Image courtesy of Vikings.com

Teddy Bridgewater is decidedly the most important player on the Vikings roster. How he performs in 2015 and beyond will have the greatest impact on how the Vikings perform as a whole.

After a promising rookie campaign, we now have 13 games and 402 throws worth of data on Bridgewater.

What we really want to know is what to expect from the young quarterback in the years to come. Understanding his level of quality, where he thrives, and where he doesn’t will give Minnesota valuable information with which to build the offense around him. And of course, we as fans just want to know how good he is and how good he will be.

To try and pry this information from the data, we can answer three important questions.

  1. How does Bridgewater’s rookie season stack up against other rookie quarterbacks past and present?
  2. What can these statistics actually tells us and which are the most meaningful?
  3. Can the numbers be tweaked to more accurately represent the performance of Bridgewater and how does he stack up after doing so?

To answer the first and second questions, I have a sample of 59 quarterbacks since 1993, all of which threw 150 or more passes in their rookie seasons.

1. How does Bridgewater’s rookie season stack up against other rookie quarterbacks past and present?

Comparing Teddy with all quarterbacks might be worthwhile, but comparing him with other rookies is a better way to gauge performance against expectations while recognizing that rookie quarterbacks often struggle.

First, let’s get the simplicities out of the way. Teddy was both very young as a rookie and successful in terms of win percentage.

Measure Average Median Bridgewater Ranking (59) Comparable
Age 23.34 23.25 21.81 5 Blaine Gabbert
Win % 36.5% 33.0% 50% 14 Geno Smith

*Age as of September 1 of rookie season

Only four quarterbacks hit the 150-attempt mark at a younger age than Bridgewater did in 2014. Teddy’s success at such a young age is promising for the future. Correlations between age and future success are not strong, but absolute youngest of quarterbacks have found success. The only ones younger than Bridgewater were Alex Smith, Matthew Stafford, Drew Bledsoe, and Vikings legend Josh Freeman.

Minnesota was also more successful in the win-loss column than most teams are that field rookie quarterbacks.

Screen Shot 2015-03-02 at 9.42.33 PM

More descriptive and more telling statistics certainly exist.

Let’s start from the top with overarching views and comparisons and then whittle it down to the fine details. First, these are the five rookie quarterbacks who compare most closely with Bridgewater over a large handful of stats by summing squared differences. Age, win percentage, ANY/A, completion percentage, touchdown percentage, interception percentage, sack percentage and other splits are included in this comparison.

  1. Joe Flacco
  2. Sam Bradford
  3. Byron Leftwich
  4. Colt McCoy
  5. Ryan Tannehill

Advanced passing statistics from Pro-Football-Reference are the best way to compare passers across different seasons and different eras. A full explanation of how the stats are created can be found here. Essentially, the number reflects standard deviations from the mean with 100 representing league average. A higher number is always preferable, even in interception percentage.

Measure Average Median Bridgewater Ranking (59) Comparable
ANY/A+ 85.9 87 93 17 Ryan Tannehill
Comp. %+ 84.7 84 110 4 Jeff Garcia
TD%+ 89.0 89 89 30 Jeff Garcia
INT%+ 90.3 93 94 29 Kyle Orton
Passer Rating+ 84.7 86 97 8 Joe Flacco
Sack%+ 92.6 95 83 40 E.J. Manuel

Bridgewater’s adjusted net yards per attempt being high is promising, because it subtracts sack yardage and penalizes interceptions heavily. Bridgewater struggled with sacks and interceptions as a rookie due in large part to letdowns from his supporting cast, so a high ANY/A shows that he found some success despite his situation at times, making lemonade out of some dry lemons. The one thing Teddy did better than anything else, completing passes with efficiency, is no secret. Only three rookies have done it better since 1993, without considering adjustments like depth of target.

Throwing touchdowns was an issue, only coming in near the middle of the pack. Norv Turner leaned heavily on Matt Asiata on the red zone, rarely giving Teddy the reins in that area of the field. That pulled his percentage down without question. The devil’s advocate would argue that Turner knows what he’s doing and the measure could actually be reminiscent of Teddy’s skills. It’s possible.

Taking sacks was his bugaboo of course. Sack percentages are more likely to be determined outside the scope of the QB’s play though. Offensive structure and quality of the offensive line are large determinants.

Consider these stats from Pro Football Focus. Out of the Vikings’ dropbacks, 42.8 percent were seven-step drops (a league high) and 35.4 percent were five-step drops, both of which ask more of the offensive line and quarterback in terms of managing the rush and avoiding sacks. Top that with an already porous offensive line that became ravaged by injuries and you have a recipe for a lot of sacks.

So all-in-all, this is a good look for Teddy. The most inclusive stat, ANY/A, has him in the top third of rookie quarterbacks. He was also especially efficient in terms of completion percentage.

But what if we compare him to only first rounders? Undrafted stiffs like Chad Hutchinson and Ryan Lindley really weigh down the averages, right? Teddy’s relation to the average and median scores actually changes very little in comparison to only those picked in the first round.

Measure Average Median Bridgewater Ranking (36)
ANY/A+ 86.3 87 93 11
Comp. %+ 83.6 79.5 110 3
TD%+ 89.9 91 89 22
INT%+ 88.3 91.5 94 16
Passer Rating+ 84.2 84 97 5
Sack%+ 93.9 96.5 83 27

Bridgewater still lands in the top third in ANY/A and right near the top in completion percentage. The only measure that took a dip was touchdown percentage, clearly falling into the bottom half. We will flesh out the touchdown numbers shortly.

So really, some of the brutal performances of late-round rookies has been outweighed by catastrophic rookie season of first rounders like Alex Smith and Ryan Leaf. There’s little difference after removing those drafted after the first round.

We can go one step further though, scrubbing out all quarterbacks who never became bona fide starters in the league. Just with the eye test, I placed 17 quarterbacks into this category. They are Bridgewater (to compare), Andrew Luck, Ryan Tannehill, Russell Wilson, Cam Newton, Andy Dalton, Matthew Stafford, Matt Ryan, Joe Flacco, Alex Smith, Eli Manning, Ben Roethlisberger, Donovan McNabb, Jeff Garcia, Peyton Manning, Jake Plummer and Drew Bledsoe.

Measure Average Median Bridgewater Ranking (17)
ANY/A+ 92.8 96 93 10
Comp. %+ 90.9 91 110 3
TD%+ 95.6 95 89 13
INT%+ 90.1 98 94 10
Passer Rating+ 91.3 94 97 6
Sack%+ 94.8 99 83 13

Touchdown percentage sinks Teddy again. Even his interception percentage is on the wrong side of the median.

Both inclusive measures, ANY/A and Passer Rating, have Bridgewater near the middle of the pack in a group of what became solid starters or more in the NFL.

Screen Shot 2015-02-28 at 4.12.10 PM

A few takeaways from the advanced statistics are not difficult to conjure. Teddy’s overall rookie stats are encouraging. His first season was certainly well above average and his ability to complete passes at a high rate was uncommon for young passers. At the same time, the lack of touchdowns really weighed him down, and threw too many interceptions for the type of passer he is.

From there, we can analyze what those conclusions even mean. Alex Smith was one of the worst rookie quarterbacks of the last 20 years and turned into a fine enough starting QB. Stafford was rather putrid as a rookie too.

What can these statistics actually tells us and which are the most meaningful?

To find out, I put a number of every quarterback’s career, weighting their season-by-season ANY/A+ scores by the number of attempts and only considering seasons after their rookie seasons. Those without sample size, like Jimmy Clausen and Lindley, are excluded. Nick Foles was also removed, because his small sample gave him nearly the same number as Peyton Manning. The ending scores make sense though. Peyton is 123.7. Roethlisberger is 110.3. Andy Dalton is 99.6. Christian Ponder is 87.3. Ryan Leaf is 72.7.

Then regressions were run to find out which stats correlated with future success as a passer, at least from this sample.

R squared is a mathematical measure of how well the independent variable (the rookie stat) explains the dependent variable (future performance). Its value ranges from 0 to 1, with a higher value representing a closer relationship. Its partner in crime is the correlation coefficient, which is simply the square root of the R squared. The coefficient will give negative values for negative relationships however.

For the stats already discussed, these are the vital statistics. Higher numbers indicate a more telling stat.

Age Win % ANY/A+ Comp%+ TD%+ INT%+ Rating+ Sack%+
CC -8% 33% 49% 36% 60% 2% 43% 23%
R2 0.01 0.11 0.24 0.13 0.36 0.00 0.19 0.05

A few outcomes were expected. Descriptive statistics tell more about a QB’s future than does his win percentage or his age. That was a given. Sack percentage is also of little meaning.

The meat of it is in the comparison of the other stats. Completion percentage performed poorly. Quarterbacks who are given simple tasks can complete a high percentage of throws, so that outcome makes some sense. When ramping it up to yardage per attempt, with bonuses for touchdowns and penalties for sacks and interceptions, the numbers get us somewhere. ANY/A is better than passer rating (which leans on completion percentage too much).

Interception percentage performing poorly was something I also expected. But an R squared of basically zero? That means interception percentage is almost meaningless in predicting future performance. The sample is big enough to say that.

Screen Shot 2015-02-28 at 5.13.33 PM

On the other hand, touchdown percentage outperforms every other measure.

Quarterbacks who threw touchdowns became good quarterbacks. The best five rookies were Wilson, Roethlisberger, Plummer, Robert Griffin III, and Peyton Manning. Graphically, the difference between touchdowns and interceptions for rookies is quite clear.

Screen Shot 2015-02-28 at 5.12.23 PM

What does this mean for Teddy? First, his interceptions are meaningless in terms of what he will eventually become. Considering a number of them clanged off the hands of his receivers and one was a Hail Mary, perhaps we already knew that.

Bridgewater’s lack of touchdowns may be amplified however. He only threw 14 of them, 4 of which were thrown behind the line of scrimmage. Finishing drives with touchdown passes will be an important part of his sophomore season. Proving he has the velocity and anticipation to make tight-window throws in the red zone will go along ways toward making him a complete passer.

But if you followed along with Teddy’s rookie season closely, you may be thinking to yourself right now “He should have had at least one or two more touchdown passes and the number of interceptions he earned is really low. What about drops? Don’t those throw off his stats?”

The answer is yes, but other rookies also had bad luck. Did they have as bad of luck as Teddy did?

There’s only one way to find out. No, I did not sort through a couple hundred thousand throws from those 59 quarterbacks. I did chart the entire rookie seasons of seven quarterbacks though, those being Bridgewater, Blake Bortles, Derek Carr, Luck, Griffin III, Tannehill, and Wilson.

Can the numbers be tweaked to more accurately represent the performance of Bridgewater and how does he stack up after doing so?

They were put through the same wringer I put the collegiate prospects through. The charting process removes throw-aways from completion percentages, adjusts for drops, and corrects for dropped interceptions, unearned interceptions, and dropped touchdowns. Essentially, I try to represent each pass as best it relates to what the QB actually accomplished.

First, here are the locations these seven passers threw to, with the numbers representing depths down the field.

Screen Shot 2015-03-01 at 6.00.00 PM

Bridgewater was especially likely to dump it down short of five yards. Minnesota’s reliance on screens (ineffective ones at that) played a part. More than anything though, it was his tendency to check it down, which was caused by a number of things. At the same time, he threw a higher percentage beyond 15 yards than each of these quarterbacks with the exception of Griffin III and Wilson.

Screen Shot 2015-03-01 at 6.06.11 PM

This chart gives a better feel for the “luck” involved in the box score stats and how they were adjusted through charting. Teddy picked up three touchdowns, one for Charles Johnson’s fumble while extending across the goal line, one for Chase Ford’s reception on the right sideline in Miami, and one for Greg Jennings’ dropped touchdown in Week 17. Carr picked up two touchdowns, but nobody else gained more than one.

Meanwhile, Teddy’s interception total remained the same. Some were removed (Asiata drop vs. Miami, Hail Mary vs. Jets) and some were added (Week 4 vs. Atlanta). All the others gained a significant amount through dropped interceptions though, except for Wilson.

Bridgewater was also the only one to have an increased AY/A, which gives touchdowns a value of 20 yards and subtracts 45 yards for interceptions. He was no Griffin III or Wilson, but clearly a class above Tannehill, Luck and certainly the other rookies.

Screen Shot 2015-03-01 at 6.11.31 PM

*INT% on attempts beyond 5 yards

How do the percentages look now?

Bridgewater actually moves past Luck and Carr in touchdown percentage, the most telling stat. His interception percentage becomes second best.

So the answer to questions about fortune prove to be reasonable, at least from this sample of quarterbacks. Teddy seemed to get a worse deal when you correct for dropped touchdowns and earned/unearned interceptions.

Screen Shot 2015-03-01 at 6.15.08 PM

The charts produce an abundance of other stats too, giving adjusted completion percentages and yards per attempt measures at numerous slices of the field.

Overall, Bridgewater’s completion percentage is among the highest. No surprises there. But even after you remove throws of 5 yards or less down the field, he stays at the top. Did he throw frequently underneath? Sure, but his completion percentage does not really skew because of that.

Screen Shot 2015-03-01 at 6.16.17 PM

Maybe unsurprisingly, Teddy’s general effectiveness dropped at each successive distance down the field. Even so, his completion percentage was second best between 16 and 25 yards and middle of the pack on deep balls. In his final six games, he completed 7-of-11 deep passes (including a drop) for four touchdowns. Another deep ball I counted incomplete was a borderline drop by Charles Johnson in the end zone against the Jets.

Screen Shot 2015-03-01 at 6.17.52 PM

Third-down conversions are another positive indicator for Bridgewater. He just edged out Wilson for second best. Considering Teddy’s biggest weakness, his average arm, the success he found in 3rd-and-long is enlightening. “There are throws he cannot make” was a common refrain before his rookie season and even at times during it. The phrase still gets thrown around, but the numbers do not show that.

Of interest to all Vikings fans is how Bridgewater fared if you isolate only the second half of each of these quarterback’s rookie seasons. Teddy was tremendous down the stretch. How did he compare with the others? (median game removed)

Screen Shot 2015-03-02 at 1.03.44 PM

 *Adj. % beyond 5 yards

Touchdown percentage takes a big jump. Interception percentage dips. Completion percentages jump as well. Bridgewater’s percentage on throws beyond 5 yards is 10 percent higher than the next QB and 30 percent higher than Carr’s!

To summarize, Teddy’s rookie season was somewhere on the spectrum of above average to good by analysis of box score statistics and other advanced statistics. His touchdown percentage was of some concern and his interception percentage came in higher than it really should have been considering his style of play.

After correcting for a few occurrences that skew his baseline stats, Bridgewater’s rookie season was closer to that of Russell Wilson than one would expect. And the idea that Carr had a better rookie campaign than Bridgewater did? Get out of here with that.

At the very least, the Vikings have themselves a worthwhile starter in Bridgewater. If his second-half form rolls into 2015, they’ll have much more than that. The tape says it, and the stats say it too.

 

Advanced Statistics via Pro-Football-Reference.