An Introduction to 'oTD,' a Replacement for Redzone Targets

Over the past five NFL seasons, 863 pass plays have been called when the line of scrimmage is the opponent’s 20-yard line. 7.4 percent of those plays have resulted in a touchdown.

Over the past five NFL seasons, 727 pass plays have been called when the line of scrimmage is the opponent’s 22-yard line. 7.4 percent of those plays have resulted in a touchdown.

Yet, when we refer to redzone data, we only include plays run from within 20 yards of the end zone. Why? Because it’s what our daddy taught us. But, this day in age, there’s no reason we should be using an arbitrary number to weight what does and doesn’t count as a “scoring opportunity.”

Should a wide receiver screen to Randall Cobb from the 19-yard line really be valued the same as a quick slant to Calvin Johnson at the two-yard line? Of course not. But, if you’re referencing redzone data, that’s exactly what you’re doing.

Today, I’m going to introduce new statistics that will replace blatantly flawed redzone target data. And I’m going to take it a step further, focusing the new statistics on where the player was on the field when targeted, rather than the original line of scrimmage.

By the end of the project, we’ll have two new statistics. One is the average distance in yardage the player was from the endzone when targeted (tADEZ). The other, more important stat is opportunity-adjusted touchdowns (oTD).

To kick off this project, I created a chart with four columns. The first shows each yard marker from 0 (which includes all 10 yards of the endzone) to 109 (the longest possible play distance). The next column includes five years (2008 to 2012) of target data for each yard line. Remember, we're charting the location of the targeted player when the ball arrived at this person, not the line of scrimmage. The third column indicates the converted touchdowns on each target. The final column is the conversion rate.

[table id=12 /]

–

This chart is more for information than anything, with its real purpose coming later when we use it to weight each of the 83,631 targets from the last five years. Of course, one of my goals here is to stop you from blindly trusting redzone data, so let’s look at it for a minute or two.

It shouldn’t be a surprise to see a progressive decline in scoring rate as you work your way down the list. What’s slightly interesting to me is that the conversion rate on a ball thrown into the endzone (36.9 percent) isn’t much higher than the conversation rate on a ball directed to a player standing on the one-yard line (36.0 percent). That’s something to keep in mind next time you trash Tom Brady for throwing a pass short of the endzone on third down.

We do see a few areas of sharp decline. One is the difference between two (32 percent conversion rate) and three (22 percent) yards out of the endzone. Next, we see a 23 percent mark from four yards out, but a dip to 15 percent at five yards. From there, the drops are evenly distributed for the most part.

What’s key here is the tiny difference between the conversion rate from 11 and, let’s say, 18 or 21 yards out. Players score when targeted at the 11-yard line 6.5 percent of the time. They convert from the 18-yard line at a 5.2 percent rate and from 21 yards out 4.0 percent of the time. And that’s all from farther than 10 yards away from the endzone. You could make a case that it’s not worth getting excited about a player’s chances of scoring until he’s targeted inside the five-yard line.

The next step in our process is to weight each target from the last five NFL seasons. From there, we can determine our “oTD” for each player. The next chart includes six columns. The first includes each player who saw at least one target during the 2012 season. The next is the total number of targets for each player. The third column is tADEZ, which was explained earlier. Note that tADEZ is here only for informational purposes and has zero impact on the final three columns. The final three columns will be our main focus. We have actual touchdowns, “oTD,” and the difference between the two.

[table id=11 /]

–

This chart is sorted by “oTD.” Before I dive into analysis, I need to make something clear. “oTD” is an adjusted touchdown number based on two items: total targets and the distance from the endzone on those targets. There is no differentiating a player with great hands vs. a player with stone hands. There’s no factor for a great quarterback vs. a poor quarterback. Remember, we’re focusing on a replacement for redzone target data, which also takes neither of those items into account. This stat is completely related to target volume and scoring opportunity based on the distance from the endzone. That’s it.

Okay, let’s analyze.

First, we’ll look at straight-up oTD data. In a world where all pass-catchers had the same exact hands and quarterback, Brandon Marshall would’ve led the league in touchdowns. A.J. Green in second is no surprise, but Calvin Johnson coming in third is fairly interesting. Johnson had some awful luck when it came to scores in 2012. He was tackled at the one-yard line a ridiculous five times (not to mention another three times inside the five). We’ll discuss the biggest gaps in actual vs. expected later, but note that Johnson came up a whopping 6.2 touchdowns below expected last year. That’s 37.2 fantasy points, my friends.

Reggie Wayne is in a similar boat to Johnson. Despite plenty of scoring opportunities, he managed to find paydirt only five times. He should’ve been closer to 10. Before you jump to the conclusion that all players with high target volumes and low touchdown totals are going to climb to the top of this list, search our list for Andre Johnson. I’ll give you a minute.

Johnson is 46th on this list despite ranking fifth in targets. Johnson’s opportunity-adjusted touchdown total is 4.8, which suggests his low touchdown total of four is legit. Why is that? Check this out:

Shown are the five most-targeted players of the 2012 regular season. First of all, notice Andre Johnson’s tADEZ, which was highest of the five wide receivers at 49.7 yards. Calvin Johnson (40.5) and Wayne (44.2) both had much lower figures. Next, check out the second-to-last column. Andre Johnson was targeted only six times while in the boundaries of the endzone last year. Calvin Johnson saw 16 and Wayne was just behind with 15. We learned earlier how important endzone targets are to touchdown totals. Our chart here is further proof.

Next, I want to look at the players with the biggest gaps between their actual and expected touchdown numbers.

The Fortunate

This shouldn’t be even a little surprising, but James Jones scored nearly double the touchdowns he would’ve under normal circumstances. Despite not clearing 100 targets, he managed to find paydirt 14 times. Note that Jones was in the boundaries of the endzone on seven of his scores and was at the two-yard line on four others. The other three touchdowns came from 10, 11, and 12 yards out. All in all, Jones was within two yards of the endzone when targeted 17 times last year and he converted 11 times.

We also see Randall Cobb on this list (seventh) and Jordy Nelson came in 11th. The presence of Aaron Rodgers clearly has a big impact here, but Jermichael Finley only scoring twice helped boost these guys, as well.

You may recall that Nelson scored a seemingly unsustainable 15 touchdowns on only 93 targets in 2011. Had I written this article last year, he would have been first on this list with 15 actual touchdowns, 6.3 expected, and a difference of 8.7. Jones was sixth in 2011 and Greg Jennings was seventh. It’s important to remember that the reason Packers rank high on this list is the elite performance of the offense. Under normal circumstances, they wouldn’t have the opportunity to put up such large numbers. This is important to remember when evaluating the 2013 scoring potential for a player like Jennings (now with Minnesota).

Dez Bryant and Michael Crabtree are other interesting names on this list. Bryant scored 12 times, but had an expected mark of 6.7. Crabtree found the endzone on nine occasions, but would’ve scored 4.2 under normal circumstances. A deeper look at the data explains the discrepancies.

Of Bryant’s 12 touchdowns, he was in the endzone when targeted on only six of them. Even more telling, he scored the other six from at least seven yards out of the endzone. And taking it even a step further, four came from outside the 20-yard line. Bryant had to run 48 yards for one of the scores (100:1 odds) and 65 for another (1,000:1 odds).

Crabtree’s resume, as you might imagine, is similar. He was in the endzone when targeted on only one-third of his nine scores. He was two, four, seven, nine, 19, and 32 yards out on the other six.

It’s obvious that these are two players who can make plays with the ball in their hands. That’s certainly a factor here. But that doesn’t change the fact that they weren’t given nearly the same opportunity to score as players like Brandon Marshall, A.J. Green, and even Calvin Johnson.

Quickly glancing over the other names here, Danario Alexander, Santana Moss, and Kyle Rudolph were already logical regression candidates, so seeing them here is no shocker. Eric Decker, Darren Sproles, and Rob Gronkowski will continue to take advantage of high-scoring pass offenses.

The Unfortunate

We already talked at length about Johnson and Wayne, so we’ll leave them be.

Despite a breakout 2012 season, Hartline scored only once, so it’s no shocker to see him here. Our data shows how fluky the one-score campaign was. He’ll score more often going forward, even with Mike Wallace in town.

Ex-Panthers WR Louis Murphy is really interesting to me because he was only targeted 55 times, but the data suggests he should’ve scored nearly six touchdowns. Digging deeper, we see that Murphy was in the endzone when targeted a whopping 10 times last season. He failed to convert any of them into touchdowns. Murphy saw four other targets when inside the three-yard line, scoring on only one. We also see Murphy’s teammate Steve Smith on this list, which makes you wonder if Cam Newton is bad or just unlucky when throwing into or near the endzone. That’s a study for another day.

The news is good for Larry Fitzgerald, who suffered through awful quarterback play last season. Bruce Arians’ arrival means his usage will be adjusted, but the presence of Carson Palmer is sure to result in a rebound season.

—

If you read and understand what I did here today, your brain probably hurts. So I’m going to stop here for now. There are certainly other areas that need to be examined. First of all, I want to look at the opportunity-adjusted touchdowns on a per-target rate. Our oTD stat is just raw data that will always be inflated for players seeing significant targets.

The next obvious study is one focused on receptions instead of targets. That would eliminate the earlier issues related to a player’s catch rate and the ability of the quarterback. Of course, that wouldn’t really be a proper replacement for redzone targets. Players on good teams and those who had lucky catch rates would be inflated. Still, there’s some value, especially if you’re only looking to adjust touchdown rates for players returning to the same team/offense the next year.

Additionally, today’s research was focused solely on the passing game. I plan on doing the study on running backs at some point this offseason.

Remember, redzone stats are terribly flawed and not a good method for evaluating scoring potential. Refer to the above tables instead and you’ll come away with significantly better feel for which players are and aren’t being put in position to score touchdowns.

Want to discuss? Leave a comment or hit up Mike Clay on Twitter at @MikeClayNFL