I Am No Scout But I Know Statistics

[Meathead]

if you put me in 1000 fights with a teacher who was also a stripper i would win probably two which by my calculations is about half

 

/thread

[bobobonators]

Trade up with whatever it takes to get to #3 and take Geno and never look back. I'd rather make a big splash and fail than to continue with this half-assed garbage we've been going through for the last decade, which has been a complete utter failure anyway.

[ndirish1978]

Ok dude check this s..t out... I clearly f...in named the title wrong... Here's an example for you... Your mother can clearly be a teacher and then be a stripper at night right? A person who calls himself defensewinzchampionshipz has nothing to do with anything else... Wat the !@#$ is your point?

 

 

Here's another stat for you... You put Mr. Weo and myself to fight a thousand times, I go 999-1000... The only win he gets is because I kneel on the floor and tap out on the last fight and pray to God to not let this man suffer anymore...

 

So, we can confirm you in fact know nothing about statistics or analogies. At this point I am shocked you spelled statistics right in the title, as you seem to be regressing with every post. It's like you're the Benjamin Button of logic.

[uncle flap]

 

As you likely know, I have done a few studies on drafting success rates......and found that each position has a different success rate.

 

 

IMO, the OP was on the right track. If the Bills have rated several QBs as 1st round prospects(with none being of Elite prospect talent).....and they believe that at least one will definitely be available at a later part in the 1st round........the best scenario(based upon my drafting history analytics) would be to trade down picking up 2 later 1st round picks.....taking a QB(which will likely have the same chance of becoming Elite as a QB selected at #8).....and also take a LB(who will also have a similar chance of success as a LB selected at #8).

 

 

 

I checked those out and they are certainly interesting and well-researched.

 

I agree with your premise but disagree with your conclusion.

 

The stats you're talking about are descriptive and speak to a particular pick's value - "pick" as in draft slot, not "pick" as in player. I don't think they can be applied position-wise because there are too many other variables in play.

 

For example (and correct me if I'm wrong), by your logic, any of the QBs the Bills could hypothetically draft at #8 (or #22 or #36, etc.) have the same chance at success/stardom. Well, that's obviously inherently flawed. While we can all disagree on who has the best shot, nobody would claim that each of the QBs chances are equal.

 

Or to put it another way, does Geno Smith have a better chance at success with the Eagles at #4, the Bills at #8, or the Vikings at #23? There's an argument for all sides, but now flip the teams. Does the draft position matter? Or is it more Geno's talent, his coaching staff, his surrounding cast, etc?

 

While draft position will impact the amount and types of opportunity a player receives (among many other factors), ultimately their individual ability is the key variable in their success. The question is, "Is there a statistical way to predict player X's success?"

 

The Holy Grail in NFL/NCAA statistics would be an analytical model to predict a player's success/stardom. Many have tried and seemingly all have failed.

 

From one of the links I posted upthread:

 

 

Given the poor predictive performance of models incorporating a variety of quantitative measures, it seems unlikely that collecting more statistics on the performance of college quarterbacks will yield a clearer picture about their likelihood of success in the NFL. Indeed, one might reasonably argue that there are few observable factors, either quantitative or qualitative, which are not already being used in a near-optimal way to predict quarterback performance. Though NFL draft “experts” at the major sports networks may object, it appears that factors which are inherently unmeasurable and/or random play a major role in determining whether a quarterback will succeed at the professional level.

 

[Dibs]

I checked those out and they are certainly interesting and well-researched.

 

I agree with your premise but disagree with your conclusion.

 

The stats you're talking about are descriptive and speak to a particular pick's value - "pick" as in draft slot, not "pick" as in player. I don't think they can be applied position-wise because there are too many other variables in play.

 

For example (and correct me if I'm wrong), by your logic, any of the QBs the Bills could hypothetically draft at #8 (or #22 or #36, etc.) have the same chance at success/stardom. Well, that's obviously inherently flawed. While we can all disagree on who has the best shot, nobody would claim that each of the QBs chances are equal.

 

Or to put it another way, does Geno Smith have a better chance at success with the Eagles at #4, the Bills at #8, or the Vikings at #23? There's an argument for all sides, but now flip the teams. Does the draft position matter? Or is it more Geno's talent, his coaching staff, his surrounding cast, etc?

 

While draft position will impact the amount and types of opportunity a player receives (among many other factors), ultimately their individual ability is the key variable in their success. The question is, "Is there a statistical way to predict player X's success?"

 

The Holy Grail in NFL/NCAA statistics would be an analytical model to predict a player's success/stardom. Many have tried and seemingly all have failed.

 

From one of the links I posted upthread:

 

[/size]

 

Actually, that's pretty much what I am claiming(Red).

 

The studies I did were not about figuring player talent......they were about the study of the ability to be able to analyze and grade that talent. I figure that you are aware of this....but bear with me.....I'll endeavour to explain my rationale. This will likely be quite a lot of typing as I want to give justice to each aspect covered.

 

 

Analyzing player talent is not an exact art......and it is impossible to determine an individual's ability to do so without having access to their private grading records. Obviously some will be better/worse than others, some will excel at grading certain positions, be deficient at other positions etc. It is possible that one(or a few) will be regularly much better than their peers.

 

Unfortunately, it is extremely unlikely that these potential super-scouts are fully recognized and their opinions will be just one of many opinions on a team that go into formulating their big board. Logically, in regards to individual teams selecting players, it all relatively evens out......some teams a bit better.....some a bit worse.....but all have their fair share of "misses" which is the important factor, as it shows that it is indeed not an exact art. This amalgam of minds doing the selecting will also be constantly changing for many and varied reasons, adding further to the impossible nature of determining individual(or team) scouting.

 

 

In relation to the QB position, the NFL scouting system as a whole has had an extremely good success rate(compared to other positions). There really cannot be much argument on this as the results clearly point to this being factual and not just conceptual.

 

Looking at the last 28 years, it is improbable to think that it is fluke that only one QB(Brady) has become an elite NFL QB after being selected past the 36th pick(2 if you prematurely count Wilson)......and only one outside the top 33 picks(3 if you prematurely count Kaepernick & Wilson). It is obvious that virtually every QB who has a glimmer of elite potential is recognized and, with the QB position being so vitally important, are selected at worst inside the top 4 picks of the 2nd round.

 

Furthering to the point, when a QB is recognized as having "special" potential, they are invariably selected inside the top 4 of the draft. Though some QB prospects(Alex Smith) are selected there due to supply/demand reasons, the success rate in those top 4 selections are far greater than other areas. Manning & Luck are great examples. The poorest of scouts could recognize their special potential. Leaf is also a good example as he was near universally considered to have special potential......which obviously didn't pan out.

 

 

Now....getting to your actual point....

We can safely state that scouts can recognize those QBs who have some sort of elite potential. We can also safely say that scouts can recognize those QBs who have special potential. In short, scouts can differentiate between those QBs who have no chance of becoming elite.....those QBs who have a chance of becoming elite......and those QBs who have a good chance of becoming elite.....

 

......but can they differentiate between different QBs inside each area? Can they determine the better/worse success chance between two QBs that fall into the middle category.....say between Geno Smith & Matt Barkley(assuming that both fall in that category)?

 

It seems logical that they can. If one can determine that Luck is has awesome potential, that Newton has very good potential, that Flacco has some potential & that Kafka has very little potential.....it makes sense that you would be able to determine some sort of difference between a Geno & Barkley.

 

I tend to think however that the basis for this is slightly skewed......in that it is likely that the players selected inside the top 4 vary greatly in potential....and any player who grades above a certain point, is highly desired, and invariably gets selected inside the top 4. Effectively, the gradings are not proportionate. Top 4 is relatively easy for the scouts to grade.....any QB who looks very good is worth spending a top 4 pick on.....regardless of just how good(Luck compared to Young as example).

 

Players who show elite potential but fall under that theoretical top 4 bar appear much more difficult for scouts to effectively grade. These guys show "something" which makes scouts believe that they could become elite. Trying to grade arm strength, intangibles, pocket awareness, accuracy etc, etc, against each other seems an extremely difficult task......and as history shows, a QB who is rated as a 8th pick or 15th type selection has just as much chance to succeed/fail as one who is rated as a 25th type selection.

 

Using a racing car analogy......grading QBs is like trying to predict which cars will go around a track faster...without the use of mathematics/physics. You have a bunch of cars varying from a F1 to a mini.....with varying engine sizes, tires, weight, suspension, driver etc. The F1 car with a good driver, or the muscle car with low weight and good suspension become obvious predictions. The mini with the small engine are also obvious predictions. All of the other combinations become impossible to predict. Does the car with the big engine & low weight but bad driver have a better or worse chance than the car with the medium engine & good suspension with a decent driver? Both have a chance......but how can you determine which one has the better chance?

 

In terms of mathematics, one may well be able to determine that player "A" is a better prospect than player "B".......but the variables involved outweigh the grading differences. Player "A" might be given a 10% chance.....player "B" a 6% chance......but there is a +/- of 90% making any grading of the players irrelevant. (Obviously my percentages here are fictional to highlight the point).

 

 

 

In regards to where a QB ends up being drafted effecting his elite potential.....history has shown this to be basically irrelevant. Bad teams generally get the top 4 picks. The elite QBs on a bad team maintain(develop) their elite level of play regardless to whether their WRs drop the balls or their OL allows constant pressure. It is usually only under duress that elite QBs show themselves as being better than their peers......basically, all QBs look good when they have no pressure, lots of time to throw & have good WRs......very few QBs look consistently good when under heavy duress.

 

Furthering to this, unless one believes that the "special" quality attributed to elite QBs can disappear, it is very telling that no QB, barring injury issues(in modern times?) has ever been considered not good enough by their drafting team......and then develops into an elite QB elsewhere.

This occurs with many players of many positions......but never at the QB position.

 

 

Sorry for the length of the response. Hopefully you didn't find it a complete waste of your time.

Edited March 19, 2013 by Dibs

[first_and_ten]

If you know stats, then you'll know we'll never be good until we get a GOOD QB.

 

fixed

Edited March 19, 2013 by first_and_ten

[uncle flap]

Actually, that's pretty much what I am claiming(Red).

 

The studies I did were not about figuring player talent......they were about the study of the ability to be able to analyze and grade that talent. I figure that you are aware of this....but bear with me.....I'll endeavour to explain my rationale. This will likely be quite a lot of typing as I want to give justice to each aspect covered.

-----

Sorry for the length of the response. Hopefully you didn't find it a complete waste of your time.

 

Not at all! I enjoyed reading it and I agree with everything you say.

 

I don't mean any offense, I just don't think there is much value to analyzing the success rate of scouts/GMs as a whole - except in terms of draft pick value a la Jimmy Johnson's chart.

 

Again, not to sound like a jerk, but it just doesn't much interest me to know that a QB drafted at #8 has X% chance to succeed based on all the previous QBs that have been drafted in or near that slot. Only because because that percentage has no bearing on a particular player's chance of success.

 

While it'd be misleading to say I'm not interested in descriptive stats, since I find that stuff interesting too, but what I'm really interested in is predictive stats.

 

It just sucks because as I said earlier, there doesn't seem to be a reliable model, and it's not for lack of trying or data.

 

I look forward to reading some more of your analyses and I'll def stay on the lookout for them!

[3rdand12]

I have a headache now.

But well worth the effort. Thanks gentlemen all , for a rather good read.