The Draft is a Total Crapshoot (Per Science)

[Beck Water]

1 hour ago, Pine Barrens Mafia said:

And yet, there are teams that nail it year after year at key positions.

 

 

OK, this could be a basis for discussion.   

 

Who, in your view, are these teams?

[FireChans]

1 hour ago, Coach Tuesday said:

https://theathletic.com/5416007/2024/04/16/nfl-drafting-methods-insight-massey-thaler/?source=user_shared_article
 

Great article from the Athletic.  Basically says that teams routinely over-estimate their own ability to predict player performance and as a result overestimate the value of their own specific picks.  Instead, they should maximize the number of lottery tickets they’re buying.

 

Thoughts?

The NFL is resistant to analytics because they don’t want the data out there.

 

The problem with arguing about draft pick accumulation is that draft pick trades are not always “more valuable pick next year for this year’s pick,” which is sort of a fallacious argument. Yes, if a team offered next years first for your second every year, you’d be stupid to not take it.

 

But what’s the data on trading down from the 5th round for a 6th and 2 7ths? Probably not as much of a slam dunk. What about your second for an extra 3rd and 4th and next years 3rd? Hard to say if that’s as much of a no-brainer.

 

I will be first to say if a team offers us a first next year for pick 60, Brandon Beane should take it. They won’t, though.

[Matt_In_NH]

1 hour ago, Pine Barrens Mafia said:

And yet, there are teams that nail it year after year at key positions.

 

Which ones?

[SoMAn]

1 hour ago, Pine Barrens Mafia said:

And yet, there are teams that nail it year after year at key positions.

 

Which teams do you have in mind that consistently restock their roster from the draft and a high percentage of players that contribute?

I'm not doubting you, just curious of which teams you considered that fit the description.

[FireChans]

2 minutes ago, eball said:

 

I'm not saying there aren't guys who are better or worse at it, but even Ozzie missed a lot.

 

The goal is to be right enough of the time at key positions...but it's still anything but scientific.

 

If they are better at it, they, by definition, are not wrong as much as everyone else. They are actually wrong the least. Because they are better at it. That’s what “better” means!

[gonzo1105]

2 minutes ago, SoMAn said:

Which teams do you have in mind that consistently restock their roster from the draft and a high percentage of players that contribute?

I'm not doubting you, just curious of which teams you considered that fit the description.

Baltimore would be the one I would think of off the top of my head without doing research 

[Beck Water]

1 hour ago, Buffalo_Stampede said:

Probably a weird way to look at it but I believe the draft will get easier with NIL deals. These players will show who they are very early in college. 

 

That's actually a valid point.

 

One of the biggest unpredictable factors is, you take a guy who may have come from an impoverished or chaotic family background and who, in any case, has not been in the habit of dealing with large amounts of money. 

 

How will that impact him? 

 

Will he still have a "fire within" to be great? 

 

Or will he start loving the lifestyle the money can give him, over the pain and effort of the game that gave him that lifestyle?  Or maybe he starts out loving the lifestyle and not as motivated and something changes him - he has a child, or someone influences him.

 

It's intangible, and like any intangible, predicting that can be highly flawed.

[Kirby Jackson]

I buy it as a concept. I think it primarily applies after the top 20. I’ll use the extreme example in MHJ. The chances of him failing are very small. Now, if you want to say that Legette and Baker for example is a coin flip, sure. I think there are high floor and low ceiling prospects. There are also boom or bust guys. Those type of guys I’ll buy that you’re better off with 2 swings than 1.

[Beck Water]

5 minutes ago, gonzo1105 said:

Baltimore would be the one I would think of off the top of my head without doing research 

 

Ozzie Newsome was one GM called out in the OP Athletic article as consistently better than most:

"Though some posted better track records than others, specifically Baltimore’s Ozzie Newsome, the answer was mostly not."

 

But get this:

 

"Coincidentally, around that time, Ravens GM Eric DeCosta said the following on a podcast: “There was a really seminal article written in 2005. It was really about the draft and how teams should trade back and always acquire picks — and never trade up.”

DeCosta doubled down in 2021 when a reporter mentioned the Ravens as one of the top drafting teams in the NFL. “We’ve probably had the most picks over that span,” he said. “That goes back to a philosophy that I think Ozzie started back in 1996.”"

 

So Ozzie Newsome and his successor Eric DeCosta may be doing better than average, because they are following the "trade back, get more lottery tickets" philosophy.

 

 

 

[FireChans]

6 minutes ago, Beck Water said:

 

Ozzie Newsome was one GM called out in the OP Athletic article as consistently better than most:

"Though some posted better track records than others, specifically Baltimore’s Ozzie Newsome, the answer was mostly not."

 

But get this:

 

"Coincidentally, around that time, Ravens GM Eric DeCosta said the following on a podcast: “There was a really seminal article written in 2005. It was really about the draft and how teams should trade back and always acquire picks — and never trade up.”

DeCosta doubled down in 2021 when a reporter mentioned the Ravens as one of the top drafting teams in the NFL. “We’ve probably had the most picks over that span,” he said. “That goes back to a philosophy that I think Ozzie started back in 1996.”"

 

So Ozzie Newsome and his successor Eric DeCosta may be doing better than average, because they are following the "trade back, get more lottery tickets" philosophy.

 

 

 

To @Kirby Jackson’s point, there are exceptions to even this philosophy.

 

Baltimore famously traded up for Flacco in 2008 and LJ in 2018. 

[TheyCallMeAndy]

2 hours ago, Coach Tuesday said:

https://theathletic.com/5416007/2024/04/16/nfl-drafting-methods-insight-massey-thaler/?source=user_shared_article
 

Great article from the Athletic.  Basically says that teams routinely over-estimate their own ability to predict player performance and as a result overestimate the value of their own specific picks.  Instead, they should maximize the number of lottery tickets they’re buying.

 

Thoughts?

Agreed, with variables. Teams with better talent development (Bills) may be more likely to hit on the prospect than teams without it a strong development system. 
 

That’s why even the stats about where WRs have been drafted and how successful needs to be taken with heavy variables in mind. A 2nd round pick or third round WR drafted with an elite QB could have a better career than a first rounder who went to a crap team. 

[GunnerBill]

I think overall the draft is far from an exact science. There are teams that do better and teams that do worse, but nobody bats close to 100. I think in general you should want to apply the more picks strategy. But that doesn't equate to never trade up and always try and trade down. It's just a general principle for how to maximise your chances. 

 

My theory on drafting is do not fixate on any one player or position. Do not reach through draft tiers. Let the board make your decisions for you. That generally favours the Ravens style "let the draft come to you" type approach. 

[Rigotz]

Ah yes, the increasingly popular opinion:

 

"I don't understand this, so it's not real"

 

Bill Belichick was notorious for trading down to accumulate an immense amount of picks because he believed the draft is mostly a gamble. He was one of the worst GMs in modern history. Of course there is luck involved, but GMs like Ozzie Newsom routinely outperform awful GMs like Matt Millen and Bill O'Brien.

 

In the new world of advanced analytics, these gaps will tighten even more, and the expected value of a player will be closer to their real value.

 

Looking at historical analysis of hit rates is a fool's errand. The world has changed.

[dave mcbride]

6 minutes ago, GunnerBill said:

I think overall the draft is far from an exact science. There are teams that do better and teams that do worse, but nobody bats close to 100. I think in general you should want to apply the more picks strategy. But that doesn't equate to never trade up and always try and trade down. It's just a general principle for how to maximise your chances. 

 

My theory on drafting is do not fixate on any one player or position. Do not reach through draft tiers. Let the board make your decisions for you. That generally favours the Ravens style "let the draft come to you" type approach. 

Going back to 2014 (a ten year period), the Patriots have had more than seven picks in 9 out of the 10 seasons (usually between 9 and 11). Yet it really didn't help them all that much: https://www.pro-football-reference.com/teams/nwe/draft.htm. There is a handful of very good players here, but overall it's pretty much a vast wasteland.

[Beck Water]

3 minutes ago, GunnerBill said:

I think overall the draft is far from an exact science. There are teams that do better and teams that do worse, but nobody bats close to 100. I think in general you should want to apply the more picks strategy. But that doesn't equate to never trade up and always try and trade down. It's just a general principle for how to maximise your chances. 

 

My theory on drafting is do not fixate on any one player or position. Do not reach through draft tiers. Let the board make your decisions for you. That generally favours the Ravens style "let the draft come to you" type approach. 

 

The OP Athletic article also doesn't discuss differences between the top of the draft and later rounds, or between the behavior of a team that is rebuilding vs one that is set at a lot of positions.  It seems to me that probably matters.

 

Someone referenced PFF data on this.  I and others have done more back of the envelop studies that indicate the probability of getting a good player is something like 30% overall in the 1st round (maybe 50% at the top of the first round).  That's not a star, just a good starter - including Jerry Jeudy with Justin Jefferson and CeeDee Lamb.  The second round is also 30% overall, falling to something like 20% in the 3rd round and maybe 10% after that.

 

Of course, there may be some self-fulfilling prophecy in that higher draft picks tend to get more development attention and more chances to start.

 

[Captain Hindsight]

The best GMs probably have a 50% hit rate. Thats why I don't get hung up on Beane missing on a second rounder sometimes. Yeah it sucks, but hes still a damn good GM and better than most 

[Bleeding Bills Blue]

1 hour ago, Buffalo_Stampede said:

Pretty much. Talent, size, and explosiveness is what scouts look for over production. But then you have players with great production and average size and explosiveness. Those are typically the players that get missed by scouts. It’s almost impossible to know who will translate to the NFL based on measurements. Then there’s the mental makeup. Something no one can predict.

 

QB is a total coin flip. I don’t care what anyone says, they have no clue.

 

Arm talent and athleticism can get you far in college - in the NFL everyone has arm talent and i'd say most QBs today have some level of athleticism.  

 

Between the ears pre and post snap, how you throw on the run and under pressure, how you sense pressure in general, how you manipulate the defense, how you manipulate passing windows, etc.  That seems to be the part that most scouts get wrong, and probably the hardest part to evaluate and project at the next level. 

[Kirby Jackson]

26 minutes ago, GunnerBill said:

I think overall the draft is far from an exact science. There are teams that do better and teams that do worse, but nobody bats close to 100. I think in general you should want to apply the more picks strategy. But that doesn't equate to never trade up and always try and trade down. It's just a general principle for how to maximise your chances. 

 

My theory on drafting is do not fixate on any one player or position. Do not reach through draft tiers. Let the board make your decisions for you. That generally favours the Ravens style "let the draft come to you" type approach. 

I saw a stat years ago when the Patriots were good at drafting. I’m going to have the details wrong but the concept will be right. They made something like 60 trades at or around the draft (over like a 10 or 15 year period). People would say, “the Patriots always trade down to get more picks.” It ended up being like 19 trade ups, 22 trade downs and 19 trades for players. It was almost equal across the board. That made sense to me and screams, “let the board determine your decisions.”

[34-78-83]

13 minutes ago, Captain Hindsight said:

The best GMs probably have a 50% hit rate. Thats why I don't get hung up on Beane missing on a second rounder sometimes. Yeah it sucks, but hes still a damn good GM and better than most 

Change that to about 33% and I agree with the rest of what you said.

[BarleyNY]

2 hours ago, Coach Tuesday said:

https://theathletic.com/5416007/2024/04/16/nfl-drafting-methods-insight-massey-thaler/?source=user_shared_article
 

Great article from the Athletic.  Basically says that teams routinely over-estimate their own ability to predict player performance and as a result overestimate the value of their own specific picks.  Instead, they should maximize the number of lottery tickets they’re buying.

 

Thoughts?

 

I think that it amounts to this: Teams have maxed out their ability to predict the success of draft prospects. The unnecessary extra work done on prospects doesn’t improve their hit rate, but only increases the confidence level and makes them overly confident in their decisions. So the first mistake teams make is overvaluing players they spend too much time on. Those are almost always the players that fit their needs. 

 

The second place teams make mistakes is when uninformed coaches or owners meddle. The Browns under Haslam were a huge offender of this until recently. The Justin Gilbert pick was to appease a DC who, at the last minute, demanded he be the draft pick because he needed a player like that to run his defense. No real work had been done on him since he hadn’t been a prospect the team was considering. So the Browns drafted a player who really didn’t like football in the top 10. Then later that night they traded back into the first round because the owner wanted Johnny Manziel. 

 

Analytics driven FOs know that there is a baseline level of uncertainty that is inherent in the NFL draft and they act accordingly. Quality teams are run with that in mind and without interference.