Jonathan Bales on draft grades

[Common Sense]

There are a few more tweets, but that's the gist of the argument. This comes hot of the heels of data that suggests that even supposedly "good" GMs don't do any better at identifying quality college prospects than picking them at random, i.e., it's a complete crapshoot. Bales argues that playing the economics of the draft and maximizing chances and probability are what actually matter. Hence the blackjack metaphor. Thoughts?

 

[ThreeandOut]

Bales is a big proponent of identifying characteristics that statistically predict success in the NFL for each position (e.g. 40 yard dash for a RB) and drafting according to those characteristics. So he is likely judging a draft selection as good or bad based on the extent that selection met the predefined standard. A player that doesn't meet the standard may be successful in his career, but statistically you are more likely to be successful by selecting a player that meets the standard. That was what is was getting at in the blackjack analogy.

He seemed to be to making the above points in the context of disliking the Giants selection of Odell Beckham. I didn't catch the reason for him not liking Beckham. I do remember Bales hating the Joseph Randall pick last year based on his slow 40 time.

 

[FuzzyLumpkins]

Sure and now he is using that same brain to pretend like his pet cat WR should have been drafted over Streets. You cannot have it both ways: you either think it is random or you do not. it is pretty obvious that he does not think that despite his protests to the contrary.

 

[jobberone]

It's a science and an art. You need both. It's both a crap shoot as well as the art of successful drafting. For me the biggest thing is to look for what makes a football player and character as well as athleticism. I don't have a problem adding any stats that identifies a player's odds of success. But it would just be part of the equation not THE answer for how to draft.

I do think if you have the right system and personnel set up you can beat the odds over the long haul. You're going to have runs of success and not so successful.

 

[visionary]

Sure and now he is using that same brain to pretend like his pet cat WR should have been drafted over Streets. You cannot have it both ways: you either think it is random or you do not. it is pretty obvious that he does not think that despite his protests to the contrary.

Way to misunderstand and (or) misconstrue his entire point

"one of many paths" does not equate to "random"

He makes a great point

 

[FuzzyLumpkins]

Way to misunderstand and (or) misconstrue his entire point

"one of many paths" does not equate to "random"

He makes a great point

I think you are not reading all of his quotes and points he is trying to make. He has said that there is no statistical difference between random picks and a GM picking over time. In essence he claims that every GM eventually returns to the mean. Its his the entire basis behind his push to stock up picks because if they all come out in the wash then it is better to have more picks than fewer.

He has all the hallmarks of sophistry and pseudoscience. He loves to conflate correlation and causation. He will sift through stats until he finds the one that fits his conclusion rather than looking at the data set and making the conclusions organically. And he uses QBR in his QB evaluations, the worst fluff stat that assumes its conclusion I have ever seen. That is just from reading his twitter feed. He is a hack.

 

[FuzzyLumpkins]

Further I would add, you are grading the draft that a particular team makes. I fail to see how he can say that evaluating the match of pick and team right off the bat is valid but seeing how well the team and the pick performed down the line is not. It's asinine.

Sure he may performed well somewhere else but you are not grading the player in a vacuum. You are grading the match of team and player. You cannot just divorce that.

Further, he is assuming that the mediot graph charts are 'correct' or 'true' like some platonic form of perfection. It's ********.

And I love his assertion that actual outcomes are meaningless as if we should go with his theorycrafting over empirical proof. Like I said: pseudoscience.

He is in the business of selling the instant evaluator draft grades. His words in the regard are akin to listening to Jerry Jones give his team a draft grade.

 

[visionary]

I think you are not reading all of his quotes and points he is trying to make. He has said that there is no statistical difference between random picks and a GM picking over time. In essence he claims that every GM eventually returns to the mean. Its his the entire basis behind his push to stock up picks because if they all come out in the wash then it is better to have more picks than fewer.

He has all the hallmarks of sophistry and pseudoscience. He loves to conflate correlation and causation. He will sift through stats until he finds the one that fits his conclusion rather than looking at the data set and making the conclusions organically. And he uses QBR in his QB evaluations, the worst fluff stat that assumes its conclusion I have ever seen. That is just from reading his twitter feed. He is a hack.

I haven't read his stuff previously except in passing
If that is his point I would not agree with him

I thought he was making a more sophisticated argument that there are a lot of variables that can affect how a player turns out over 3 years, player dynamics, team dynamics, late bloomer, coaching, etc

So the "draft evaluation" can almost be a 2 part process, the obvious one 2-3 years later to see how the player actually turned out and a more short term one to determine how and why a team picked certain players, is the thinking and the process of draft selection a sound one. There will obviously be more different opinions here and that is why there will be "reaches" but as long as the process is a good one, teams will make better decisions in the long run on average. In this context when a GM and a team has a history of making sound decisions, "draft graders" give them better grades (eg Baltimore) when a GM and team has a history of making poor, flashy and impulsive decisions (eg cowboys, skins, raiders) they are likely to get a worse grade shirt term. That is not the media "hating" on that team.

The best thing would be (and I am sure teams ready do this) to build in a quality control and feedback loop where someone objective is looking at how your previous picks turned outing term and trying to build in strengths and change deficiencies so the team can improve it's draft process

Anyway, my point was that one could "evaluate" the draft both short term and long term

His other point about a long shot hitting does not mean the decision was a sound one is a very very good point

 

[FuzzyLumpkins]

I haven't read his stuff previously except in passing
If that is his point I would not agree with him

I thought he was making a more sophisticated argument that there are a lot of variables that can affect how a player turns out over 3 years, player dynamics, team dynamics, late bloomer, coaching, etc

So the "draft evaluation" can almost be a 2 part process, the obvious one 2-3 years later to see how the player actually turned out and a more short term one to determine how and why a team picked certain players, is the thinking and the process of draft selection a sound one. There will obviously be more different opinions here and that is why there will be "reaches" but as long as the process is a good one, teams will make better decisions in the long run on average. In this context when a GM and a team has a history of making sound decisions, "draft graders" give them better grades (eg Baltimore) when a GM and team has a history of making poor, flashy and impulsive decisions (eg cowboys, skins, raiders) they are likely to get a worse grade shirt term. That is not the media "hating" on that team.

The best thing would be (and I am sure teams ready do this) to build in a quality control and feedback loop where someone objective is looking at how your previous picks turned outing term and trying to build in strengths and change deficiencies so the team can improve it's draft process

Anyway, my point was that one could "evaluate" the draft both short term and long term

His other point about a long shot hitting does not mean the decision was a sound one is a very very good point

Again, he is making the assumption that what he sees as a "long shot" is correct in the first place as if he and the other mediots draft charts are "true."

If the team evaluated someone as being a sure shot and he determined it to be a long shot then what you have is a difference of opinion. Neither is an objective "truth." There are no objective probabilities before the fact. If the player works out then it validates the team in thinking someone was going to work out. He is just doubling down on his initial find and blustering.

 

[visionary]

Again, he is making the assumption that what he sees as a "long shot" is correct in the first place as if he and the other mediots draft charts are "true."

If the team evaluated someone as being a sure shot and he determined it to be a long shot then what you have is a difference of opinion. Neither is an objective "truth." There are no objective probabilities before the fact. If the player works out then it validates the team in thinking someone was going to work out. He is just doubling down on his initial find and blustering.

That is where your drafting history as a GM and an organization comes into play

If the ravens or the niners "reach" for a player in rd 1 or 2, very few will call that a reach because of their history of talent evaluation and the belief that their process of drafting has been shown to work. Most in the media would be scrambling to find out what they themselves missed in the evaluation of that player...... For good reason

There is a reason we are 8-8 year after year under jones/Garrett, poor player evaluation and poor coaching

In Dallas the emperor has no clothes but no one is willing to tell him

 

[FuzzyLumpkins]

That is where your drafting history as a GM and an organization comes into play

If the ravens or the niners "reach" for a player in rd 1 or 2, very few will call that a reach because of their history of talent evaluation and the belief that their process of drafting has been shown to work. Most in the media would be scrambling to find out what they themselves missed in the evaluation of that player...... For good reason

There is a reason we are 8-8 year after year under jones/Garrett, poor player evaluation and poor coaching

In Dallas the emperor has no clothes but no one is willing to tell him

You didn't need to tell me you think that we are doomed to failure because of Jerry Jones. I already knew that.

That has no bearing on the discussion at hand. It's ad hominem frankly and an oversimplified copout.

 

[DFWJC]

There are a few more tweets, but that's the gist of the argument. This comes hot of the heels of data that suggests that even supposedly "good" GMs don't do any better at identifying quality college prospects than picking them at random, i.e., it's a complete crapshoot. Bales argues that playing the economics of the draft and maximizing chances and probability are what actually matter. Hence the blackjack metaphor. Thoughts?

I disagree with Bales here.

Just because some analysts thinks a team should or should not have picked a player, does not remotely mean they are right.

Real results in real NFL games is proof one way or the other...or at least a close to proof that we'll ever get.

The rest is theory...period.

 

[xwalker]

Bales is a big proponent of identifying characteristics that statistically predict success in the NFL for each position (e.g. 40 yard dash for a RB) and drafting according to those characteristics. So he is likely judging a draft selection as good or bad based on the extent that selection met the predefined standard. A player that doesn't meet the standard may be successful in his career, but statistically you are more likely to be successful by selecting a player that meets the standard. That was what is was getting at in the blackjack analogy.

He seemed to be to making the above points in the context of disliking the Giants selection of Odell Beckham. I didn't catch the reason for him not liking Beckham. I do remember Bales hating the Joseph Randall pick last year based on his slow 40 time.

According to Bales theory, Emmitt Smith was a bad pick.

 

[visionary]

You didn't need to tell me you think that we are doomed to failure because of Jerry Jones. I already knew that.

That has no bearing on the discussion at hand. It's ad hominem frankly and an oversimplified copout.

I will take that to mean that you know you were wrong

Bales made some very good points and you were bringing your usual blame the messenger schtick to try and refute them because they point out deficiencies that apply to the cowboys

 

[visionary]

According to Bales theory, Emmitt Smith was a bad pick.

There are exceptions to every "theory"
Exceptions don't invalidate an entire theory

 

[visionary]

I disagree with Bales here.

Just because some analysts thinks a team should or should not have picked a player, does not remotely mean they are right.

Real results in real NFL games is proof one way or the other...or at least a close to proof that we'll ever get.

The rest is theory...period.

But if most analysts think so, maybe just maybe there is a reason for it?

One could look at this short term and long term

Both analyses are valid and have different objectives

 

[FuzzyLumpkins]

I will take that to mean that you know you were wrong

Bales made some very good points and you were bringing your usual blame the messenger schtick to try and refute them because they point out deficiencies that apply to the cowboys

You can bluster whatever you like. All you said was that Jerry Jones track record makes you have no faith in him.

What good points did Bales make that I did not refute and if I did refute then what was your rebuttal. When you judge who wins or loses a debate, that is the paradigm. Your bluster frankly is boring.

Are you ever going to address my point that there is no objective probability to base his assertion, the whole thing about his draft charts being 'true,' on or did you just not understand what that means and are resorting to this tactic instead?

 

[DFWJC]

But if most analysts think so, maybe just maybe there is a reason for it?

One could look at this short term and long term

Both analyses are valid and have different objectives

All I say is that the vast majority of analysts loved....

Todd Marinavich
Lawrence Phillips
Heath Shuler
Jamarcus Russell
Ki-Jana Carter
Akili Smith
Ryan Leaf
Tony Mandarich

 

[FuzzyLumpkins]

But if most analysts think so, maybe just maybe there is a reason for it?

One could look at this short term and long term

Both analyses are valid and have different objectives

I am sure that there is a reason for it but that reason is not necessarily that they have uncovered the 'truth.' Quite frankly, most of the draftniks seem to think with the same brain all too often.

You familiar with the concept of groupthink?

 

[Common Sense]

One possibility I keep coming back to is that Bales' overall theory is essentially correct, but his criteria for player selection and odds calculation are flawed. Maybe he's right about what is essentially risk aversion, but isn't very good at determining what represents actual risk (Say, the Emmitt Smith/running back's 40 time example).