, 9 rushing touchdowns
A. J. Brown: 61 catches on 105 targets for 795 yards and 6 touchdowns
Corey Davis: 56 catches on 96 targets for 698 yards and 4 touchdowns
Jonnu Smith: 53 catches on 80 targets for 582 yards and 4 touchdowns
The rest of the player and team final stats are shown in detail in the above chart (click the table to open it in a new window if it’s too small to read).
Remember that Tannehill’s completion percentage, total yards, and yards per attempt are determined by the receivers’ variables for catch percentage and yards per reception. “Determined” in this context simply refers to how the data is input. That doesn’t necessarily mean Tannehill isn’t the one to blame if the receivers’ see an overall decrease in catch percentage because, for example, passes are off target.
It’s sort of a backwards way to calculate Tannehill’s numbers, but it’s really the only way to project the receivers’ stats in a way that adds up.
What’s interesting about these projections, aside from the huge drop in production across the board, is Tanenhill’s stats are kind of like Marcus Mariota numbers. This projection is sort of a 2017-2018 type of scenario.
If the Titans offense regresses in the way many national media members have predicted, this is where they might land.
Let’s Get Aggressive
Note again: the chart above includes the 2019 Titans stats, as well as the highest in the league, lowest in the league, and NFL average stats across the board to use for reference.
Let’s flip the script on that last projection and get aggressive.
Imagine the Titans offense doesn’t regress. Instead, they pick up where they left off last year, and the offense takes another step forward in Tannehill’s second season. Henry continues his reign of dominance, Brown emerges as a top receiver in the league, and Davis has the breakout season we’ve all been waiting for.
What might the Titans stats look like in that scenario? Here are the estimates used to calculate my Aggressive Approach:
Pace of Play: The Titans will never be among the leaders in pace of play unless they vastly change the style in which they operate. A run-heavy, play-action based offense with a bend-don’t-break defense isn’t going to lead the league in total snaps. But with less long touchdowns and more sustained drives, plus an improved pass rush (hello Clowney) and playmakers on offense, we may see them jump up a bit in this category from last year’s 3rd-fewest total plays. This projection of 67 plays per game puts them around the top third of the league based on last season.
Run-Pass Split: Similar to pace of play, even the most aggressive projections for the Titans are going to keep them near the top of the rushing percentage charts. A 44% run rate, as I’m estimating here, would place them as the 12th-most run heavy team by last year’s numbers. That seems appropriately aggressive to me.
Rushing Shares: I bumped Henry back up a bit over my conservative estimate to a 63.5% share here, still down from his 68% last year. That’s again based again on having an effective second back to ease some of the workload. Even with a smaller workload, we can see Henry hit 300 carries in this scenario simply because the offense is projected to run more plays.
Yards per carry: This is the aggressive approach, so I estimated that Henry can nearly duplicate his 5.1 yards per carry average from last season with 5.0 here. That’s slightly above his 4.8 career average and would be Henry’s second-best season, falling squarely between his 2018 and 2019 numbers.
Rushing TD %: Henry managed to score on a little over 5% of his carries last year, so let’s project him right at 5% again in 2020.
Target Share: My target share estimates do not change based on the approach. This rate stat is an estimate that should apply appropriately to all three approaches.
Catch Percentage: Again using a weighted average of last season number’s plus career averages, I estimated a much more aggressive catch percentage for each Titans receiver. This leads to a corresponding bump in Tannehill’s completion percentage and total passing yards.
Yards Per Reception: Using the same process, I created more aggressive yards per reception estimates for each player. Here we see the more optimistic possibility for Brown’s yards-per-catch regression, dropping from his 20.2 rookie average to 15.0, much higher than our conservative approach.
Touchdown Percentage: After throwing a whopping 7.7% touchdown percentage last year, we’re left to wonder if Tannehill can sustain that efficiency for a full season. This aggressive approach assumes that he basically can, as I put him at 7%. Like target share, receiving touchdown share didn’t change from the conservative estimate.
Interception Percentage: Let’s say Tannehill maintains his low interception % also. FootballOutsiders had Tannehill as the second unluckiest quarterback in terms of interceptable throws converting to interceptions. If he plays as well as he did last year and gets a bit luckier here, we can aggressively estimate him at 2%.
Sack Percentage: Let’s say Tannehill also improves his ability to evade sacks behind an offensive line that knows how to play together and shouldn’t take half the season to “get going.” This estimate drops his sack percentage to 7%. Yards lost per sack also came down a yard from our conservative estimate to 6.5 yards lost.
That was the logic behind the Aggressive Approach. This gives us what I consider to be the “ceiling” for the Titans offense this season. It’s how we can imagine the offense if they don’t regress at all from last year’s brilliant performance down the stretch.
So what do the totals look like for the headliner players in this scenario?
Tannehill: 376/558 (67.3%) for 4,752 passing yards (8.5 YPA), 39 touchdowns, 11 interceptions
Derrick Henry: 300 carries for 1,498 yards, 15 rushing touchdowns
A. J. Brown: 82 catches on 128 targets for 1,233 yards and 11 touchdowns
Corey Davis: 74 catches on 117 targets for 1,005 yards and 7 touchdowns
Jonnu Smith: 68 catches on 98 targets for 841 yards and 8 touchdowns
That’s right, the Titans could have two 1,000-yard receivers. But this projection assumes a sizable increase in target share for Davis. Last season, he was at just 15.4% (again, Brown was at 18.75%). Smith was at 9.8%, but Delanie Walker accumulated 6.9% of the targets in the time he was on the field, so it’s unfair to estimate Smith so low this season.
As I mentioned up top, 22% of the team’s target share from last year, including Walker’s, is up for grabs in 2020. It’s possible to expect Davis, Brown, and Smith all to see bumps of about 5% or more, with the rest going to Darrynton Evans, Adam Humphries, and maybe Kalif Raymond.
The rest of the player and team stats are shown in detail in the above chart (click the table to open it in a new window if it’s too small to read). If you look at the top portion, you’ll see that this projection has the Titans about 100 yards shy of the most productive offense in the NFL last season.
Projecting a likely outcome
Note for the final time: the chart above includes the 2019 Titans stats, as well as the highest in the league, lowest in the league, and NFL average stats across the board to use for reference.
Finally, for our Probable Projection, we’ll land most of our variables somewhere between the last two approaches.
Here what I think is most likely as far as variables go for the Titans this season:
Pace of Play: 62.5 plays per game would’ve ranked 23rd overall last year. The Titans are a slower-paced team in general, finishing 30th in this category in 2019, so this seems like a fairly likely outcome if the defense maintains the level they played at last year but manages to intercept more than 14 passes. It’s a slight bump from the 2019 pace of play, but a minor regression in 50+ yard touchdowns could lead to a handful more plays in each game on longer sustained drives.
Run-Pass Split: I settled in at a very slightly more pass heavy split than we saw a year ago, but this estimate is very close to 2019’s numbers. Ultimately, this is who the Titans want to be on offense, and I don’t see much reason for that to change this season.
Rushing Shares: I left Henry at 63.5% rushing share, the same as his aggressive estimate but still down from his 68% last year. I think it’s likely we’ll see a few less carries from Henry overall this season, but he comes close to his 2019 total in this projection.
Yards per carry: For this estimate, I have Henry at a “modest” 4.5 yards per carry. It’s under his career average of 4.8, but with defenses likely turning their attention to the Titans run game this season, Henry could see a drop from the 5.1 yards he averaged last year. 4.5 is still a pretty good per carry average.
Rushing TD %: Again I think we’ll see a bit of a drop from last year’s 5.08 touchdown %, so I estimated Henry at 4% in this projection.
Catch Percentage: I settled in between my floor and ceiling estimates for this projection for each player.
Yards Per Reception: Similar to catch percentage, I found a happy medium between my last two estimates for each player.
Touchdown Percentage: I think it’s most likely that we see Tannehill fall somewhere between his career average of 4.5% and last season’s stellar 7.7%. I gave him 5.5% for this estimate, which would’ve been 8th-highest in the NFL last season.
Interception Percentage: For this estimate, I used Tannehill’s career number of 2.5%.
Sack Percentage/Yards lost per sack: Again I used Tannehill’s career numbers, which are 8% and 7.5 yards, respectively.
So here they are, the most likely stats for the Titans 2020 skill players, based on a bunch of subjective variables as decided by me:
Ryan Tannehill
Low end: 288/458 (62.9%) for 3,308 passing yards (7.2 YPA), 21 touchdowns, 14 interceptions
Mid range: 319/492 (64.8%) for 3,951 passing yards (8.0 YPA), 27 touchdowns, 12 interceptions
High end: 376/558 (67.3%) for 4,752 passing yards (8.5 YPA), 39 touchdowns, 11 interceptions
Derrick Henry
Low end: 284 carries for 1,191 yards, 9 rushing touchdowns
Mid range: 295 carries for 1,329 yards, 12 rushing touchdowns
High end: 300 carries for 1,498 yards, 15 rushing touchdowns
A. J. Brown
Low end: 61 catches on 105 targets for 795 yards and 6 touchdowns
Mid range: 70 catches on 113 targets for 1,018 yards and 7 touchdowns
High end: 82 catches on 128 targets for 1,233 yards and 11 touchdowns
Corey Davis
Low end: 56 catches on 96 targets for 698 yards and 4 touchdowns
Mid range: 63 catches on 103 targets for 826 yards and 5 touchdowns
High end: 74 catches on 117 targets for 1,005 yards and 7 touchdowns
Jonnu Smith
Low End: 53 catches on 80 targets for 582 yards and 4 touchdowns
Mid range: 59 catches on 86 targets for 697 yards and 6 touchdowns
High end: 68 catches on 98 targets for 841 yards and 8 touchdowns
The rest of the player and team stats are shown in detail in the above chart (click the table to open it in a new window if it’s too small to read).
The most mysterious variable to me is target share. It’s possible Brown could dominate the target share the way the top receivers in the league do, around 25-27%. If he does, it’s likely at the expense of Davis and maybe Humphries or Smith. That would shift Brown’s projection up and Davis’s down.
At the same time, you could imagine swapping the target shares between Davis and Smith or bringing them a bit closer together. There is wiggle room on the ends of these ranges for each player based on the variables I set. Ultimately though, no matter how the targets are divided, they all add up to the same total pass attempts.
The same idea goes for the rushing distribution — I didn’t project any carries for Jonnu — and it also assumes there are no fake punt pass attempts or Tim Tebow plays with Henry in the wildcat.
The final so-called “probable” projection comes out with slightly fewer total yards than the Titans posted last season.
