NF Post: Don't draft in the 7th round.

[thunderspirit]

or something like that. link.

i'd like more data -- say, a three or four year span (well, five would be better, but as most rookie deals are four years now that has its own problems), along with UDFA considerations -- but the concept is interesting.

what do better statheads than me think?

[anhur]

It seems like the numbers are skewed by guys not making the team in the late rounds, and their entire cap number being counted as 1 snap. I don’t see how it’s really even a useful tool to evaluate value. His example of a guy being part of a trade is flawed too. When you trade for a guy, you don’t pay a signing bonus. I assume that the dollar value for players who were cut is exclusively their bonus, but if you sign a street free agent they usually get a small signing bonus too. If you cut them, it counts against your salary just like the bonus of a cut 7th round pick. If the point is that less talented players play less snaps, and provide less value per snap than sure, that makes sense. Every roster needs to be stocked with players like that though, so I don’t really see this as useful information.

[TheCheat]

There are roster limits and a salary cap. This isnt that useful. If a team doesnt have or use 7th round picks, that will use that roster spot on a UDFA or FA who will cost as much or more. The cost per snap problem gets transferred over to another player. Many 7th rounde guys are inactive mostly until cut if not cut ouright. You have to have inactive guys.

Maximizing snaps with late round picks is a philosophy the Bears used with the DB draft frenzy in late rounds for ST. You have to weigh the cost with the chance for finding a late round steal. Either way, there are only X snaps and Y players for a total Z dollars. Its a shell game of sorts.

What it tells us is what we all already know. 2nd and 3rd rounders are the best value picks. The Patriots have been talked about ad nauseum for this. Late round picks may as well be packaged up for a higher pick for a higher value section. Difficulty with this is finding the team willing to take late picks at any real value.

What really makes the difference is finding good players that contribute early. Working stats is helpful to some degree, but your cost per snap looks like garbage if you trade around and wind up with 2 Dan Buzuin guys in round 2.

Interesting thought though

[janos_51]

Its not a salary cap quantitative analysis, its the cost per snap for each player from a draft standpoint. Guys that can at least get in games lower this figure while guys who don't play a snap still have to be paid.

[BearsFanInMN]

I think the important part is I was right about picking Dennard day 3. :-)

[LT2_3]

This is useless. It entirely ignores the concept of developing players before needed. So, a team that stays healthy and has competent veterans playing while developing young talent for the future gets a bad grade. That's moronic.

[Wounded Bear]

I want a per on-field step analysis.

[thunderspirit]

thanks for the thoughts, all.

my initial thought is any attempt to quantify value in draft picks is a Good Thing [tm]. i'm certainly not convinced this is the way to go, but i'm wondering: is the reason it's not effective that it's lacking sufficient data, or that it's taking the wrong approach in trying to value what it's measuring?

[LT2_3]

It's the wrong approach. Measuring a player's productivity merely in their first year makes no sense. Also, there is no adjustment for a player's position as some positions are more likely to play as rookies than others.

It also doesn't address how easily the numbers would be skewed by both fewer draft picks by team or injuries.

Then there is the issue that it's counter-intuitive in it's assumption that playing rookies for many snaps is a good thing.

[Buffaloaf]

Another thing it doesn't consider is that if you are not filling out the bottom part of your roster with 7th round draft picks, the next thing you're looking at is "FA Acquisitions price per snap" with guys likely making more than a 7th round pick.

[nomoreshoop]

Another thing it doesn't consider is that if you are not filling out the bottom part of your roster with 7th round draft picks, the next thing you're looking at is "FA Acquisitions price per snap" with guys likely making more than a 7th round pick.

Trade the next 40 7th round picks for the first pick in the draft

[brantcruz50]

It's the wrong approach. Measuring a player's productivity merely in their first year makes no sense. Also, there is no adjustment for a player's position as some positions are more likely to play as rookies than others.

It also doesn't address how easily the numbers would be skewed by both fewer draft picks by team or injuries.

Then there is the issue that it's counter-intuitive in it's assumption that playing rookies for many snaps is a good thing.

Agree. It also ignores the exponentially beneficial (IMO; and realize I am exaggerating given this isn't really an exponential function) value of having a key contributor who is also cheap. When you find one.