Tannehill

[Dolfanalyst]

I'm not sure where your stats came from but according to NFL.com Bortles was sacked 51 times to lead the league. Rodgers came in second with 46 and then Tannehill tied with Smith & Wilson at 45 to finish third.

Tannehill was 13th in the league in sack rate (percentage of pass dropbacks resulting in a sack) and wasn't significantly different from the average in the league in that regard.

The average in the league in 2015 was 6.16%, and Tannehill's 7.1% wasn't a standard deviation (1.86%) higher than the average.

In other words, he experienced nothing significantly different from the league norm in terms of the percentage of the time he was sacked when dropping back to pass.

On top of that, the correlation between sack rate and QBR (ESPN's QB rating) in 2015 was -0.02, meaning that there was no relationship between those two variables.

In other words, there are a good number of quarterbacks who are sacked a larger percentage of the time who nonetheless play very well, and conversely a good number of quarterbacks who are sacked a smaller percentage of the time who nonetheless play poorly.  The two variables aren't related.

[EKnight]

In other words, it's time for the Tannehill apologists to start realizing he's just not that great as a QB, and the continued excuse of the Oline doesn't hold much merit.

[Dolfanalyst]

In other words, it's time for the Tannehill apologists to start realizing he's just not that great as a QB, and the continued excuse of the Oline doesn't hold much merit.

Over the past two years of his career he's been about exactly average overall, in comparison to the other quarterbacks in the league.  Right smack in the middle of the road.

There are people who believe he'd be doing better if what was going on around him on the team was different, but there's really no way of confirming or refuting that with any certainty.  It's an unknown, and an unfalsifiable argument.

You can win big with an average QB, but it takes exceptionally good circumstances (or variables) elsewhere on the team.

[Rich]

Over the past two years of his career he's been about exactly average overall, in comparison to the other quarterbacks in the league.  Right smack in the middle of the road.

There are people who believe he'd be doing better if what was going on around him on the team was different, but there's really no way of confirming or refuting that with any certainty.  It's an unknown, and an unfalsifiable argument.

You can win big with an average QB, but it takes exceptionally good circumstances (or variables) elsewhere on the team.

I feel like I have read this post 73 times before...

[Spider-Dan]

In other words, it's time for the Tannehill apologists to start realizing he's just not that great as a QB, and the continued excuse of the Oline doesn't hold much merit.

Don't you mean he's a bust?  I mean, that is what you said after halftime of his first game in Week 1 of his rookie season, 63 consecutive starts ago.

[Rich]

Tannehill was 13th in the league in sack rate (percentage of pass dropbacks resulting in a sack) and wasn't significantly different from the average in the league in that regard.

The average in the league in 2015 was 6.16%, and Tannehill's 7.1% wasn't a standard deviation (1.86%) higher than the average.

In other words, he experienced nothing significantly different from the league norm in terms of the percentage of the time he was sacked when dropping back to pass.

On top of that, the correlation between sack rate and QBR (ESPN's QB rating) in 2015 was -0.02, meaning that there was no relationship between those two variables.

In other words, there are a good number of quarterbacks who are sacked a larger percentage of the time who nonetheless play very well, and conversely a good number of quarterbacks who are sacked a smaller percentage of the time who nonetheless play poorly.  The two variables aren't related.

I'm not sure where you are pulling these numbers from, but aside from the 7.1%, they are wrong.

7.1% is 23rd out of 32 teams, not 13th. The low is 3.4%, the high is 9.2%. So by either metric, he is nowhere near the league average, which is 5.7%.

Also, why would there be a correlation between sacks and QB rating? If the QB is being sacked, a pass is not being throw.

Also, just looking at sacks as not looking at the bigger picture. There are a plethora of other things to consider when correlating pressure to QB performance.

[Dolfanalyst]

I'm not sure where you are pulling these numbers from, but aside from the 7.1%, they are wrong.

7.1% is 23rd out of 32 teams, not 13th. The low is 3.4%, the high is 9.2%. So by either metric, he is nowhere near the league average, which is 5.7%.

The 7.1% is the average sack percentage of the QBs on this page:

http://espn.go.com/nfl/qbr

The sack percentage data is taken not from that page, but from here, however:

http://www.pro-football-reference.com/years/2015/passing.htm

Also, why would there be a correlation between sacks and QB rating? If the QB is being sacked, a pass is not being throw.

There shouldn't be, but one could make the argument perhaps that the more often a QB is sacked, the more he's "beaten up" so to speak, and the worse his performance should be.  That isn't found in the numbers, however.  One could also view sacks as an analogue of pressure and think that QBs who are sacked more often must also be experiencing greater or more frequent pressure.

Also, just looking at sacks as not looking at the bigger picture. There are a plethora of other things to consider when correlating pressure to QB performance.

Let's get those things out there and take a look at them.

[Rich]

The sack percentage data is taken not from that page, but from here, however:

http://www.pro-football-reference.com/years/2015/passing.htm

Your second link has Tannehill ranked 24th, which is one spot worse than the 23rd I found. Which means he is in the bottom of the league in sacks per pass attempt. Again, I don't see your 13th ranking. Therefore, Tannehill is sacked more frequently per pass attempt than the average NFL QB, whether we look at your source or mine.

http://www.sportingcharts.com/nfl/stats/sacked-percentage/2015/

So you are providing a source that says 24th and I am providing a source that says 23rd.

There shouldn't be, but one could make the argument perhaps that the more often a QB is sacked, the more he's "beaten up" so to speak, and the worse his performance should be.  That isn't found in the numbers, however.  One could also view sacks as an analogue of pressure and think that QBs who are sacked more often must also be experiencing greater or more frequent pressure.

There isn't always a correlation between sacks and pressures. A QB that is getting rid of the ball quicker may be getting sacked less and pressured more behind terrible blocking. Conversely, a QB that holds onto the ball longer may be getting great protection and get sacked more because no matter how great the protection is, the QB is staying longer in the pocket to look downfield or because he can't figure out the defense... whatever reason.

Too many variables to just look at sacks and pressures and reach a conclusion.

Let's get those things out there and take a look at them.

The only way to really get them out there is to look at film of every game and track them. For the number of stats and data sites out there today, the amount of information disclosed is still limited.

[Dolfanalyst]

Your second link has Tannehill ranked 24th, which is one spot worse than the 23rd I found. Which means he is in the bottom of the league in sacks per pass attempt. Again, I don't see your 13th ranking. Therefore, Tannehill is sacked more frequently per pass attempt than the average NFL QB, whether we look at your source or mine.

http://www.sportingcharts.com/nfl/stats/sacked-percentage/2015/

So you are providing a source that says 24th and I am providing a source that says 23rd.

If we use the site linked above (Sporting Charts), the league average is 6.15%, and the standard deviation is 1.68%.  Tannehill's sack percentage of 7.1% would therefore be well within a standard deviation of the league average.

League ranks can be misleading when there isn't much variation from the average in the league.  23rd in the league may not be a whole lot different from 14th under those circumstances, in other words.

You want to see how far someone deviates from the league norm, not where he's ranked in the league, to get an idea of how exceptional he is, if at all, in some area.  If he doesn't deviate significantly from the norm, then he's part of the norm.  He isn't an exception, and he therefore isn't exceptional (in either the good or the bad direction).

[Rich]

If we use the site linked above (Sporting Charts), the league average is 6.15%, and the standard deviation is 1.68%.  Tannehill's sack percentage of 7.1% would therefore be well within a standard deviation of the league average.

League ranks can be misleading when there isn't much variation from the average in the league.  23rd in the league may not be a whole lot different from 14th under those circumstances, in other words.

You want to see how far someone deviates from the league norm, not where he's ranked in the league, to get an idea of how exceptional he is, if at all, in some area.  If he doesn't deviate significantly from the norm, then he's part of the norm.  He isn't an exception, and he therefore isn't exceptional (in either the good or the bad direction).

Looking at league average could be misleading as well.

The worst team is 9.2%. The best team is 3.4%.

The Dolphins are 2.1% points away from the worst team and 3.7% points away from the best team.

They are 56% closer to worst than to best.

[Dolfanalyst]

Looking at league average could be misleading as well.

The worst team is 9.2%. The best team is 3.4%.

The Dolphins are 2.1% points away from the worst team and 3.7% points away from the best team.

They are 56% closer to worst than to best.

That's a good point, in that there may be a skewed distribution.  We'd have to use some different statistics in that event.

[fyo]

There isn't always a correlation between sacks and pressures. A QB that is getting rid of the ball quicker may be getting sacked less and pressured more behind terrible blocking. Conversely, a QB that holds onto the ball longer may be getting great protection and get sacked more because no matter how great the protection is, the QB is staying longer in the pocket to look downfield or because he can't figure out the defense... whatever reason.

Too many variables to just look at sacks and pressures and reach a conclusion.

The only way to really get them out there is to look at film of every game and track them. For the number of stats and data sites out there today, the amount of information disclosed is still limited.

FootballOutsiders do exactly that. They look at the tape of each and every play and chart a whole slew of statistics. I've seen sacks broken up by time (quick, medium, slow) there in articles, but I don't know if the raw data is available.

(Just to add another sack percentage data point, FO have Tannehill in 24th place with 7.5% and a league average of 6.4%.)

[Rich]

FootballOutsiders do exactly that. They look at the tape of each and every play and chart a whole slew of statistics. I've seen sacks broken up by time (quick, medium, slow) there in articles, but I don't know if the raw data is available.

(Just to add another sack percentage data point, FO have Tannehill in 24th place with 7.5% and a league average of 6.4%.)

PFF does the same thing, track play by play.

The problem still remains that I could look at every play and you could look at every play and we would come up with different results.

I'd rather watch the film and see the results for myself.

[Rich]

That's a good point, in that there may be a skewed distribution.  We'd have to use some different statistics in that event.

You can chart team by team and remove any outliers to levelize/normalize our results and derive a true average from there.

Or you can simply watch the game and see Dallas Thomas and Jason Fox getting beat on a regular basis.

[Dolfanalyst]

Or you can simply watch the game and see Dallas Thomas and Jason Fox getting beat on a regular basis.

The problem with that approach is that it wouldn't tell you how often the linemen on other teams are getting beaten similarly, and so you'd risk ending up with an inaccurate appraisal of the Dolphins' line in comparison to those of other teams.

If what you're looking for in the end is an idea of how much to attribute Tannehill's play to the offensive line, then it's crucial to determine first whether his line is functioning significantly differently from the average line in the league.  If it isn't, then it can hardly be reasonably said that Tannehill's play is a function of his line.

What you'd also need is a league-wide correlation between pass blocking effectiveness and quarterback play.  If that correlation was weak or non-existent, then the argument that Tannehill's play is a function primarily of the offensive line's pass blocking would be similarly weak.

The strongest argument that Tannehill's play is a function of the Dolphins' offensive line would come from both 1) a strong league-wide correlation between pass blocking and quarterback play, and 2) quantitative evidence that the Dolphins' offensive line functions significantly more poorly than that of the average team in the league.