DawgTalkers.net Forums The Archives 2016 NFL Season 2016 NFL Draft The 2016 Quarterback class

http://profootballtalk.nbcsports.com/201...need-some-time/

One G.M. thinks quarterback draft class might need some time

Posted by Darin Gantt on January 27, 2016, 9:06 AM EST
AMES, IA - AUGUST 30: Quarterbacks Carson Wentz #11 and quarterback Easton Stick #12 of the North Dakota State Bison celebrate with fans after defeating the Iowa State Cyclones 34-14 at Jack Trice Stadium on August 30, 2014 in Ames, Iowa. North Dakota State defeated Iowa State 34-14. (Photo by David Purdy/Getty Images)
Getty Images
Many teams will be looking for quarterbacks in the first round of this year’s draft.

But at least one evaluator who may be in that market suggested that he’s not sure there are many immediate contributors.

Bills General Manager Doug Whaley said from the Senior Bowl that he thinks most of the quarterbacks in this year’s class will be eventual contributors, as opposed to Day One starters as have come out of some recent drafts, such as Jameis Winston and Marcus Mariota last year.

“I think it’s a deep crop and I think it’s a crop that’s going to need some time,” Whaley said, via Tyler Dunne of the Buffalo News. “Now, it depends on their situation, who takes them and what they do. But I would say it’s not as top-heavy as last year where those top two guys would still be the top two guys. How quickly these guys can matriculate into the league, I think it’s going to take a little bit of time. Maybe a year, maybe a half of a year.”

North Dakota State’s Carson Wentz leads this week’s group at the Senior Bowl, while underclassmen Jared Goff and Paxton Lynch will have to be inspected later.

Whaley said he wants to study the juniors more, but said it’s “a future down the road” for this year’s quarterback class.

“The guys I’ve seen, I like them and there are guys that could be potential future starters,” he said. “I just think it’s going to be a process.”

Of course, it could also be the beginning of a process of misdirection, for a guy who has admitted he’d “absolutely” consider using a high pick on one.

I wonder if this GM would like one of these QBs to fall to him. Or does he want to tie his future to Tyrod Taylor?

Yeah cause those 18 other teams above him wait to see what he says before they set their draft board.

That wouldn't be your normal smokescreen. You would need at least a few tons of Napalm for that kinda smoke.

It doesn't hurt to put it out there.

GMs are rarely honest during this time of the year. What advantage does Whaley gain by saying this?

It settles the fans down when the guy he wants is off the board and they are trotting out Tyrod for another year.

Similar to my thinking. Misdirection.

(Taylor was surprisingly good last year.)

After looking at this, I'd be curious to see the team's records for each QB.

Despite Roethlisberger, Ryan, Sanchez and Flacco being in the second group with the lesser average stats, their teams went to the playoffs. But from that first group, I don't think any of the names listed went to the playoffs in their first season of starting... did they?

Kaepernick did and lost the Championship game.
Philip Rivers did and lost to the Patriots.

Kaepernick is an odd case as he did start the Super Bowl a year earlier. I'm sure that isn't factored in the analytics... or is it...

I get that but knowing how Cleveland is if we over reach on a QB he will just be crucified and ruined anyways because we will rush him in and not let him develope. I think Paxton will do a lot better early on because he is a really good pocket passer. I think people get distracted by his ability to run it well but man this kid can throw it really, really well. I think he is a stronger and more physical Mariotta.

Maybe it just means that the QBs that started in year 1 were starting out of necessity because the team that took them was that much worse and had no better option... thus the lower QB rating..

But what I get from that list is that the % of those who started early and those who started later that went on to have long term success, isn't that much different...

This stuff is all bunk.

1, how does anybody know that those qb's who were successful after sitting a year would not have been just as successful or even more successful had they started right away.

2, how does anybody know that those qb's who started immediately and failed would not have failed even if they had sat a year?

3, and what about those qb's whostarted right away and succeded? Had the sat a year, perhaps they would have even more success.

Understandable points, Jester. I do not know the answers, but "The Truth is out there" as they say in the X-Files. But as an additional thought for your list, how many who get started or play first year have careers messed up by injury as well as failure.

Take the best QB when available and get him some hands on a big body. Smurf ball doesn't convince me you are competitive. And we need some thumper monsters in the middle and on the edges.

Lots to consider.

My thinking as well. The Bills are at 19. If you didn't want one, I would talk them up so a few more players I might want fell to my position.

This is the time of year you can't listen to these guys. They are doing what they do, try to put their team in the best position possible.

I mean, why would this guy try to talk us out of taking a QB if he didn't like them and he wanted a non-qb player player? I think he hopes one falls far enough he can trade up at a cheaper price to get one.

Not sure it's all bunk but I agree that you can find examples to fit whatever narrative you want to push.. maybe if we draft a QB, let him flounder, send him to Canada or some semi-pro developmental league, get him a job at Piggly Wiggly, then bring him back... 11 years from now we'll have Kurt Warner...

In the end, the QB has to have the physical and mental skills.. then they need to be with a decent staff that can teach and be patient, the system has to fit the skills.. there are a ton of variables.... I assume some could benefit from waiting and some are ready to go from day 1... so that is just one of many variables as to whether a QB succeeds or not.

The law of large numbers. The more QBs you measure, the more accurate the fact that sitting at least helped them start better.

This article doesn't say it necessarily makes them a better QB at the end, just that they start better, which makes sense because they get the fundamentals down and learn the offense. Notice how while their QBR is higher in the first year, it only goes up a little in the second. The 2nd group has a bigger jump in QBR in their second year, but still hasn't caught up because they are having to learn as they go some of the things the first group already learned while sitting.

I just dont like dealing with all this talk as "fact"

Every player is different.

I dont think you can just say every QB should or shouldn't sit..

Neither am I.

Luck shouldn't have sat. Manning shouldn't have sat. It was pretty to see those guys were ready to start right away. You really can't see that with these guys though.

But the comparison is not equivalent. They are comparing a qb's 1st and 2nd year to another's 2nd and 3rd.

The article is perfect and you shouldn't question it.

Even their 2nd years aren't as good as the other groups first years.

Right, if you can't figure out way to attack it's validity use sarcasm to try and shoot it down. That always works.

Why ten starts? Isn't that an arbitrary number? (Yes.)



Inconclusive.

Asking me why someone else picked the number that they did is a loaded question but...

First I notice that both you and CHS use the number 10 religiously instead of the number 5. While that does sound better in your attacks, it is actually 5 games = a season.

https://en.wikipedia.org/wiki/Sample_size_determination


https://en.wikipedia.org/wiki/Statistical_power



So lets look at what we are comparing. QBR

QBR is more accurate the more snaps a QB takes. If we looked at the QBR for a QB on his first pass only, it could look entirely different than the QBR for the entire first quarter, which would look entirely different than the entire first game, first season, or career.

Generally speaking though, the more data that we get, the more accurate of a description we get of that QB's performance.

http://www.nfl.com/stats/categorystats?t...s=PASSING_YARDS

Lets take a look at a few QBRs there were 3 perfect QBRs this year at 158.3


Antonio Andrews TEN RB 1 for 1 41yds TD
Ryan Nassib NYG QB 5 for 5 68yds TD
Cecil Shorts HOU WR 1 for 1 21yds TD

So if we count these guys, all three are better QB's than Drew Brees, Aaron Rodgers, and Tom Brady as far as QBR goes.

Using the same logic, there were also 3 1 pass wonders that had a QBR of 0. All of them were WR's

Obviously, counting everyone that threw the ball is probably not the best idea unless you think that Cecil Shorts is a really good QB.

So maybe we should make a cutoff at a game.

http://espn.go.com/nfl/player/gamelog/_/id/1428/peyton-manning

Peyton Manning
Week 3 vs Detroit QBR 101.7
Week 9 vs Chiefs QBR 0.0

So if we look at just one game what kind of year did Peyton have? If we set it at two games and take the average, are we certain that his QBR is 50.85 QBR is accurate and either the first game or the second game isn't a fluke?

Ok so it looks like at least 3 games should be included, but what if the QB plays two really strong opponents or two really weak opponents in that 3 game stretch? Are we at a fair comparison?

Also how do we know that a QB played a complete game? There really isn't a stat for that in the NFL. I suppose you could look to see if anyone else threw a pass, but what about the teams that run the wildcat for a series or what if he took a hard sack and sat out for a play or two?

Jimmy Garropolo only had 4 attempted passes, but played 3 games. Did he have a season that should be figured into the math?

Drew Stanton played in 7 games but only had 25 total attempts. He would be included in this study, but should he be?

The main point of your argument is that all these guys were thrown out of the study and that slanted the results toward sitting a QB.

What you ignore is that the outliers (both good and bad) are thrown out. For every perfect QBR thrown out, there was also a zero QBR. For this particular year it seems to be equal amounts, 3 and 3. That is most likely not the case most years, but I'd be willing to bet that the average of each is pretty close.

What I don't like is that in my little examination there was at least one QB that I would consider an outlier still included. Drew Stanton played 7 games, but only had 25 passes. I don't think personally that is a big enough sample size to be relevant.

So that brings us back to the total sample size.

A total of 79 QBs over 15 seasons means we are still looking at about 5.267 QBs for every season.

I can't think of a draft that had 5 meaningful QB's in it, can you?

Would adding another 20 or 30 QBs that never put together any production help this study or hurt it?



I don't know about you, but the first column in that chart sure looks like a bell curve to me. At least the sample that we have looks like what we would expect it to look like.

So is this perfect? I think the answer to that is probably a No. We could get more QB's in the sample by going back even more years, but how far do we go back before we get a shift the complexity of the game? I think in the last 15 years, it's probably a stretch to say that there were 80 relevant QB's, but weeding this out any more would give a sample size that would be too little to trust. As it stands, any one QB is 1.25% of the sample. That seems trust-able. While I'd like to get the Drew Stantons of the world out of the study and take it even farther above your pain threshold cfrs15, I wouldn't want to see one QB become 3% or 4% of the sample size. Then you start to get one QB having too much influence on the study.

Although a 5 game season might be a little low for my taste, it does seem to be a sweet spot and certainly not arbitrary.

For those of us that are upset that it doesn't say what we want it to say though, I think any process would be called arbitrary.

Studies can please some of the people, some of the time.

Did you write that article?

not worth

Good play going back to the sarcasm.

If you can't attack the post, attack the poster.

QBR and QB Rating are not the same thing. I know what you are talking about, but it is a little confusing.

Drop the outlier?

http://www.theanalysisfactor.com/outliers-to-drop-or-not-to-drop/

That was a very well written post. It sounds very professional. If you wrote it then kudos to your writing ability. If not then admit it. Either way I don't care. It was well written and informative. That's all that matters to me.

I did actually think about that towards the end of my post, but I didn't want to filter through it again and figured most people would get my meaning.

tab, sorry for the headache.

The post was me, the article was most definitely not me.

1) If it is obvious that the outlier is due to incorrectly entered or measured data, you should drop the outlier:

This also applies to a situation in which you know the datum did not accurately measure what you intended. For example, if you are testing people’s reaction times to an event, but you saw that the participant is not paying attention and randomly hitting the response key, you know it is not an accurate measurement.

This is exactly why you set the 5 game minimum. To have enough snaps to accurately measure QBR.

How come five games is enough to accurately measure the QB rating? Why is QB rating an accurate measure?

I would argue that the number would need to much greater than five games to get an accurate sample.

(This is my last comment on this subject. It is obvious we are getting nowhere and I don't actually care enough to continue the conversation at the level I feel it requires.)

It's not arbitrary, it's beneficial. When you use QBR as the only way to calculate how good a QB is then you need to have somewhat of a sample size. However, when you do that, you also disqualify QB's, who have sat and learned, but never started in the NFL because they were deemed to be backups. It seems disingenuous when they're completely removed from the study.

That was actually probably your best comment on the subject.

I'll start with why using Quarterback rating is the measure.

It's the best we got right now.

I would argue as well that more than 5 games would be a better measurement. But more games being measured equals less QB's in the sample size. When one QB starts to get much more than his current influence on the study than his current 1.25%, the study starts to become tainted by sample size.

You also remove QB's that started a game or two in their first season, then sat and learned but never started in the NFL because they were deemed to be bad draft choices.

What it comes down to is did they get a shot whether it's in the first season or after that. The QB that was forced in first season or second season for a couple of games because of injury didn't get that shot. That is because he wasn't good enough.

What you really want to do is to include all the bad draft picks with the guys that sat a year. What you forget is that you start including guys like Johnny Manziel in the first year when you do that. What was his QB rating again?