For those with The Athletic: the science of the draft...teams know it but rarely follow it

[msgkings322]

NFL teams know the best way to draft, so why aren't they doing it?

Only a few teams are curious enough to think differently, and even fewer are disciplined enough to act differently.

theathletic.com

The science:

First, he wondered, who wrote this?

Richard Thaler, an economics professor at the University of Chicago who would win a Nobel Prize in 2017, and Cade Massey, a business professor then at Duke University.

Their hypothesis?

Teams overestimate their abilities to delineate between stars and flops, and because of that they overvalue the “right to choose” in the draft.

And what were the findings after examining every draft pick and trade from 1988 to 2004?

Teams massively overestimate their abilities to delineate between stars and flops, and because of that they heavily overvalue the “right to choose” in the draft.

Meers combed through the paper and uncovered some highlights:

• The treasured No. 1 pick in the draft is actually the least valuable in the first round, according to the surplus value a team can create with each pick.
• Across all rounds, the probability that a player starts more games than the next player chosen at his position is just 53 percent.
• Teams generated a 174 percent return on trades by forgoing a pick this year for picks next year.

Thaler and Massey suggested that teams should accumulate picks by trading back and into the future more often. The more darts you have, the better your chance of eventually hitting the bull’s-eye.



Owners like Jerry Jones met with these guys, but most teams still don't operate using these findings...basically it says it's almost always better to trade back and get more picks, because the hit rate isn't very good on top picks and you get way more value with more picks later on. Some say QBs are an exception.

 

[shopson67]

It seems like they over-generalized. As you noted, QBs will be an exception (and not all teams struggle the same with their selections).

 

[HaroldSeattle]

Takes a trade partner to trade down and I think most the time teams can't find that trade partner unless a QB is involved.

 

[Clayton]

NFL teams know the best way to draft, so why aren't they doing it?

Only a few teams are curious enough to think differently, and even fewer are disciplined enough to act differently.

theathletic.com

The science:

First, he wondered, who wrote this?

Richard Thaler, an economics professor at the University of Chicago who would win a Nobel Prize in 2017, and Cade Massey, a business professor then at Duke University.

Their hypothesis?

Teams overestimate their abilities to delineate between stars and flops, and because of that they overvalue the “right to choose” in the draft.

And what were the findings after examining every draft pick and trade from 1988 to 2004?

Teams massively overestimate their abilities to delineate between stars and flops, and because of that they heavily overvalue the “right to choose” in the draft.

Meers combed through the paper and uncovered some highlights:

• The treasured No. 1 pick in the draft is actually the least valuable in the first round, according to the surplus value a team can create with each pick.
• Across all rounds, the probability that a player starts more games than the next player chosen at his position is just 53 percent.
• Teams generated a 174 percent return on trades by forgoing a pick this year for picks next year.

Thaler and Massey suggested that teams should accumulate picks by trading back and into the future more often. The more darts you have, the better your chance of eventually hitting the bull’s-eye.



Owners like Jerry Jones met with these guys, but most teams still don't operate using these findings...basically it says it's almost always better to trade back and get more picks, because the hit rate isn't very good on top picks and you get way more value with more picks later on. Some say QBs are an exception.

PFF used to talk about this all of the time but I think the 2 analysts moved on since then.

I think their analysis was that the worst thing you can do is 'reach' based on consensus. (The Raiders were commonly used as an example) The best thing you can do is take a QB. Then WR or CB. Trade down otherwise. Get your oline to average. Running backs dont matter.

 

[fastforward]

Interesting article. It's behind a paywall. I'm not a subscriber, but I got around the paywall. I'm not sure what this site's rules are regarding the posting of whole articles or information about how to bypass the paywall.

 

[Iggloo]

It's a good article, I am a subcriber. The TLDR version is this:

1) Teams overvalue their ability to pick players better than other people, and thus overvalue higher picks because they downplay how many flop.

2) Trading down and acquiring more picks is a better strategy, data show, but few teams really do it because of ego and human nature.

 

[shopson67]

It's a good article, I am a subcriber. The TLDR version is this:

1) Teams overvalue their ability to pick players better than other people, and thus overvalue higher picks because they downplay how many flop.

2) Trading down and acquiring more picks is a better strategy, data show, but few teams really do it because of ego and human nature.

Extra picks is only good to a point. There are only so many roster spots for draft picks.

 

[Iggloo]

Extra picks is only good to a point. There are only so many roster spots for draft picks.

That's true, but if teams were honest they could weed out players after an initial training camp and not keep guys they drafted who didn't look good, just because they had been draft picks.

Right now, a lot of teams draft guys in, say, the third round who are outplayed in training camp, but they still make rosters due to the sunk costs. Teams with a lot of picks would need to say, this guy isn't working out, we're cutting him and keeping the 7th rounder, and again it comes down to ego on why that isn't happening now.

 

[fastforward]

That's true, but if teams were honest they could weed out players after an initial training camp and not keep guys they drafted who didn't look good, just because they had been draft picks.

Right now, a lot of teams draft guys in, say, the third round who are outplayed in training camp, but they still make rosters due to the sunk costs. Teams with a lot of picks would need to say, this guy isn't working out, we're cutting him and keeping the 7th rounder, and again it comes down to ego on why that isn't happening now.

Then again teams select prospects with very high ceilings in the 3rd & 4th rounds who they know will take a year or 2 to develop. Maybe the way forward is a change in roster sizes from 53 to 47 + any rookies you select in the Draft.

 

[Iggloo]

That's true, and it would have to be just one part of the draft evaluation for that reason. You would also have to factor in more time for known projects.

But I think we all know what I am describing: GMs who stick with picks because they made them and are hoping against hope that they show signs of progress. That's a problem in all sports and is hard to avoid because it is about people not wanting to admit their mistakes.