Separate names with a comma.
Discussion in 'New York Jets' started by JethroTull, Mar 17, 2018.
If you believe in analytics, this trade was the problem. The one proven analytic is that no team can beat the draft. The fact that our GM ignored the value chart in his move up implies that our GM already ignored what is proven analytics. The one proven is that teams that pick in the top 5 pick the best player in that draft 10.3 % of the time. It gets worse as you go down the draft board. Analytics support the trade we made is really bad for the Jets. Even though are odds went up on this pick it went down when you consider what we gave up to get it.
The other problem with analytics is as the numbers and data get better the differential between players will get smaller. In fact as that happens if you happen to draft a player who is way better than the mean, it means luck played a bigger part in the pick than skill not less.
thanks 101, throughout the draft process this year, I have valued your opinions on the QB prospects as it appears you have "done your homework" more than anyone here. It has made the draft discussion more enjoyable for me. If we could pick one TGG poster to hire as next Jets GM I would put you atop my list. cheers
The studies on draft value pool the positions together. They don't consider the extra value of a QB. They also don't consider a year where there are 4 consensus top QBs, and picking at 6 all but guarantees you get none of them.
I reviewed the analysis of the data and agree with its findings as I have watched a lot of Allen and Mayfield. I don’t care what your opinion is. Clearly it doesn’t align with mine and that is fine with me.
Nothing wrong with that. We just don’t see eye to eye. I’m cool with that.
thats very true but that is what analytics is FOR. It is not supposed to factor in all the intangible stuff you are referring to. By analytics standards, the trade was a bad one but it doesn't mean the numbers will be proven right.
I can agree with that for sure, but if someone on a forum can "show their work" i'd personally accept it over a PFF
lets not forget PFF is a business, they charge for access to the data and sell data to teams i'm sure as well. they just started doing college in 2016. Now lets say an agent comes along and says here is a couple hundred grand for you to bump my player up, knowing the difference between a 2nd round and a top 5 pick is 3 mil in the 1st year alone, Maybe the person doing the stats would be willing to skew them a bit? Not saying that is what happens, but in todays NFL competitive-ness where everyone is trying to gain an edge (i mean even rumors came out that mccowns agent push cousins to the vikes so he can land mccown with the jets) where teams will record practices and pull fire alarms there is always an edge to gain. It even could be another team trying to bump up someone they don't want. NFL owners are billionaires and i'm sure could influence a site like PFF.
That was not my intention and I don't want to be the one to put you over the top. But everything you said was based on what Bowles supposedly said and really he said none of that. As a matter of fact he really didn't compare Wilson and Mayfield at all. He was answering questions by the media and was giving generic answers and there were several questions about Mayfield. Bowles offered up nothing basically. Every time someone says Mayfield is too small people bring up Wilson. That is all he did. He also brought up Fran Tarkenton.
Btw, Mayfield is essentially the same size as Steve Young. The best QB's that see the field the best understand the field the best.
There is no extra value in a bust QB. Draft position doesn't change that. In fact if you have several QB's going in the first round that makes you even more wrong. That has been born out in several multiple QB drafts. 1983 HOF QB's went at 1, 14 and 27. Outright busts went at 7 and 15. We got O'Brein at 24. This draft has 5 potential 1st round picks and 2 possible 2nd round picks. The trade up reduced our chances of getting a successful QB at the NFL level.
Mayfield's fit in the Jets offense, compared to what they ran last season:
Analytics is for getting the right answer. Which means if you want to talk about trading up for a QB, you need to make a model that factors in the extra value of a QB. Very much possible, just requires a little extra work to quantify it. Ignoring the intangibles is junk analytics and gives analytics a bad name.
There's no extra value in a bust QB, but there is extra value in an elite QB. A HOF QB is more valuable than a HOF of any other position, and you have to consider that. Several QBs going the first round only applies if you're in position to get one of them. If we stayed at 6, it isn't like we'd get whoever of Darnold/Rosen/Mayfield/Allen was left over. All of them would be gone, and we'd be left with Jackson. If you like him as much as the other 4, then you don't trade up. But most scouts rate Jackson considerably lower.
Always thought that Bates is part of this process, and this scenario is very possible. Mayfield moves around easily and pump fakes very well. He's always looking, observing, as he play actions as well. I think he see's the field really well. He's good back there.
Also, the old standards of judging the QB's and whether they're a 'prototype' is not the same anymore. It's a different league and the game is evolving.
Some stats on the evolution of the shotgun in the NFL:
Edit: There's also these stats from last year:
That said, I don't know much about Warren Sharp and would take that with a grain of salt unless confirmed by a more trustworthy source.
Just becuase they ran it, doens't mean it worked. the niners according to that ran 99% of snaps out of the shotgun, they also had the 2nd worst offense in the NFL. meanwhile ATL who had the lowest amount of snaps in shotgun, had the number 2 offense in the NFL. Only 1 team (the packers) were top 10 (8th) in offense on the most snaps in shotgun side, 2 teams in the bottom 5 ATL (2nd) and Arizona (9th) were in the top 10. If anything, I think it shows balance is the best way to a top offense. out of the top 10 offenses, 7 of them weren't in the bottom 5 or top 5 in shotgun snaps.
If you model the QB at a higher value how does that change the value of your pick? The picks you traded could be used for a QB and that value would also go up. If on the other hand you value QB's higher and they succeed at a lower rate and are drafted at a higher rate, their value may be less than other positions? I'm not sure what kind of model you're talking about?
You seem to want to use analytics to make your argument for who to pick and what position to pick but depend on luck in terms of valuing the actual draft.
If you're really interested in analytics let's go ahead and throw out team wins and player height and actually use analytics to make the argument.
What stats matter to you beside the ones that show up on the first 3 lines of a player description?
Self Preservation has a lot to do with it. McCagnan needed to make sure he secured a QB, and Mayfield gives him the best chance of the four to be ready this season (5th year senior).