Miami Heralds Greg Cote: The Dolphins Should Find A QB They Believe In

first of all, we aren't getting cousins so no need to even talk about him. Cousins will get the biggest contract in probably NFL history, miami doesn't have the cap space for that. 2nd yes miami needs to think about the future at QB whether it's the first round 2nd 3rd whatever. I rather have a guy they trust as a back-up than have to deal with what happened this past year. If tannehill lights it up then great, we now have a position of depth and strength and hopefully the FO makes the right move with it. If tannehill doesn't light it up, then great we have the future being groomed. win/win.

 

I don't think I do. I mean if i were a betting man which I am and someone made me bet, gun to my head I would say no. I'm not sure how I would feel about Tannehill succeeding with the Pats offensive weapons i mean I know they got gronk but he's only been apart of 1 superbowl so far because of injuries so he may not be a factor in the playoffs/super bowl.

 

Still BS.

Travis is not writing technical manuals. Hyperbole is part and parcel to sports writing.

And again, the problem isn't that someone disagrees. It is the reason someone disagrees then they don't back up their claim as to why they disagree. If a person doesn't think Thill runs like a RB because of things they see in Thill running, then fine, present the evidence specific to Thill. Arguing with Travis because a person just thinks he's biased because his stance disagrees with their bias, is the problem.

 

Whether you agree with them or not you'll learn far more from the Aquas of the world than the Greg Cotes or Omars and we should be glad that they're willing to share their content in the mains.

 

They should ask Patrick Chung if he thinks Tannehill runs like a running back.

 

About 10 years ago, Michael Vick returned to the league and had an away game vs Atlanta (his former team). One of the news outlets I occasionally wrote for asked me to cover the story, so they sent me the press credentials and I made the 4 hour drive south to Atlanta's stadium. Vick didn't play particularly good that day but from the moment he walked into the stadium, applause broke out like I've never, ever heard for an opposing player. They delayed the kickoff almost ten minutes because the arena was in such a frenzy with their golden boy returning home, and the stadium would explode every time Vick completed a pass. It was insane....were they rooting for their own team or for Philly?

Anyway, I knocked out my little 1,000 word article and labeled it- "Michael Vick- The Canine Killer's Sweet Revenge". Now, I will admit that I was going for a little click-bait because I knew the title would piss some people off. The story itself was about how he turned around his life, did massive charity work, lost everything he had and finally scraped his way back into the NFL...only to be greeted in Atlanta by massive respect and praise. I also went into how his dog-fighting days were deplorable and he honestly didn't deserve a 2nd chance, but none of that mattered to the readers. It got ugly fast.

Here's what happened- Yahoo and Google both picked up the article and I had millions of views within the first day....which meant that all segments of the US population were reading it....including animal rights activists. And just like that, these folks were writing opinion pieces on their websites absolutely trashing me for supporting Vick and accepting that he had some success. Now, most of what they were posting had nothing to do with what I wrote- it was the title alone that drove them into a frenzy. By the end of that first week, I was approaching 80 million views and the hate was coming from everywhere around the globe.

So there I am trying to defend myself against the troll horde saying, "But I love dogs and what he did was deplorable...that's not even what the article was about!" But I dug my own internet grave with that title; I bet 1 in 50 didn't read past the first paragraph. I was literally receiving death threats and I had internet stalkers coming out of the wood-works. I'm guessing I was one of the most viral journalists in the world that week, but do you know what? I was back to being an anonymous nobody about 10 days later. A few of the PETA folks continued to comment on my other work for a few weeks but even they got bored and gave up.

Here's the thing you have to accept as a writer though- do you think my employer was mad at me? Hell no...he offered me a full time staff position! I got him more hits in seven days than all his other articles combined delivered in six months, plus the ad revenue was literally through the roof. That article earned hundreds of thousands in ad revenue and picked up just as many bookmarks- that traffic is worth its weight in gold in the digital age. 80M+ hits can't be wrong.

And guess what else happened? Job offers started coming in from all over the place- mainly from internet marketing firms. That's really the week I stopped being a writer and started on the path that got me to where I am today- I became a marketer then a copywriter then an SEO and a digital marketing specialist. That one stupid article put me on a trajectory for a strong six-figure income, my own digital marketing firm and countless Fortune 500 clients.

Another quick point- all those angry people who were quoting me and linking to my profile...that was insanely powerful backlink juice that any company in America would kill to have. All of my profiles instantly ranked better by association and in turn, all my clients sites did better if my name was in the byline. Those haters did me a massive favor without even realizing it and every ugly comment was telling Google it should pay more attention to anything that I write.

Looking back, that week after the article dropped was painful- but it's a step every writer has to take to reach the next level of their career. All you can do is write from your heart and pick subjects that others are passionate about- it makes no difference if they're "on your side" or not. If the work sucks, then nobody reads it and they certainly don't comment...so take EVERY interaction as a compliment. If people love you, then great! If they hate you, then it's just as good! Regardless, it means that you're delivering quality and you're headed on the right path.

 

I’m not a tape reviewer like aqua or CK. I like to play with stats.
Leaving aside the issues of performance of the OL, D or other factors I did some number crunching a 2 years back, which I can’t find the link to.
Basically Mike Shernabks offense was so out of date it was worth about a -10 penalty to a QB’s passer rating.
Lazor’s playcalling fell apart when the team fell behind, which was worth about a -5 penalty to Tannehill’s passer rating while he was OC.

I’m reasonably happy that post-rookie Tannehill would be about a 95 passer rating QB, even with all the other problems he’s faced, if he had a competent OC. Now its up to others to do a film study to say ifthat is Tannehill’s ceiling or if he has potential to do better. But I do note that the people who do deep dives into the film almost always come away saying breakdowns on other parts of the team are hurting Tannehill’s record of production.

That’s why I firmly believe that Belichek would be able to win SBs if RT17 was his QB.

 

Matt Cassell couldn't even make the playoffs with them.

 

If Tannehill post rookie was a 95 passer rating QB with a competent OC. Than your saying he's in the same company as Manning, Brees, Young, Rodgers, Brady, Romo, and Wilson because those are the only QB's in NFL history with a QB rating with 95 or higher for their career. 5 of those 7 are HOF/future HOF QB's, with Wilson being a potential 6 depending how his career plays out. That would make Tannehill a certifiable elite QB.

I think Tannehill can be a good QB with more consistency, but I will never be convinced he is ever going to be an elite QB outside of him proving it on the field. Even in his breakout 2016 playoff season he was on track to throw for about 23-24 td's which is his median the last 4 years. He was also on track to throw 14-15 ints which would have been a career high. Those numbers aren't great.

Giving it a second though though I'm even going to change my stance and say Tannehill could win a super bowl under Belichick but only if the defense doesn't give up more than 17 points like in 2001. So with an elite defense that gives up no more than 17 points throughout the playoffs yes Tannehill probably wins a Superbowl with Bill.

 

95 rating isnt elite anymore, just good. Passer rating have been going up over the last 15 years.
Elite is 105 now.

 

This is a great example of why it is absolutely necessary to adjust stats to a common year when comparing across eras.

Since you're new here let me just go through the steps. First of all, note how league average passer rating has changed over time, from the 60's in the 1970's to the mid-70's in the 1990's to mid-80's today:
https://www.pro-football-reference.com/years/NFL/passing.htm

If you don't adjust for passer rating inflation you get crazy results like Tannehill = Marino because Tannehill so far has a combined 86.5 rating and Marino had a combined 86.4 rating! And no one in their right mind thinks those two QB's belong anywhere near each other.

So how to adjust?

Let's do this by example. First let's adjust Brady's 86.5 rating in 2001 to the year 2017.
https://www.pro-football-reference.com/players/B/BradTo00.htm

All you need to do is divide 86.5 by the league average PR in 2001 which is 76.6 and then multiply it by the league average in 2017 which is 85.1. So Brady's 2001 passer rating adjusted to 2017 is 86.5/76.6*85.1 = 96.1.

If you want to combine passer ratings across years, then you first adjust each year the way I just described to the reference year (2017) and then compute what's called the weighted average, where the weight is passing attempts in that year because passer rating is an efficiency stats based on passing attempts (you don't want to overweight passer ratings from years where the QB had very few attempts).

The way to compute a weighted average is to multiply each passer rating by its weight (attempts in that year), then sum those numbers, and then divide by the sum of all the weights (sum of all passing attempts).

For example, let's say we want to combine Brady's 2017-adjusted passer ratings for the years 2001 and 2002. In 2001 Brady had 413 passing attempts with an adjusted passer rating of 96.1. In 2002 Brady had 601 passing attempts with an adjusted passer rating of 85.7/78.6*85.1 = 92.8. The weighted average is:

(96.1*413 + 92.8*601)/(413+601) = 94.14

When you do that for Brady's entire career his 2017-adjusted passer rating is 101.04, which already shows it's not the "mid-90's" career passer rating a modern day QB should have to compare to Brady. For the other QB's on your list you get career 2017-adjusted ratings of 101.29 for Peyton, 99.5 for Brees, 99.91 for Romo, 97.6 for Wilson, 104.81 for Rodgers and 109.98 for Young. Oh, and you get 99.63 for Marino.

What about Tannehill? His 2017-adjusted career passer rating is 85.28, and if you exclude his rookie year it's 87.08.

As you can see those numbers make more sense because Tannehill isn't Marino or Peyton Manning by any stretch. Basically the threshold in today's game for an "elite" QB is around 100 career passer rating.

 

If only 7 players in NFL history have a career avg of 95 qb rating or above and 5 of those 7 players are HOF/future HOF players. I would very much say you are in elite territory. But at the same time having a really high QB rating is not really the only metric I would use. TD's , Int's, y/a, how you play against playoff caliber teams regardless if you win or lose, personal accolades like pro bowls, all-pros, MVPs, all decade teams, records, etc.

I understand players like Kirk Cousins have had a couple of seasons above a 95 QB rating, but the big difference is those other QB's did it over an extended period of time (which is what makes great players great) while a player like Cousins has gradually gone down the last 3 years. Even in the last 6 years there are only 9 QB's with a minimum of 1000 attempts who have been able to maintain a 95 or higher QB rating. 5 of which are HOF, 1 of which is an MVP winning/caliber QB. Achieving a 95 qb rating for a couple seasons is definitely easier nowadays but I think keeping it above 95 over 10+ years is still plenty hard to do and still an accomplishment as of right now, as its only been done 7 times. As of 2014 only Brees and Brady have had a QB rating of 95 or above each consecutive year, thats how hard it is to accomplish yearly.

But like i said the whole picture matters as well so I would never rely solely on QB rating.

 

http://dailydolphin.blog.palmbeachp...white-the-right-qb-choice-for-miami-dolphins/

 

The passer rating stat has steadily increasing by about 0.7 points a year since 2002.
The NFL average QB has roughly a 90 rating. So a QB who is 95 in 2017 is like an 85 rated QB in 2002.

As Cbrad says we’ve been adjusting stats to a base year on this site so we can fairly compare modern performance with historical performance, over a 3 or 4 year period the inflation does not make that big a difference but when you go 5 years or longer it starts making a noticeable difference.

 

Cbrad, if we adjust based on numbers of attempts, shouldn’t that also be set to the base year too? Pass attempts per year have gone from the about 400 range to around 600. Otherwise later years are getting over weighted

 

Passer rating is a "per attempt" efficiency stat and is naturally biased towards games or seasons with more attempts than others when you combine.

For example, when the NFL (or pro-football-reference) reports Brady had a 102.8 rating for 2017 they're first adding up all the component stats of passer rating across all games in 2017: completions, attempts, yards, TD's and INT's, and then plugging those combined totals into the formula as if all those stats came from a single game.

So if you have two games, one with 40 passing attempts and another with 30, the formula (or its use) will naturally weight those two games 4:3. In other words, we WANT to weight later years for a given QB by more attempts if the goal is to remain consistent with how they currently do things.

It's also worth keeping in mind that this bias isn't affecting the comparison across eras per se, but is instead weighting later years of every QB more than his earlier years. QB's in an era with 550 passing attempts aren't being weighted more than those in an era with 400 passing attempts. But if passing attempts changed during a QB's career, then those years with more passing attempts will be weighted more.

EDIT:
I thought about this a little more and there's a more technical reason for not adjusting by differences in attempts across eras.

The goal is to remove the effect of any and all factors (including passing attempts) on passer rating across eras or years. And the way to remove the effect on passer rating of all factors is to adjust passer rating directly.

If we adjusted by passing attempts after adjusting passer rating, we would not only NOT be adjusting for the effect on passer rating, but we'd be double-adjusting by attempts. So combine that with the argument above by how passer rating naturally weights by attempts and I think we're doing the right thing by reporting the weighted average of adjusted PR (the weighted average when the weight is by "attempts" gives the same natural weight to PR in different years as the formula itself does).

 

Travis does have a huge bias in regards to RT, always has...and that's fine. He's a Dolphins fan.

But even when some of it makes me chuckle, it is flat out BULLJIVE to dismiss the time he puts in to watch games and come to conclusions on multiple player's (no not just RT)

I suffered through watching 5 games on the coaches tape and even condensed it took over half my day. All to just come back and get some idiotic responses from folks who nitpicked one or two comments. That may have been the first time in my 11 years at this site I was actually pissed off as opposed to chuckling at the usual trolls. And why? Because it is/was very time consuming and not something I wanted to really do, but was asked.

It didn't frustrate me anyone disagreed, that is and should be expected if it's me Travis or Peyton Manning discussing football. But the way some people attack others for at least putting in that time is beyond petty and reflects horrible on that person as a poster.

We gotta do better round here..

 

Oh so adjusting the QB rating is a site wide agreement. That seems like it would be impossible to pull off considering the different personalities. Kudos to the unification.

 

Well not all the site, just us stat monkeys.

 

He won 11 games coming off of the bench and not starting a game since high school. Winning 11 games and not making the playoffs is a rarity.

 

I saw all that a long time ago.

 

I've stated it, and it has been stated over and over and over by the non fan of Tannehill group. His shortcomings have been well documented on this board several times, don't really see a need to keep posting the same things. Of course many were run off the board for such statements. I'm glad you see what others saw early on.

 

How many times will you use that same lame LIE!....How many? People got banned because they did not follow rules, not for their opinion. It has become embarrassing that you keep saying it. Please stop, because it is incredibly disingenuous. And it is not the first time you have been corrected about it.....and you know it

 

So long as the dink and dunk is the preferrd method of moving a football down the field a 95 rating isn' much to brag about. In a few Years we'll probably see defenses force longer passes and take away the short game. I expect qb ratings to go down when that happens

 

Just how will that of Tannehill can recover it is up for grabs.

 

I also think league average passer rating has plateaued (4 consecutive vowels in that word!!).

It's been relatively stable the past 5 years and the game is already "offense friendly" enough IMO that the league should see no need to keep systematically changing or reinterpreting the rules to make defenses even less relevant. We'll see.. but I expect the current mid-80's (max 90) average rating to hold for quite some time.

 

Combining the last ten years (2007 to 2017) the top 10 y/a leaders have a median of 7.77 y/a.
From 1990 to 1999 the top 10 y/a leaders have a median of 7.50
From 1980 to 1989 the top 10 y/a leaders have a median of 7.81

The difference between the 80's and 07-17 is 0.04.

 

Completion % has gone up though, iirc from about 55% to about 65% in the same period.
So by simple maths the average pass completion must have gotten shorter to keep the average yards per attempt the same.

But since iterception% has also come down a long way it shows that the ‘dink and dunk’ is more effective than heroically lobbing it deep downfield every other down,

Edit
Yards per catch fell from 12.6 or so in the mid 80s to high 11s, low 12s in the mid 90s and has plateaued at 11.5 - 11.3 range in the last 10 years or so.

 

Just did a little bit of allok at some numbers.

From the 1980s to early 2000s teams averaged low 500s for pass attempts per year,
In the last 4 years the pass attempts per team have been, 558, 571, 571 and 546. So an average season with an average number of attempts would now be weighted about 10% more heavily than a season from 10 years ago if we use raw snap numbers.

My thought is that if we are trying to equalize for era we should do so for snap mumbers as well as passer rating.

 

Y/A and Y/C are 2 different things. Marino's 1984 season he had 14.0 y/c but 9.0 y/a.

 

The difference between ‘dink and dunk’ and ‘hero ball’ is how they get to the average. Both types of play can have the same ypa, but one can get there with high completion% and low ypc and the other can get there with low completion% and hugh ypc. What I was getting at is that quoting ypa alone doesn’t prove or disprove ‘dink and dunk’ or ‘hero ball”. When we look at ypc and completion % you can see that although ypa has remained about the same, ypc dropped about a full yard.

 

Yeah.. it's definitely a VERY good question Pauly. And let's debate this thoroughly if necessary. But I still stand by the argument I gave in the post you quoted. Let's start with something we probably agree on: the purpose of adjusting stats is to remove the effect on a stat of any changes in the nature of the game between two different years.

If the goal is to remove the effect on passer rating of not just different average passing attempts but really everything else that changed between two different years, then all you have to do is to look at how passer rating itself changed between those two years (because that directly shows the effect). Once you've made that adjustment, you're done!

That is.. the effect of different average passing attempts is already incorporated into the adjusted passer rating. You shouldn't adjust for attempts TWICE by adjusting again for average passing attempts afterwards.

Point is, passer rating is a "per attempt" efficiency stat that naturally weights games or seasons with more passing attempts MORE than those with fewer passing attempts. Combine two games, one with 40 passing attempts and another with 30, and passer rating naturally weights those 4/7 and 3/7. You wouldn't suggest we try to remove that "bias" would you? Because after adjusting passer rating, that's the only bias left. That is.. if you pass less often, then it should count less in calculating "efficiency" than if you pass more.

 

From 2007 to 2016 the median y/c was 11.52
From 1990 to 1999 the median y/c was 11.87 a difference of 0.35.

In that time frame QB ratings have gone up 5-10 points peaking at 88.4. Whats your theory on that?

 

No theory, it’s interception % and completion%.

When the passer rating was introduced 4 key components were identified, completion%, ypa, TD% and interception%.
When you plug in 1970s stats into the passer rating formula each of the 4 components is equally weighted.
When you put 2010s stats into the formula completion% and int% are roughly weighted the double of TD% and ypa.

The interesting thing though is that team passer rating has held as continuing to have a high degree of correlation to win% even though the Effective weightings have changed.

 

I was a little surprised because with all the talk of up tempo offenses and the NFL becoming a passing league I thought that there would have been a bigger difference between then and now.

I am very persuaded by the ‘you don’t keep adjusting’ argument. I’ve had situations where I’ve had to reconstruct databases from the original files because people kept adjusting the adjustments until it wasn’t actually a record anymore but the opinion of the people doing the adjusting.

I would be more inclined to think it should be discussed further if we were going from an average of 400 attempts per year to 600 attempts per year. I have averaged by games started when I did adjustments in the past, but I do think attempts is a better and more complete stat than games started.

 

Don't be upset with me that it took you years to see what I stated when I first saw him play. Not just me, but several others on the board too.

 

How many more TD's would Marino have thrown with that average?

 

It’s about another 10% more, but only if he started his career in the last 5 years.

 

Actually.. it should be MORE obvious that we should NOT adjust twice if the discrepancy is huge. Because what's really going on here (the true technical argument) is that you're equating different sample sizes with different variances when you should treat two samples as a single one.

Try this out on Y/A instead of passer rating. The logic is the same because all passer rating does is sum 4 such efficiency stats (comp%, Y/A, TD%, INT%). So what is true for any component will be true for the entire formula.

Suppose we have two years, one where you only had 100 total passing attempts, and another where you had 1000 total passing attempts, possibly because of a strike-shortened season (which actually occurred in the NFL, but not with that big a discrepancy in passing attempts).

Would you seriously consider weighting Y/A in the year with 100 total passing attempts equally to the year when you had 1000 total passing attempts? If you do, you're equating variances you shouldn't be equating (larger variance for the smaller sample), and obviously you have better information on how the QB is when he passed 1000 times than 100 times.

It's actually a great way of seeing which approach is better in many mathematical contexts: look at extreme cases. So if equating makes "logical sense" when sample sizes are similar, then it should also make sense if in one year you had precisely 1 passing attempt and in another year you had 10,000. Clearly it doesn't.

 

I have to say, my gripes of Ryan initially included those things.

Luckily for our sake, he has overcame many of them.