Ryan Tannehill 2013: QB Pressure, Completion %, Deep Passing, & YPA Statistics | Page 18 | FinHeaven - Miami Dolphins Forums

Ryan Tannehill 2013: QB Pressure, Completion %, Deep Passing, & YPA Statistics

Gravity said:
If that's true, I'd encourage you to start a thread that sums up your findings in its original post.

And remember, the points of interest for me at least are: 1) how sacks affect individual QB play in general, and 2) how they affected Ryan Tannehill's individual play in 2013.
Click to expand...

They don't affect QB's play as measured by YPA for any team, including the Dolphins. They affect the team's play.

Here is a study:

http://www.thelionsinwinter.com/2010/05/defensive-line-and-secondary.html

"We have found that there is no inverse correlation between that pass rush effectiveness and passing effectiveness; on plays where a sack does not occur, per-play pass yardage is unaffected.

The correlation coefficient is .068; approaching zero."

That doesn't mean sacks aren't important, it means they don't affect the pass plays that don't include a sack.

Here is a study that looks at the impact of sacks on scoring.

http://www.advancednflstats.com/2008/11/value-of-sack.html



The same is true for running the ball. It doesn't affect ANY team's passing efficiency.

http://www.advancednflstats.com/2007/07/what-makes-teams-win-part-1.html

"If a good running game gives a team an advantage in passing then we would see a positive and significant correlation between offensive running and passing efficiency. In fact, the correlation is 0.13, which is very weak and not statistically significant. "
 
Gravity said:
And I think all that certainly makes sense, but in the absence of even the most foundational level of evidence that he would play better with a better running game (i.e., a correlation between running game variables and his individual play, game-to-game), should we believe he'd play better with a better running game, or should we believe the matter is simply an unknown?

In other words, the more sophisticated the analysis we need to do to support the idea that he'd play better with a better running game, the less willing we should be to believe the idea has merit.

If the correlation between running game variables and his individual play were present, we could stop there and say the idea likely has some merit. When that correlation doesn't exist, however, and we need to propose and conduct more sophisticated analyses of the variables in question, how much merit should we believe the idea to have, until and unless those analyses are conducted?
Click to expand...

But that's what we do here for everything. We speculate whether we have the necessary information or not. Just look at the opinions floating around about certain gm candidates. Look what's gonna happen regarding the players we draft. How many times have ever seen you someone post "I'm sorry, I can't formulate an opinion on that matter due to lack of information".
 
if you seriously think that sacks and pressure don't affect the qb and offense then you have some screws loose...if putting pressure on qbs and getting sacks doesn't affect the qbs performance why have pass rushers? why blitz? I mean if sacks and pressure don't affect the qbs play then why waste time rushing the qb?
 
FinfanInBuffalo said:
They don't affect QB's play as measured by YPA for any team, including the Dolphins. They affect the team's play.

Here is a study:

http://www.thelionsinwinter.com/2010/05/defensive-line-and-secondary.html

"We have found that there is no inverse correlation between that pass rush effectiveness and passing effectiveness; on plays where a sack does not occur, per-play pass yardage is unaffected.

The correlation coefficient is .068; approaching zero."

That doesn't mean sacks aren't important, it means they don't affect the pass plays that don't include a sack.

Here is a study that looks at the impact of sacks on scoring.

http://www.advancednflstats.com/2008/11/value-of-sack.html



The same is true for running the ball. It doesn't affect ANY team's passing efficiency.

http://www.advancednflstats.com/2007/07/what-makes-teams-win-part-1.html

"If a good running game gives a team an advantage in passing then we would see a positive and significant correlation between offensive running and passing efficiency. In fact, the correlation is 0.13, which is very weak and not statistically significant. "
Click to expand...
So then is your conclusion that the YPA associated with Ryan Tannehill's play is unlikely to change as a function of a decrease in sacks or an improvement of the running game?
 
Gravity said:
So then is your conclusion that the YPA associated with Ryan Tannehill's play is unlikely to change as a function of a decrease in sacks or an improvement of the running game?
Click to expand...

The YPA associated with the ANY TEAM'S offensive is unlikely to change as a function of a decrease in sacks or an improvement in the running game.

An improvement to the running game and a reduction in sacks will help the team sustain drives (leading to more points), deal with bad weather conditions, hold leads, makeup for poor days by the pass offense, etc. etc. YPA is not the be all end all stat and it is not a function of the QB alone.
 
Exactly. Well said. I'll add my two cents.

1) What is the average yardage required to attain a first down on 2nd and 3rd down forTannehill vs Brady?

2) What relationship, if any, is there between down and distance with YPA or QB rating?

Tannehill, with his lousy OL and running game, I assume had among the highest average yards required to attain a 1st down on either 2nd or 3rd down. While Brady is typically facing 3rd and 3, Tannehill is typically facing 3rd and 9.

Suppose a QB faces a lot of 3rd and longs. One QB might be prone to attempt to complete a more difficult pass in order to pick up the first down, to the detriment of his YPA and QB rating, because he is concerned with winning as opposed to protecting his statistical measurables. Another QB might decide to throw a safe underneath pass which, while elevating his YPA and QB rating, fails to pick up the 1st down. Would this be winning football? Is this the mark of being a better QB? Maybe. Maybe not. What yard line are you on? What is the score of the game? How much time is remaining?

So to say that a QB's OL play and running game have no impact on a QB's YPA or QB rating is ludicrous. These statistics are dramatically influenced by both down and distance as well as field position and the score of the game. If it's 3rd and 20 and you're winning a QB can throw a safe completion that the defense gives you, thereby augmenting your statistics; conversely, if it's 3rd and 20 and you're losing, you are obliged to incur more risk which has a greater chance of lowering your statistics.

Conversely, when it's 3rd and 3 and you have a good running game that QB also benefits from the defense not knowing whether to play the run or the pass. Here again, Tannehill is at a distinct disadvantage as compared with Brady, irrespective of talent, as the defense will most likely play the pass given the futility of the Dolphins' running game. Against Brady on 3rd and 3, the likelihood of either a run or a pass is nearly equal.

Bottom line is that it's a team game and the team's success in other phases of play dramatically affect the situations which allow a QB to improve upon his individual statistics, even when the underlying talent of two QB's under comparison is equal. And I would argue that these situational variations caused by the team's performance excluding the QB are too complicated to be controlled for when presenting basic statistical comparisons and analyses.


FinfanInBuffalo said:
Because the assumption is that the only difference between NE (the division winner) and Miami is Brady vs Tannehill. That is delusional. WE WERE THE WORST PASS BLOCKING LINE IN THE LEAGUE AND THE 26TH RANKED RUSHING TEAM. Why the hell to posters on this message board keep ignoring the fact that this team had serious issues that are unrelated to the QB?
Click to expand...
 
Last edited by a moderator:
nyashfan said:
Exactly. Well said. I'll add my two cents.

1) What is the average yardage required to attain a first down on 2nd and 3rd down forTannehill vs Brady?

2) What relationship, if any, is there between down and distance with YPA or QB rating?
Click to expand...

The bigger relationship is between down and distance and drive efficiency or points, not YPA. I suspect there are many things at play. Things tend to cancel each other out. For example, a 14 yard pass on 3rd and 15 looks great for your YPA but doesn't continue the drive. An incomplete pass on 3rd and 15 also stops the drive but hurts the YPA. Where they happen on the field is also important. If you have 3rd and 15 and you are 5 yards from FG range, a 5 yard pass might result in a FG rather than a punt.

Finally the time in the game and the score matter also. What is considered a successful play on 3rd and long varies. When down by 6 with 1 second to go on 3rd and long your own 20, only an 80 pass is meaningful.
 
nyashfan said:
Exactly. Well said. I'll add my two cents.

1) What is the average yardage required to attain a first down on 2nd and 3rd down forTannehill vs Brady?

2) What relationship, if any, is there between down and distance with YPA or QB rating?

Tannehill, with his lousy OL and running game, I assume had among the highest average yards required to attain a 1st down on either 2nd or 3rd down. While Brady is typically facing 3rd and 3, Tannehill is typically facing 3rd and 9.

Suppose a QB faces a lot of 3rd and longs. One QB might be prone to attempt to complete a more difficult pass in order to pick up the first down, to the detriment of his YPA and QB rating, because he is concerned with winning as opposed to protecting his statistical measurables. Another QB might decide to throw a safe underneath pass which, while elevating his YPA and QB rating, fails to pick up the 1st down. Would this be winning football? Is this the mark of being a better QB? Maybe. Maybe not. What yard line are you on? What is the score of the game? How much time is remaining?

So to say that a QB's OL play and running game have no impact on a QB's YPA or QB rating is ludicrous. These statistics are dramatically influenced by both down and distance as well as field position and the score of the game. If it's 3rd and 20 and you're winning a QB can throw a safe completion that the defense gives you, thereby augmenting your statistics; conversely, if it's 3rd and 20 and you're losing, you are obliged to incur more risk which has a greater chance of lowering your statistics.

Conversely, when it's 3rd and 3 and you have a good running game that QB also benefits from the defense not knowing whether to play the run or the pass. Here again, Tannehill is at a distinct disadvantage as compared with Brady, irrespective of talent, as the defense will most likely play the pass given the futility of the Dolphins' running game. Against Brady on 3rd and 3, the likelihood of either a run or a pass is nearly equal.

Bottom line is that it's a team game and the team's success in other phases of play dramatically affect the situations which allow a QB to improve upon his individual statistics, even when the underlying talent of two QB's under comparison is equal. And I would argue that these situational variations caused by the team's performance excluding the QB are too complicated to be controlled for when presenting basic statistical comparisons and analyses.
Click to expand...

FinfanInBuffalo said:
The bigger relationship is between down and distance and drive efficiency or points, not YPA. I suspect there are many things at play. Things tend to cancel each other out. For example, a 14 yard pass on 3rd and 15 looks great for your YPA but doesn't continue the drive. An incomplete pass on 3rd and 15 also stops the drive but hurts the YPA. Where they happen on the field is also important. If you have 3rd and 15 and you are 5 yards from FG range, a 5 yard pass might result in a FG rather than a punt.

Finally the time in the game and the score matter also. What is considered a successful play on 3rd and long varies. When down by 6 with 1 second to go on 3rd and long your own 20, only an 80 pass is meaningful.
Click to expand...
I believe both of your posts are addressed by the analytical system described on this page:

http://www.footballoutsiders.com/info/methods

Note Ryan Tannehill's league ranking within that system on this page:

http://www.footballoutsiders.com/stats/qb
 
Holy ****, I've haven't seen this much of an all-consuming sexual fixation since Hinckley's on Jodie Foster!!!

In consideration of your obvious fantasies, can we now assume, based on your name change that you see yourself as "the bottom"??

k9yi6e-1.jpg
 
FinfanInBuffalo said:
Do you finally admit that no correlation between sacks and YPA exists and it is not "thread worthy" to point out that Tannehill's YPA doesn't vary with sacks on a per game basis?
Click to expand...
Sure, but what I find ironic is that we seem to spend more time focusing on pointing out how that game-to-game correlation is to be expected based on the general league correlation, rather than using the general league correlation to dispute the folks who seem to believe that Tannehill is just a decrease in sacks away from having a higher YPA. Why is the focus more on the former instead of the latter?
 
A little statistical analysis I came up with here. Number of times sacked vs the YPA reduction incurred due to lost yards (since sack yardage is subtracted from your passing yardage). I am equating YPA with "QB play" since that is the preference of these threads. YPA lost is the difference between YPA and YPA calculated with lost sack yards added back. If the number of sacks had nothing to do with reduced QB play, I would think that the YPA reduction would be the same no matter how many times you are sacked. But by these numbers, it appears that the YPA increases has a "moderate" correlation of r = 0.498.

Ygwy5m9-1.png
 
wonderl33t said:
A little statistical analysis I came up with here. Number of times sacked vs the YPA reduction incurred due to lost yards (since sack yardage is subtracted from your passing yardage). I am equating YPA with "QB play" since that is the preference of these threads. YPA lost is the difference between YPA and YPA calculated with lost sack yards added back. If the number of sacks had nothing to do with reduced QB play, I would think that the YPA reduction would be the same no matter how many times you are sacked. But by these numbers, it appears that the YPA increases has a "moderate" correlation of r = 0.498.

Ygwy5m9-1.png
Click to expand...
I think what you were perhaps trying to get at there was the difference between net YPA, which accounts for sacks and sack yards, and YPA, which does not.

The correlation between net YPA and YPA this year was 0.95, which tells me they were functionally equivalent.

The difference between net YPA and the "new" YPA number you're arriving at above is that you're eliminating the number of sacks from the net YPA equation. Net YPA = (passing yards - sack yards)/(pass attempts + sacks). Your equation above = (passing yards - sack yards)/pass attempts.
 
Would like a similar thread with all these fancy net ypa and wpa and correlations, standard deviations, or years 1 and 2, in comparison to ryan tannehills year one and two, for the following quarterbacks:

Tom brady
Aaron rodgers
Peyton manning
drew brees
And phillip rivers...

Please.
 
Gravity said:
I think what you were perhaps trying to get at there was the difference between net YPA, which accounts for sacks and sack yards, and YPA, which does not.

The correlation between net YPA and YPA this year was 0.95, which tells me they were functionally equivalent.

The difference between net YPA and the "new" YPA number you're arriving at above is that you're eliminating the number of sacks from the net YPA equation. Net YPA = (passing yards - sack yards)/(pass attempts + sacks). Your equation above = (passing yards - sack yards)/pass attempts.
Click to expand...

No. YPA is yards divided by attempts. YPA with sacks counted as dropbacks would be yards per dropback, not yards per attempt. My terms are clearly defined. Yards per dropback is different and I didn't calculate it.

Edit - I just calculated it, and the correlation that way is 0.532, when you do dropbacks instead of just total attempts. So it indeed appears that getting sacked more has a moderate correlation with worse YPA and worse YPD.
 
You must log in or register to reply here.
Back
Top Bottom