Tannehill

[Dolfanalyst]

So then, coming back from a 95% improbability to win is a good thing?

http://www.pro-football-reference.com/boxscores/201412210mia.htm - comeback from 94.70% improbability to win
http://www.pro-football-reference.com/boxscores/201309220mia.htm - comeback from 94.00% improbability to win
http://www.pro-football-reference.com/boxscores/201310310mia.htm - comeback from 97.90% improbability to win

The irony here is that you are citing Aaron Rodgers as your chosen example of clutch performance, when the knock on Rodgers for most of his career has been his lack of 4th quarter comebacks and poor record in close games.

Of the great QBs in the league, you picked the worst one to make your point with.

You're failing (again) to grasp the fact that I'm not making a point about a certain quarterback, but about a certain kind of play.

And so you found three games (and there could be more) in which Tannehill's clutch-weighted expected points added on plays with pass attempts would be high.  However, he was slightly below the league average in 2015 with regard to the statistic that measures that best.

Again, couple that with the sacks he takes on clutch plays (ones that don't involve pass attempts), and you have a problem.

[Spider-Dan]

If your argument is that Tannehill did not have his best year in 2015, I agree.  His 2013 and 2014 were certainly better.

But you continue to throw out citations and statistics that don't prove any point worth proving.  If I actually believed your statistical "analysis," I would come to the conclusion that both Tannehill and Rodgers were not very good QBs.

[fyo]

So while the point made by "fyo" is a good one

Why is my name in quotes? Seems kind of "odd".

[fyo]

This is how in fact the clutch "weight" is determined.  If a play is associated with no change in a team's probability of winning, then that play has zero weight.

So, basically, they just decided to redefine clutch. Okay, whatever.

It is funny in some sense, though, because maybe the site that really got the ball rolling on football analytics (FootballOutsiders.com) has always been on a crusade against describing anyone as "clutch" and completely tearing down all the BS surrounding it. Of course, that's as the term has been used for decades by talking heads, media, and the rest of us. With this newfangled definition of clutch, FootballOutsiders' own statistics are clutch-weighted. That's a nice dose of irony.

It really doesn't matter much to me what term is used to describe these statistics, but clutch-weighted seems an awful label in this case.

Whatever. Not going to spend more time on that, just pointing it out.

[Dolfanalyst]

If your argument is that Tannehill did not have his best year in 2015, I agree.  His 2013 and 2014 were certainly better.

But you continue to throw out citations and statistics that don't prove any point worth proving.  If I actually believed your statistical "analysis," I would come to the conclusion that both Tannehill and Rodgers were not very good QBs.

Well now if you expand that analysis and look at the clutch-weighted plays as a whole:

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

...you'll see that Rodgers was better than Tannehill in 2015 not only with regard to those plays in which pass attempts were made, but also the ones were there were sacks, runs by the QB, and penalties (i.e., throws that drew pass interference calls).

When you consider all of that as perhaps indicative of a big part of playing the quarterback position as a whole, I think you have to wonder whether Tannehill has what it takes to make the team competitive with the best teams in the league in a playoff situation.

[Dolfanalyst]

So, basically, they just decided to redefine clutch. Okay, whatever.

It is funny in some sense, though, because maybe the site that really got the ball rolling on football analytics (FootballOutsiders.com) has always been on a crusade against describing anyone as "clutch" and completely tearing down all the BS surrounding it. Of course, that's as the term has been used for decades by talking heads, media, and the rest of us. With this newfangled definition of clutch, FootballOutsiders' own statistics are clutch-weighted. That's a nice dose of irony.

It really doesn't matter much to me what term is used to describe these statistics, but clutch-weighted seems an awful label in this case.

Whatever. Not going to spend more time on that, just pointing it out.

But yet even their statistics are clutch-weighted.  Note the following:

Every single play run in the NFL gets a "success value" based on this system, and then that number gets compared to the average success values of plays in similar situations for all players, adjusted for a number of variables. These include down and distance, field location, time remaining in game, and the team’s lead or deficit in the game score.

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

[Dolfanalyst]

Why is my name in quotes? Seems kind of "odd".

Only because that can't possibly be your real name.

Or is it?

[Spider-Dan]

Well now if you expand that analysis and look at the clutch-weighted plays as a whole:

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

...you'll see that Rodgers was better than Tannehill in 2015 not only with regard to those plays in which pass attempts were made, but also the ones were there were sacks, runs by the QB, and penalties (i.e., throws that drew pass interference calls).

This is a really long-winded way of saying that Rodgers' QBR was better than Tannehill's in 2015, which is a point that was never in dispute.  However, looking at the total QBR does not tell you anything specific about how those QBs performed "in the clutch," only how they played under all conditions (including "the clutch").

But originally, you didn't just cite Rodgers' QBR and say, "look, Rodgers is better than Tannehill... here are their QBRs" because that would be a waste of time (as everyone already knows Rodgers is better).  Instead, you lauded Rodgers' performance at the end of a pair of games while citing sack EPA and pass EPA to "prove" Tannehill's incompetence... while somehow failing to notice that those same stats indicate Rodgers is also bad.

[Dolfanalyst]

This is a really long-winded way of saying that Rodgers' QBR was better than Tannehill's in 2015, which is a point that was never in dispute.  However, looking at the total QBR does not tell you anything specific about how those QBs performed "in the clutch," only how they played under all conditions (including "the clutch").

Those statistics are clutch-weighted, so while yes, they tell you how quarterbacks played under all conditions, there is greater weight assigned to plays made in the clutch.  Consequently, you could have two QBs who have identical traditional QB ratings, but one will have a higher QBR than the other if he made a greater number of clutch plays, or avoided a greater number of negative plays in the clutch.

But originally, you didn't just cite Rodgers' QBR and say, "look, Rodgers is better than Tannehill... here are their QBRs" because that would be a waste of time (as everyone already knows Rodgers is better).  Instead, you lauded Rodgers' performance at the end of a pair of games while citing sack EPA and pass EPA to "prove" Tannehill's incompetence... while somehow failing to notice that those same stats indicate Rodgers is also bad.

Once again, I lauded the kinds of plays both Rodgers and Palmer made at the end of a game, and noted that Tannehill's incidence of such plays has been comparatively rare.  Those were simply examples of clutch plays made by quarterbacks, irrespective of who made them.  We could find similar plays John Beck of all people made as a rookie in 2007 (if they exist), and the point would be the same.

You seem to be out to expose some sort of inconsistency I have with myself in this area.  Let me save you the trouble -- it isn't there.

[Dolfanalyst]

This is the most reasonable position about Ryan Tannehill currently, in my opinion.  It would be foolish for the team to think the changes made around Tannehill (Gase, etc.) will guarantee his improvement, without having a fallback option.

The Dolphins need to draft a quarterback.
 
They need to do it this draft.
 
And next draft.
 
And every draft, if they must, because there are myriad reasons this approach is the right thing to do.
 
And I'm not talking draft a quarterback in the sixth round -- perhaps maybe around the 199th selection (for those of you that get where I'm going) -- or as an afterthought in the final round. I believe the Dolphins should set their draft board and if a quarterback is the best player on that board in any round, they should pick that quarterback.
 
That means second round ...Third round...Fourth round.
 
Even in the first round.
 
Yes, in the first round. And I'm talking about in the 2016 first round, if necessary.

http://miamiherald.typepad.com/dolphins_in_depth/2016/01/the-right-path-to-finding-an-elite-qb-is-taking-multiple-paths.html?

[Brian Fein]

Why are you guys using Aaron Rodgers, one of the league's best QB's, as a comparison?  If he's the benchmark, then everyone else in the league also sucks.

[Dolfanalyst]

Greg Cosell's view of a couple of the plays I mentioned earlier in this thread:

http://sports.yahoo.com/blogs/nfl-shutdown-corner/greg-cosell-s-film-review--the-two-big-packers-cardinals-plays-211724878.html

[Dolfanalyst]

Why are you guys using Aaron Rodgers, one of the league's best QB's, as a comparison?  If he's the benchmark, then everyone else in the league also sucks.

One can make a decent case that winning a Super Bowl in this day and age is fairly unlikely unless a team has a QB the caliber of Rodgers or the other QBs in the league like him.

Since the league changed the rules to favor the passing game in 2004, the winning Super Bowl QBs have been Brady, Roethlisberger, Brees, Rodgers, Manning (Payton and Eli), Flacco, and Wilson.

If you distinguish Eli Manning and Flacco unfavorably from the rest, then you're left with 3 of the past 12 Super Bowls in which one of the league's elite QBs was not the winner.  That in turn leaves you with a 75% probability of failing to win the Super Bowl without a QB of that caliber.

So having Rodgers et al. as a benchmark may be the smart thing to do, if the goal is winning a Super Bowl.  It can be done without a QB of that caliber, but it isn't likely.

Moreover, when Eli Manning and Joe Flacco won Super Bowls, they were playing statistically at an elite level virtually throughout the playoffs.

So you need either an elite QB, or one who can play like one throughout the playoffs.  That doesn't augur well for the average Joe in the league, who isn't that caliber and hasn't demonstrated any ability to play well in the playoffs.

[Brian Fein]

I disagree with your assessment that you need an elite QB.  Plenty of contrary examples in NFL history.  Frankly the discussion of "elite" with respect to QB's is a tired one.

Any team can be successful in the NFL with the right puzzle pieces.  Plenty of teams have had success with average-to-mediocre QB's and dominant defenses.  Concurrently, the "elite" label is self-fulfilling, as most talking heads will consider any QB who has won a Super Bowl as "elite" (see: Russell Wilson).

[Dolfanalyst]

I disagree with your assessment that you need an elite QB.  Plenty of contrary examples in NFL history.  Frankly the discussion of "elite" with respect to QB's is a tired one.

Any team can be successful in the NFL with the right puzzle pieces.  Plenty of teams have had success with average-to-mediocre QB's and dominant defenses.  Concurrently, the "elite" label is self-fulfilling, as most talking heads will consider any QB who has won a Super Bowl as "elite" (see: Russell Wilson).

The complete history of the game, however, may not apply as strongly after the point at which the rules of the game are changed.

And they were, in 2004, to favor the offensive passing game.