83 Comments

Data Entry: Searching for Stats to Contextualize Trubisky’s Rookie Season

| January 30th, 2018

I recently looked at Trubisky’s rookie performance in “quarters” – four-game sets – and found that he showed continual growth in both usage and efficiency (in all areas but throwing touchdowns) as the season progressed.

Now I want to look at how that growth compares to other recent quarterbacks in their rookie seasons. Do quarterbacks who are going to be good show more growth during their rookie season? Do those who stay the same, or get worse, tend to bust?

The Set-Up

I looked at all QBs drafted in the 1st round who played at least twelve games of their rookie season within the last 10 years and tracked their progress in four-game samples. All data was compiled using the Pro Football Reference game play finder. Allow me a brief explanation of my 3 limits:

  • 1st Round Picks. I wanted players similar to Trubisky, who were drafted with the expectation of playing early. Later round picks often have to earn the job so I didn’t want to include them and skew the data.
  • In the Last 10 years. The NFL passing game continues to evolve, as does the college passing game that prepares them for the NFL. Comparing rookie QBs now to rookie QBs from 20 years ago just isn’t reasonable. Heck, even comparing now to 10 years ago isn’t great, but cutting it much shorter than that really limits the sample size, which is already pretty small.
  • Who Played Twelve Games as Rookies. I’m tracking growth in four-game samples, and two sets of data isn’t really enough, so twelve games gives some sort of growth trend through at least 3 sets.

These stipulations gave me a sample size of 16 quarterbacks: Matt Ryan, Joe Flacco, Mark Sanchez, Sam Bradford, Cam Newton, Blaine Gabbert, Robert Griffin III, Ryan Tannehill, Brandon Weeden, Andrew Luck, Teddy Bridgewater, Blake Bortles, Jameis Winston, Marcus Mariota, Carson Wentz, and Mitch Trubisky.

Before doing this study, it seemed fairly logical to me that most rookie QBs would naturally improve as the season wore on. After all, they’re brand new at this and facing a steep growth curve. And you usually get better at your job within the first few months, right?

Also, take into consideration that most of these quarterbacks were starters from day one of training camp, let alone the regular season. Trubisky faced the unique scenario of not seeing first-team reps until after the first quarter, as the Bears prepared to face what was then the league’s best defense.

Nevertheless, we look at the numbers.

Not-So-Steady Improvement

It turns out that’s not actually the case.

There were some QBs who got better as the season wore on. Matt Ryan, Joe Flacco, and Teddy Bridgewater all showed a clear jump in performance after their first 4 games.

But there were also QBs who got worse after a good start, including Carson Wentz and Cam Newton. Some didn’t really change much from their beginnings, whether those were great (Robert Griffin III), awful (Blaine Gabbert, Mark Sanchez), or somewhere in between (Sam Bradford, Andrew Luck).

In short, there is no one-size-fits-all improvement narrative for rookie quarterbacks. Which makes sense if you think about it; these players come from different college offenses, have different NFL offenses, with different coaches and coaching approaches, and play in different contexts regarding their supporting casts and how many points they need to score to compete.

No QB really seemed to match Trubisky’s rookie year perfectly. Statistically, his first four games were pretty rough, with a completion percentage below 50% and a passer rating in the mid-60s. Those numbers compare to the first few games from Blaine Gabbert and Mark Sanchez, which is not good company to be in. But Trubisky got better from there in ways that those two didn’t, which should give Bears fans some hope. There were QBs who started out solidly and got better, but nobody really went from bad to solid like Trubisky did in his rookie year.

In terms of trends, QBs who improved over their rookie season (mainly Ryan, Flacco, and Bridgewater) seem to have built on that in their careers, barring Bridgewater’s unfortunate injury. But that’s a sample size of 3, so it’s not like this is anything definitive. That is the main theme you pick up on in this article. Nothing about predicting Trubisky’s career is certain based on statistical profiling.

Looking for Indicators

Looking at growth through the rookie season didn’t really show me what I was looking for, but as long as I had all of this data, I figured I’d use it to look at a few different statistical markers. My goal was to see if anything could help predict whether a QB was likely to become a stud or a bust based on their rookie performance.

I compared all of the rookie quarterbacks in the sample in terms of:

  • Passer rating
  • Completion percentage
  • Yards per attempt
  • Touchdown percentage
  • Interception percentage
  • Touchdown/interception ratio

I looking for any stat that helps group the better quarterbacks towards the top of the lists and the worse ones towards the bottom. Nothing worked terribly well for helping identify the best quarterbacks  -which makes sense given that Andrew Luck and Carson Wentz both had mediocre rookie seasons – but I found that averaging them all together worked pretty well to group most of the busts at the bottom (shown in red).

A few thoughts here:

  • You have to basically just ignore Robert Griffin III. He had a phenomenal rookie season, but a combination of injuries, attitude problems, and an unsustainable gimmick offense took his career off the rails in a hurry. Besides him, the four QBs shown at the bottom have to be considered the worst of this group, with only Bortles having a faint chance to change that.
  • If you consider the Tannehill/Gabbert as kind of a Mendoza line, Trubisky finds himself on the right side of it, if only barely. Every other quarterback who in the white region has at least established themselves as a solid NFL starter (save injury), so that’s a good sign for Bears fans.
  • I wouldn’t put too much stock in ranking QBs within the white region. We still don’t know who’s better out of, say, Marcus Mariota and Carson Wentz, or Matt Ryan and Andrew Luck, for that matter.
  • It’s interesting to look at where Trubisky did well and poorly compared to his peers. He was middle of the pack in rating and completion percentage, one of the best at avoiding interceptions, but the absolute worst at throwing touchdowns. As I’ve said before, I’m curious to see if he can get a jump in touchdowns without a corresponding one in interceptions going forward.

Conclusions

I set out to try and find comparable rookie seasons to what Mitchell Trubisky had, and couldn’t find anybody with quite the same progression as him, though recent quarterbacks who improved during their rookie seasons were mostly successful in the NFL. When comparing total rookie seasons statistically, most of the worst quarterbacks group themselves at the bottom, and Trubisky avoids that grouping.

These stats make me tentatively hopeful that Trubisky will at least avoid being a bust (barring an unforeseen injury), but please let me caution that this is far from absolute. As we saw looking at growth throughout the rookie season, every situation is unique, which is why you can never predict the future perfectly.

Still, after watching Trubisky’s rookie season, I came away thinking that (barring injury) he will at least be a serviceable NFL quarterback. Nothing I have found statistically through any of this digging has suggested that to be an illogical belief. I’m very curious to see just how good Trubisky will be (great? good? ok?), but there’s not really any data yet to give any sort of clear indication in that regard.

Tagged: , , , , , , , , , , , , , , , ,