Ja'Wuan James

[jim1]

My response is simple…I just do not think Martin is there at 19.

That could very well be, because he's that good. His looks deceive, in terms of athletic builds he's somewhat in in the John Kruk mode, but that dude can play. I would also take Koundjio over James, provided that his knee did in fact check out, and not think twice about it. People talk about the Dolphins liking Koundjio at #19 as smoke, I'm wondering the same about James. Good player, but I just do not see a first round pick there.

 

[ckparrothead]

Most of the guys you listed were drafted (or signed as college free agents) when fullbacks were still alive. Dotson and Waddle are exceptions to that. You could make the same argument that because Tom Brady, Drew Brees, and Russell Wilson were picked in the 6th, 2nd, and 3rd round respectively that you can find elite quarterbacks outside of the first round.

This is exactly what I'm talking about though. Anyone that truly understands the statistics would know not to make that argument. But a person who doesn't really have a grounding in them and isn't aware of their limitations especially in either scientific or financial settings...may end up making that kind of argument. But they'd be doing so a bit blindly, IMO.

Or just because Arian Foster went undrafted, you shouldn't take Adrian Peterson or Marshawn Lynch in the first round.

Which would be more examples of the incorrect application of statistics, and most likely would be made by someone who doesn't really have a good background and understanding of their uses and the limitations associated with things like sample size and confirmation bias.

The RT position is definitely on the rise in the league. It's a good point though. I don't think there is any arguing the 5 I listed are in the top 10. Also, I think Austin Howard and Joe Barksdale are "JAGs" and not up to par with the others you listed despite what PFF says.

I listed Howard mostly because he was actually getting play on the market and received a pretty good contract for a right tackle. He signed a 5 year, $30 million contract.

DJ Fluker was looking good at RT. Lane Johnson played really good as well, especially towards the end of the season. There will always be exceptions to every rule, but that doesn't mean you should change the rule. I think if you are looking for a good right tackle, you should now look in the first 2 rounds. It's a passing league, and that's the way the trend is going now.

I'm not sure what you're arguing, as it relates to what I was saying...Are you disagreeing? Or agreeing?

But you made an excellent point in your last paragraph. For whatever reason my messaging was turned off on here but I'm going to change that. I've been working on some stuff in regards to that. I'll shoot you a message in the next couple days with some stuff I've been working on.

Interesting. I'd love to hear if you have any background in this stuff. I firmly believe that successful integration of analytics into the scouting process will help bring scouting success rates to a new level (which would be reflected in better league-wide correlations between draft position and player performance).

 

[roy_miami]

He's going to check out 10 out of 10 off the field as far as his character is concerned.

I've had my eye on the guy for years.

http://www.finheaven.com/showthread...eniors&p=1064583006&viewfull=1-post1064583006



Going back to the 2012 tape and evaluating guys like Dallas Thomas, Zach Fulton, Antonio Richardson and JaWuan James it was always clear to me which of the four players had the brightest pro future.

He's big at 6'6" and usually runs in the 320 lbs range during the season I think. He's one of those naturally big guys too, has an 82 inch (6'10") wing span. Unusual agility for the kind of player.

I wish I'd gotten more chance to see him against the likes of Jadaveon Clowney. I'll have to settle for the likes of Devin Taylor, Jarvis Jones, Cornelius Washington, Dee Ford, Michael Sam and Dante Fowler. Most of those guys he absolutely shut down. Some of like like Devin Taylor, Dee Ford and Michael Sam looked borderline undraftable in the games they played against JaWuan James.

He's not unbeatable. Markus Golden (whom I've got tabbed for watching closely for 2015) got him on a nice speed to power move, sacked the quarterback. I think Ronald Powell was able to cut inside of him one time though he didn't even end up hurrying the quarterback if I'm not mistaken. Dante Fowler got a sack but it wasn't exactly a fair fight as the play saw James down blocking initially before having to redirect back out to try and pick up Fowler who had blitzed from a linebacker position.

There's an added benefit with James that if for some reason you don't end up liking him at tackle he could still perhaps move inside to right guard and be viable. It's a chance, a projection, but what is on the tape that makes you think the projection worthwhile is enough to offer a little added slice of value onto JaWuan James draft stock. He has the makings of being a versatile player.

Right tackle is not a well understood position when it comes to the normal conventions of evaluation, in my opinion. If you were to regress PFF grades against draft position you will find that the correlation between draft position and grades is only half as strong among right tackles as it is left tackles. What does that mean? It means when it comes to a right tackle a guy drafted in the 6th round way too commonly outplays the guy drafted in the 3rd round...which means the NFL isn't getting it right. Left tackle is much more orderly.

This is the primary reason a right tackle is not considered a position you should be drafting high. It's not because the position itself lacks value. Try telling your coach the position lacks value when you just lost because Tyson Clabo couldn't come even close to blocking Mario Williams during a critical portion of the game. There's too much chaos to noise in the evaluation and so the 6th and 7th rounders often outperform the 2nd and 3rd rounders. That's the reason it's considered to be a position you don't need to draft in the 1st round. If the NFL actually were getting it right on the right tackles, and the talents were being sorted into a more orderly fashion, suddenly you'd need to draft one early to get one.

Knowing the conventions aren't getting it right, my gut tells me to reduce things down to basics. I want guys I've seen playing there. Guys I've seen play in a lot of games. Guys I've seen play talented players. Guys that have the standard physical measurements (i.e. fit the prototype). This is why if I have an immediate need at right tackle, to hell if I'll draft a Zack Martin and roll the dice that way. I want guys I've seen do it multiple years like Jake Matthews, Morgan Moses or JaWuan James.

Would I take JaWuan James at 19? Ultimately I intend to do better than filling immediate needs. But then, I wouldn't have walked into the draft NEEDING to think about forcing a JaWuan James pick, either. Ideally you want to have viable options at every position before the draft so that you're free to pick the best value.

If you truly believe Ryan Tannehill is a star in the making, that he and Mike Wallace are going to get their chemistry and that this team is on the right track...I think absolutely you go ahead and fill your immediate need with guys like Morgan Moses or JaWuan James at 19 overall. You do that because allowing a HOLE in the roster to persist can prevent your star players from being star players. One of the most commonly cited rules of roster building is that you don't have to be great or even good everywhere, but you can't be bad anywhere.

This is a great post.

The only thing I would add is while it may appear that RT is a gaping hole I believe hickey sees some, if not all of the guys still on the market as viable options, so even though McKinnie isn't officially signed as a Dolphin I would guess Hickey has told his agent to speak with us before he inks any deal (not saying it has to be McKinnie just using his name as an example). It might be wishful thinking but if I'm right I love the way he's handled pre-draft free agency.

 

[futurescout]

This is exactly what I'm talking about though. Anyone that truly understands the statistics would know not to make that argument. But a person who doesn't really have a grounding in them and isn't aware of their limitations especially in either scientific or financial settings...may end up making that kind of argument. But they'd be doing so a bit blindly, IMO.



Which would be more examples of the incorrect application of statistics, and most likely would be made by someone who doesn't really have a good background and understanding of their uses and the limitations associated with things like sample size and confirmation bias.



I listed Howard mostly because he was actually getting play on the market and received a pretty good contract for a right tackle. He signed a 5 year, $30 million contract.



I'm not sure what you're arguing, as it relates to what I was saying...Are you disagreeing? Or agreeing?



Interesting. I'd love to hear if you have any background in this stuff. I firmly believe that successful integration of analytics into the scouting process will help bring scouting success rates to a new level (which would be reflected in better league-wide correlations between draft position and player performance).[/QUOTE]

I took it as if you were arguing with me, when in fact, we are on the same page.

I've been working on things like true position value, measurables for elite prospects (basically prototypes if you will), current trends of where elite-good starters are being drafted in recent years, philosophies in free agency. Things like that in regard to roster building. I wouldn't consider this "analytics" I mean it is, but not some ridiculous formulas or anything like that. I'm not a numbers guy (barely got through thru math in college) more along the lines of studying current trends. At the end of the day, film trumps everything. But I think I have found some interesting theories, and have plenty of examples to support them. That's the key IMO. Is it logical, and is there evidence to support your position? I cover those two areas.

 

[ckparrothead]

I've been working on things like true position value, measurables for elite prospects (basically prototypes if you will), current trends of where elite-good starters are being drafted in recent years, philosophies in free agency. Things like that in regard to roster building. I wouldn't consider this "analytics" I mean it is, but not some ridiculous formulas or anything like that. I'm not a numbers guy (barely got through thru math in college) more along the lines of studying current trends. At the end of the day, film trumps everything. But I think I have found some interesting theories, and have plenty of examples to support them. That's the key IMO. Is it logical, and is there evidence to support your position? I cover those two areas.

I have to warn you though...much of the reason the NFL has likely resisted the heavy integration of analytics into the process is because as soon as they get a real statistics expert in the office that person is more likely to spend his days saying "no" than saying "yes". A person with a firm grounding in statistics would tend to lament the insufficient sample sizes and lack of statistical significance involved in any study involving football-related data sets. Some of the stuff you're asking about (e.g. 'recent trends') strikes me as more likely to fall into traps than to be meaningful, most often due to insufficient sample size. I think once you get beyond that barrier if a statistics guy really begins to understand the sport and the drivers of success then perhaps he can start to figure out more valuable things that analytics actually can offer the sport instead of just pointing out the endless list of things it can't.

Just my point of view on the matter. I work in equities research, was an econ major in college. I don't have what I would deem a professionally strong statistics or math background (which to me would entail a graduate degree or at least undergraduate degree in math or stats), but I do have some background there and understanding of them.

 

[roy_miami]

I have to warn you though...much of the reason the NFL has likely resisted the heavy integration of analytics into the process is because as soon as they get a real statistics expert in the office that person is more likely to spend his days saying "no" than saying "yes". A person with a firm grounding in statistics would tend to lament the insufficient sample sizes and lack of statistical significance involved in any study involving football-related data sets. Some of the stuff you're asking about (e.g. 'recent trends') strikes me as more likely to fall into traps than to be meaningful, most often due to insufficient sample size. I think once you get beyond that barrier if a statistics guy really begins to understand the sport and the drivers of success then perhaps he can start to figure out more valuable things that analytics actually can offer the sport instead of just pointing out the endless list of things it can't.

Just my point of view on the matter. I work in equities research, was an econ major in college. I don't have what I would deem a professionally strong statistics or math background (which to me would entail a graduate degree or at least undergraduate degree in math or stats), but I do have some background there and understanding of them.

I think you're more likely to get mostly "inconclusives" rather than "no's." I've seen very few analytics that are powerful enough to give firm answers one way or the other and if you're are going to use them they should be used as a starting point rather than the deciding factor.

 

[Roman529]

I think James would likely still be around after #19 but before we pick again at #50.....so if the top guys are off the board at #19, maybe try to trade down and get James a little later and another quality guy before we pick again at #50. I still think there is a decent possibility that Lewan or Zack Martin drops down to us.

 

[futurescout]

I have to warn you though...much of the reason the NFL has likely resisted the heavy integration of analytics into the process is because as soon as they get a real statistics expert in the office that person is more likely to spend his days saying "no" than saying "yes". A person with a firm grounding in statistics would tend to lament the insufficient sample sizes and lack of statistical significance involved in any study involving football-related data sets. Some of the stuff you're asking about (e.g. 'recent trends') strikes me as more likely to fall into traps than to be meaningful, most often due to insufficient sample size. I think once you get beyond that barrier if a statistics guy really begins to understand the sport and the drivers of success then perhaps he can start to figure out more valuable things that analytics actually can offer the sport instead of just pointing out the endless list of things it can't.

Just my point of view on the matter. I work in equities research, was an econ major in college. I don't have what I would deem a professionally strong statistics or math background (which to me would entail a graduate degree or at least undergraduate degree in math or stats), but I do have some background there and understanding of them.

I literally had to re-read this post twice shows my mathematical knowledge and understanding..

I agree that sample size is a limiting factor in this. That's why this should be taken with a grain of salt. Not an "end all be all" kind of thing. Rather, a way to look at things. How can you look at recent trends that have a small sample size? Not a rhetorical question. I believe things like recent trends should be something to look at and try and draw conclusions from. At most, it would be a very small piece of the pie. For example the trend of Right Tackles. I think it's safe to say that the position is trending, and good ones are going higher now than, say in 2004.

I'm not a statistics or analytics guy trying to apply football knowledge I barely understand. I'm a football guy trying to apply statistics or analytics I barely understand.

I'm not trying to come up with a money ball scenario or formula. Just look at trends and the way successful teams are doing things, and what conclusions can be drawn from it. But I would rely heavily on my football knowledge when doing this. Not some "holy grail formula"

 

[futurescout]

I think you're more likely to get mostly "inconclusives" rather than "no's." I've seen very few analytics that are powerful enough to give firm answers one way or the other and if you're are going to use them they should be used as a starting point rather than the deciding factor.

This is your best post I've seen. I think they should be used as a starting point to look at way things are being done. For example, trying to find strange correlation between crazy statistics and how much they equate how much a QB is responsible for wins by using YPA is beyond ridiculous.


Anything shouright/gravity has posted is rubbish

However, stats like the correlation beween turnover differential and winning is a valid one.

I've found the deeper you dive into formulas, the farther away you get from football, and the less weight it will hold with football people.

Simple stats like 3rd down % redzone % and turnovers and how they relate to winning are good places to start. A lot easier to measure.