Pats 2022 draft class

rodderick

Member
SoSH Member
Apr 24, 2009
12,751
Belo Horizonte - Brazil
I'm just going to have to appeal to authority on this one. Well-run NFL teams--and not just New England--consistently think that it is worth it to spend a lot of resources on the o-line. Karras actually kind of proves this point--he was adequate in the role as far as I can see and now he's making five million a year?

Andy Reid and BB and the Ravens and the 49ers and the Rams and pretty much every other team you can think of is willing to pay some big bucks or use high drift picks on guards.
Let me flip this on you, then. Per Spotrac the average salary for an NFL guard is lower than the average salary for a Kicker. The only two positions it sits above are Punter and Fullback. So NFL teams usually don't value the position all that much, it seems.
 
There are a lot of people making the point that it only takes one othe team to be looking at "your" player to mean you need to take him earlier (or in a couple of cases with the Pats this year much earlier) than consensus.

But this seems like classic winner's curse territory to me. If consensus is that a player should go in the 70s but one other team would take him at say 32 so you need to take him at 30 to get him....well you may well be better off letting the other guys overpay instead. It's not a case of take-him-now-or-lose-him-forever, it's take-him-now-or-you-know-take-some-other-guy-instead. It is surely more likely in such case that you're wrong about the player than the 30 teams who have him ranked much lower?

One of my professors at b-school described the ethos of the investment firm he worked for as essentially "we don't believe anyone knows anything. We don't think we're smarter, but we don't believe the guy who really wants to buy a particular stock knows anything either. So we're happy to let him pay extra for the stocks he really wants and take the other side of that trade". It seems to me the draft is like that too.

There's also a lot of anecdotal evidence being point to here - it's important to understand that the value of the consensus big board is the consensus part. Wisdom of crowds stuff is a real thing.

So what makes me happy is when Bill (a) trades down because some other team absolutely has to have their guy (b) trades into next year because discount rates in the NFL are absolutely insane and any GM with job security should be exploiting the hell out of that (c) doesn't pick people way higher than they are expected to go.

There's a huge amount of noise when it comes to drafting so there will always be numerous counter-examples someone can point to. But why wouldn't you want to shift the percentages in your favor when you can?
 

MillarTime

Member
SoSH Member
Oct 31, 2013
1,338
How do we reconcile the beating the Pats took from the draft experts for the Strange and Thornton picks with the fact that nearly all of the them thought Malik Willis was going to be a first rounder and many of them believed he'd be a top 10 pick?
 

Shelterdog

Well-Known Member
Lifetime Member
SoSH Member
Feb 19, 2002
15,375
New York City
Let me flip this on you, then. Per Spotrac the average salary for an NFL guard is lower than the average salary for a Kicker. The only two positions it sits above are Punter and Fullback. So NFL teams usually don't value the position all that much, it seems.
So Joe Thuney is being paid in Monopoly money? The marLet value of a good guard-not the league average-is in excess of ten million.
 

ColonelMustard

Member
SoSH Member
Oct 23, 2006
219
One of my professors at b-school described the ethos of the investment firm he worked for as essentially "we don't believe anyone knows anything. We don't think we're smarter, but we don't believe the guy who really wants to buy a particular stock knows anything either. So we're happy to let him pay extra for the stocks he really wants and take the other side of that trade". It seems to me the draft is like that too.

There's also a lot of anecdotal evidence being point to here - it's important to understand that the value of the consensus big board is the consensus part. Wisdom of crowds stuff is a real thing.
The argument here is that the market doesn't reflect all available information. In passive investment management theory "the efficient-market hypothesis postulates that equilibrium market prices fully reflect all available information."
 
The argument here is that the market doesn't reflect all available information. In passive investment management theory "the efficient-market hypothesis postulates that equilibrium market prices fully reflect all available information."
No one really believes the pure efficient-market hypothesis is true. Obviously information isn't fully and equally known. Yet passive management still works.

It doesn't require perfect knowledge, or anything close to it, for it to be right to go with consensus over your own opinion that consensus is wrong.
 

ZMart100

Member
SoSH Member
Aug 15, 2008
3,188
The consensus has less information than NFL teams and is vulnerable to herding. I think a better estimate for how other teams evaluate a player is how your team evaluates that player. If you think a guy is worth a late first round pick, assume the rest of the league sees them similarly. The best strategy is to draft players who fit your roster where you think they belong.
 

Shelterdog

Well-Known Member
Lifetime Member
SoSH Member
Feb 19, 2002
15,375
New York City
Cool, now tell me what the Joe Thuney equivalent pass rusher or cornerback makes. It's not even comparable.
What's your point? The fact that a top pass rusher makes more than a top guard doesn't mean a top guard isn't valuable which is my point.

A top guard is absolutely a good use of a first round pick.

EDIT --absolutely agree that you wouldn't pick a top guard over a top pass rusher or corner or wide receiver.
 
Last edited:

RetractableRoof

tolerates intolerance
Lifetime Member
SoSH Member
Dec 1, 2003
3,836
Quincy, MA
There are a lot of people making the point that it only takes one othe team to be looking at "your" player to mean you need to take him earlier (or in a couple of cases with the Pats this year much earlier) than consensus.

But this seems like classic winner's curse territory to me. If consensus is that a player should go in the 70s but one other team would take him at say 32 so you need to take him at 30 to get him....well you may well be better off letting the other guys overpay instead. It's not a case of take-him-now-or-lose-him-forever, it's take-him-now-or-you-know-take-some-other-guy-instead. It is surely more likely in such case that you're wrong about the player than the 30 teams who have him ranked much lower?

One of my professors at b-school described the ethos of the investment firm he worked for as essentially "we don't believe anyone knows anything. We don't think we're smarter, but we don't believe the guy who really wants to buy a particular stock knows anything either. So we're happy to let him pay extra for the stocks he really wants and take the other side of that trade". It seems to me the draft is like that too.

There's also a lot of anecdotal evidence being point to here - it's important to understand that the value of the consensus big board is the consensus part. Wisdom of crowds stuff is a real thing.

So what makes me happy is when Bill (a) trades down because some other team absolutely has to have their guy (b) trades into next year because discount rates in the NFL are absolutely insane and any GM with job security should be exploiting the hell out of that (c) doesn't pick people way higher than they are expected to go.

There's a huge amount of noise when it comes to drafting so there will always be numerous counter-examples someone can point to. But why wouldn't you want to shift the percentages in your favor when you can?
Because you can't win if you don't play (the draft). That is where the supply of Tom Bradys and Vince Wilforks come from in a salary cap league. Winning the draft isn't accumulation of assets (though as you say leveraging the trade discount/premium is a real strategy for moving assets into the future), it's the mining of players sufficient enough (when combined with other roster construction) to be coached into a winning team.

The average team has 7 picks, meaning 224 picks total for the league (out of roughly 300 players who get evaluated?). If your only strategy is to evaluate the players and rank them according to your criteria and then pick the highest ranked player (or highest at your desired position) - then the draft plays out and you have what/who you have. Whittle away the list until you have removed the players you believe could never be successful on your team - within your scheme(s) or physically inadequate, etc. What happens if other teams end up with a list of 100-150 players players and your list is 30-35? When access to those players is limited you leverage the premiums for trading down, into future years and build future value. If your goal (or need) in any given year is to get your hands on as many of those higher on your list as possible - does the hypothetical of paying a premium by drafting a player 'early' matter if you are ending up with players you believe meet your criteria for winning? If over a few years you realize that you are only likely going to have access to maybe 1 of your top 10 players, and you find yourself looking at a chance to grab a 2nd player by giving up 2 round 4 picks, does it really matter where they are expected to go? Should our answer change if his evaluation of the player is the highest at his position in X years? Or if this is a player that occupies one of his most important positions on the field (regardless of how other teams value the position)? I don't know what criteria that BB is using for any of these scenarios/situations, but I do know the man is smart enough to evolve his roster construction strategies to leverage inefficiencies where ever he can. Example: waiting until day 2 or 3 of FA to land value free agents, until conversely most of the league last year was limited by cap and he had a surplus, so he spent lavishly). If there is leverage or an advantage to be had being on the opposite side of a trend, I'm comfortable that BB will exploit it if his situation allows it. Doesn't mean he won't fail miserably with a given strategy or in a given year, but if I'm looking for one person to build a roster for BB the coach - In Bill We Trust (the GM) is a decent choice.
 

P'tucket rhymes with...

Member
SoSH Member
Dec 12, 2006
11,624
The Coney Island of my mind
How do we reconcile the beating the Pats took from the draft experts for the Strange and Thornton picks with the fact that nearly all of the them thought Malik Willis was going to be a first rounder and many of them believed he'd be a top 10 pick?
It's behind the pay wall, I think, but a piece in the Athletic identifies Willis as one of the biggest steals of the draft. The consensus board cannot fail, it can pnly be failed.
 

RetractableRoof

tolerates intolerance
Lifetime Member
SoSH Member
Dec 1, 2003
3,836
Quincy, MA
No one really believes the pure efficient-market hypothesis is true. Obviously information isn't fully and equally known. Yet passive management still works.

It doesn't require perfect knowledge, or anything close to it, for it to be right to go with consensus over your own opinion that consensus is wrong.
Question, isn't one of the premises/assumptions of market systems that there is always another deal to be had, another day to succeed? What if the stock market was only open for 3 days a year? What changes when the person pulling the trigger knows he had a bad year last year and needs to save his job? What if it's his first year in the hot seat, and he's desperate to make an impression. Or the owner tells him he better get a marketable QB out of this draft? Or when the commodity performance isn't a known, but a human who might not thrive because playing for that coach "isn't fun"?

Isn't there a practical limit to approaching the draft like a closed system? Find imbalances and exploit them when you can, but aren't we far enough along in modeling systems that if it were possible to efficiently model the draft a team worth billions of dollars would have spent millions to gain that advantage?

Just questions, not challenges...
 
The consensus has less information than NFL teams and is vulnerable to herding. I think a better estimate for how other teams evaluate a player is how your team evaluates that player. If you think a guy is worth a late first round pick, assume the rest of the league sees them similarly. The best strategy is to draft players who fit your roster where you think they belong.
You might be right! Clearly your points on information and vulnerability to herding are both valid.

But you might not. There seems to be an assumption here that the consensus big board is not a good estimator of where other teams rate players. That could be true. But it seems to me that all the evidence we have points to the opposite - that the consensus actually is a decent estimator of how teams value players. Not perfect, far from it. But then teams are obviously not perfect either, far from it. After all, the correlation between consensus and actual draft position is high, notwithstanding all the noise.

You're basically saying "If I think this guy is #30 that probably means the other teams think this guy is also #30" [I'm not trying to put words in your mouth here, that's genuinely what I think you're saying so feel free to correct if not] whereas I'm saying "If I think this guy is #30 and the consensus board thinks this guy is #75 then the other teams probably think this guy is somewhere between those two numbers and I'm not sure where". If the actual answer is #32 then you've basically nailed it and yes, the concept of "reaching" is meaningless. If it's actually #65 then I'd generally prefer to avoid making those picks.

At the very least this implies you might want to be wary about picks where you're going strongly against consensus.

Of course, this is actually testable - if someone has historic consensus big board data its straightforward analysis to look at this stuff and see how "reach" picks fare v non-reaches.
 

rodderick

Member
SoSH Member
Apr 24, 2009
12,751
Belo Horizonte - Brazil
Okay, so I had the slowest of days at work yesterday and embarked on a little project. With all the talk about the "consensus" and how often the Patriots zig when other teams zag, I set out on a quest to figure one thing out: using the consensus big board at nflmockdraftdatabase, how much AV would the Patriots have gained/lost in the past five drafts had they picked the highest ranked available player by that measure instead of the guy they actually selected? It took me a while, but the results are now in! These will be spoilered because otherwise the post would be unbearably long.

2017 - plus 7 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Carl Lawson (EDGE) 11 career AV -> minus 11 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Caleb Brantley (DT) - 2 career AV -> plus 20 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by consensus Isaiah Ford (WR) - 5 career AV -> minus 2 AV

2018 - minus 2 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Harold Landry (EDGE) - 30 career AV -> minus 14 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Josh Jackson (CB) 7 career AV -> plus 14 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 1 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Tyrell Crosby (OT) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam - 0 career AV. Best available by consensus Equanimeous St. Brown (WR) - 4 career AV -> minus 4 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Quenton Meeks (CB) - 1 career AV -> plus 8 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> minus 5 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Allen Lazard (WR) - 14 career AV -> minus 9 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Desmond Harrison (OT) - 3 career AV -> minus 2 AV

2019 - Minus 35 AV

Pick 32 - Patriots select N’Keal Harry (WR) - 6 career AV. Best available by consensus Jawaan Taylor (OT) - 20 career AV -> minus 14 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus DK Metcalf (WR) - 30 career AV -> minus 27 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chauncey Gardner-Johnson (CB) - 12 career AV -> minus 2 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Hakeem Butler (WR) – 0 career AV -> plus 13 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Amani Oruwariye (CB) - 10 career AV -> minus 8 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus David Long (CB) - 9 career AV -> minus 9 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Deionte Thompson (S) - 4 career AV -> minus 3 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Kelvin Harmon (WR) – 3 career AV -> plus 6 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Emanuel Hall (WR) - 0 career AV -> plus 6 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Gerald Willis (DL) - 0 career AV -> plus 3 AV

2020 - plus 4 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Kristian Fulton (CB) - 5 career AV -> plus 4 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Josh Jones (OT) - 9 career AV -> minus 6 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Bryce Hall (CB) - 7 career AV -> minus 4 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 0 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Jacob Eason (QB) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Curtis Weaver (EDGE) - 0 career AV -> plus 0 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Donovan Peoples-Jones (WR) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Hunter Bryant (TE) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Eno Benjamin (RB) - 1 career AV -> plus 0 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Thaddeus Moss (TE) - 0 career AV -> plus 0 AV

2021 - minus 9 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Jeremiah Owusu-Karamoah (LB) - 4 career AV -> minus 1 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Jabril Cox (LB) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Daviyon Nixon (DL) – 0 career AV -> plus 5 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Trey Smith (G) - 10 career AV -> minus 10 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Quincy Roche (EDGE) - 2 career AV -> minus 2 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Marvin Wilson (DL) - 0 career AV -> plus 0 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Dylan Moses (LB) – 0 career AV -> plus 0 AV

So that's 35 AV lost over a period of five drafts just by going "BPA by the big board" instead of the guy they ended up taking. Now, I'm fully aware that this isn't a complete picture and that measuring AV of 2020 and 2021 draft picks especially is a little foolish considering many of those guys could still develop/get injured and the equation could change rapidly in the coming years. So they basically ended up pretty much near what you'd expect in every season, aside from 2019, which was a bloodbath (kinda tracks with my overall perception too). In essence: four meh drafts, one terrible draft.

After I finished this up, though, I began thinking this was an incomplete picture. Just going BPA makes little sense. The 2019 BPA draft had them taking 3 CBs and 4 WRs, for instance, which is insane. So I decided to refine the comparison a little and find out the AV gained/lost by the Patriots had they simply gone BPA at the position they ended up picking over the past five drafts. Here are the results.

2017 = plus 22 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Antonio Garcia (OT) - 0 career AV -> plus 0 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Joe Mathis (DE) - 0 career AV -> plus 22 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by Conor McDermott - 3 career AV -> plus 0 AV

2018 = plus 19 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Connor Williams (OT) - 23 career AV -> minus 7 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 19 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Isaiah Oliver (CB) - 11 career AV -> minus 8 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Genard Avery (LB) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam (LB) - 0 career AV. Best available by consensus Jack Cichy (LB) - 0 career AV -> plus 0 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> plus 4 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Kurt Benkert (QB) - 0 career AV -> plus 0 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Holton Hill (CB) - 4 career AV -> plus 1 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Marcus Baugh (TE) - 0 career AV -> plus 1 AV

2019 = minus 31 AV

Pick 32 – Patriots select N’Keal Harry (WR) - 6 career AV. Best available by DK Metcalf (WR) - 30 career AV -> minus 24 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus Greedy Williams (CB) - 7 career AV -> minus 4 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chase Winovich (DE) - 10 career AV -> plus 0 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Damien Harris (RB) - 13 career AV -> plus 0 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Yodney Cajuste (OT) - 2 career AV -> plus 0 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus Michael Jordan (G) - 11 career AV -> minus 11 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Jarrett Stidham (QB) - 1 career AV -> plus 0 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Charles Omenihu (DL) – 4 career AV -> plus 5 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Mitch Wishnowsky (P) - 6 career AV -> plus 0 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Saivion Smith (CB) - 0 career AV -> plus 3 AV

2020 = plus 3 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Grant Delpit (S) - 3 career AV -> plus 6 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Terrell Lewis (LB) - 3 career AV -> plus 0 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Jonathan Greenard (LB) - 6 career AV -> minus 3 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Adam Trautman (TE) – 4 career AV -> minus 4 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Brycen Hopkins (TE) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Rodrigo Blankenship (K) - 3 career AV -> minus 3 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Josh Simpson (OL) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Markus Bailey (LB) - 4 career AV -> minus 3 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Darryl Williams (IOL) - 0 career AV -> plus 0 AV

2021 = minus 2 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Christian Barmore (DT) – 3 career AV -> plus 0 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Carlos Basham (DE) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Kenneth Gainwell (RB) – 5 career AV -> plus 0 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Dylan Moses (LB) - 0 career AV -> plus 0 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Ar’Darius Washington (S) - 0 career AV -> plus 0 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Stone Forsythe (OL) - 1 career AV -> minus 1 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Cade Johnson (WR) – 0 career AV -> plus 0 AV

The Pats have gone from negative 35 AV to plus 12 AV over the past five years with the positional adjustment. The 2019 draft is still god awful, but now 2017 and 2018 look much better! But in what is the best representation of how weird this exercise is, the surplus value that accounted for this reversal came mostly from the selections of Sony Michel and Deatrich Wise Jr., two players I don't think most Pats fans would rank as particularly great draft picks.

So there you have it. What can we take from this? Who knows. Drafting is hard, mostly. And even role players that aren't stars or entrenched starters are generally valuable even when picked reasonably high.

For funsies, I thought it'd be fair to go back and look at how the 2016 Patriots draft (which I consider to be a very good draft) would have fared under the "what if they had simply gone BPA" parameters and the results are simply ridiculous:

2016 = plus 114 AV

Pick 60 – Patriots select Cyrus Jones (CB) - 4 career AV. Best available by consensus Andrew Billings (DT) 14 career AV -> minus 10 AV

Pick 78 – Patriots select Joe Thuney (G) - 48 career AV. Best available by consensus Jalen Mills (S) 26 career AV -> plus 22 AV

Pick 91 – Patriots select Jacoby Brissett (QB) - 24 career AV. Best available by consensus Kevin Hogan (QB) - 2 career AV -> plus 22 AV

Pick 96 – Patriots select Vincent Valentine (DT) - 2 career AV. Best available by consensus Joshua Perry (LB) - 1 career AV -> plus 1 AV

Pick 112 – Patriots select Malcolm Mitchell - 4 career AV. Best available by consensus Jeremy Cash (S) - 0 career AV -> plus 4 AV

Pick 208 – Patriots select Kamu Grugier-Hill (LB) - 18 career AV. Best available by consensus Charome Peake (WR) - 1 career AV -> plus 17 AV

Pick 214 – Patriots select Elandon Roberts (LB) - 34 career AV. Best available by consensus Scooby Wright III (LB) - 1 career AV -> plus 33 AV

Pick 221 – Patriots select Ted Karras (G) - 25 career AV. Best available by consensus Kenny Lawler (WR) - 0 career AV -> plus 25 AV

Pick 225 – Patriots select Devin Lucien (WR) - 0 career AV. Best available by consensus Bryce Williams (TE) - 0 career AV -> plus 0 AV

That's a whopping surplus of 114 AV in a single draft. They basically got Tony Romo's entire career (115 AV) just on added value when compared to the players they would have drafted by going big board BPA.
 
Last edited:
Question, isn't one of the premises/assumptions of market systems that there is always another deal to be had, another day to succeed? What if the stock market was only open for 3 days a year? What changes when the person pulling the trigger knows he had a bad year last year and needs to save his job? What if it's his first year in the hot seat, and he's desperate to make an impression. Or the owner tells him he better get a marketable QB out of this draft? Or when the commodity performance isn't a known, but a human who might not thrive because playing for that coach "isn't fun"?

Just questions, not challenges...
I think these are good questions.

I think one of the advantages the Pats have is that Bill isn't coaching/GMing for his job each year. Obviously some guys are. If a GM NEEDS to win now to keep his job, its completely rational for him to overpay for something that will pay off this year but cost in the future. Bill isn't in that position, and therefore ought to be in a position to exploit those people who are.

I think the draft is enough of a crapshoot that the noise hides a lot of the smart/dumb stuff. It's not like you can do everything right (trade down for more value, accumulate mid-round picks who provide good value-for-money performance v high-round picks (good but expensive) and low-round picks (not very good), trade for future picks etc) and therefore win. I see it more like you're 50/50 to have a better than average draft as default...you're 55/45 if you do things right and 45/55 if you do things wrong. There's just a lot of noise.

But 55/45 over the long term does accumulate a worthwhile advantage.

Isn't there a practical limit to approaching the draft like a closed system? Find imbalances and exploit them when you can, but aren't we far enough along in modeling systems that if it were possible to efficiently model the draft a team worth billions of dollars would have spent millions to gain that advantage?

Just questions, not challenges...
That teams aren't doing things that seem like obviously good ideas to me when the rewards for doing them should be high is a huge problem for my thinking. Logically it should therefore be much more likely that I'm wrong than they are. I believed this to be the case for a long time.

Then the financial crisis happened and I got to see the way people at the very top were thinking and oh boy no, sometimes people who you'd think were (a) smart (b) informed (c) heavily incentivised to do smart things are still doing really really dumb shit. So now I consider it possible that teams are just doing dumb shit too.
 

ColonelMustard

Member
SoSH Member
Oct 23, 2006
219
You might be right! Clearly your points on information and vulnerability to herding are both valid.

But you might not. There seems to be an assumption here that the consensus big board is not a good estimator of where other teams rate players. That could be true. But it seems to me that all the evidence we have points to the opposite - that the consensus actually is a decent estimator of how teams value players. Not perfect, far from it. But then teams are obviously not perfect either, far from it. After all, the correlation between consensus and actual draft position is high, notwithstanding all the noise.
Your premise here is that consensus reflects accurate "market value" because of the passive management theory that the "market is always right". This is a wrong premise because the markets are not comparable. Passive management works on the principle of efficient markets.

More importantly, passive investment strategy applies to public markets because they facilitate information (10k, 10q, etc.). The market for NFL draftees is more akin to investing in start-ups. You know literally nothing and you're investing in management teams. While the financial results of VC can be argued, I would not recommend taking a passive approach to investing in startups.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,054
Hingham, MA
It's behind the pay wall, I think, but a piece in the Athletic identifies Willis as one of the biggest steals of the draft. The consensus board cannot fail, it can pnly be failed.
It occurred to me this morning what while we have all been discussing the concept of a "reach" in relation to the Pats picks, I think I hate the term "steal" even more. A "steal" only happens because no teams valued a player as highly as the consensus / media ranked him. So Willis was ranked whatever he was, then taken a hell of a lot later, therefore the pundits proclaim STEAL! Or like our infamous Pats / Ravens Sergio Kindle example from year back. You can't call a guy a steal until they actually perform. I think there is actually more merit in the term reach than a steal. If Strange TRULY wasn't going to be picked until (making this up) the 3rd round, then the Pats could have waited until 54, and therefore they reached, even if he becomes a Pro Bowl player - they unnecessarily drafted him early.

But you can't make the same argument about a "steal". Sergio Kindle was not, in fact, a steal. He was a completely wasted pick. Julian Edelman was a steal. Tom Brady was a steal. Aaron Rodgers, Patrick Mahomes, Russell Wilson, and Lamar Jackson were steals. But only because their performance in the NFL greatly exceeded the expectation given where they were picked, not because of where they were drafted in comparison to the big board.

Hope this makes sense. It makes sense in my head.
 

Silverdude2167

Member
SoSH Member
Oct 9, 2006
4,682
Amstredam
Okay, so I had the slowest of days at work yesterday and embarked on a little project. With all the talk about the "consensus" and how often the Patriots zig when other teams zag, I set out on a quest to figure one thing out: using the consensus big board at nflmockdraftdatabase, how much AV would the Patriots have gained/lost in the past five drafts had they picked the highest ranked available player by that measure instead of the guy they actually selected? It took me a while, but the results are now in! These will be spoilered because otherwise the post would be unbearably long.

2017 - plus 7 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Carl Lawson (EDGE) 11 career AV -> minus 11 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Caleb Brantley (DT) - 2 career AV -> plus 20 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by consensus Isaiah Ford (WR) - 5 career AV -> minus 2 AV

2018 - minus 2 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Harold Landry (EDGE) - 30 career AV -> minus 14 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Josh Jackson (CB) 7 career AV -> plus 14 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 1 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Tyrell Crosby (OT) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam - 0 career AV. Best available by consensus Equanimeous St. Brown (WR) - 4 career AV -> minus 4 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Quenton Meeks (CB) - 1 career AV -> plus 8 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> minus 5 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Allen Lazard (WR) - 14 career AV -> minus 9 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Desmond Harrison (OT) - 3 career AV -> minus 2 AV

2019 - Minus 35 AV

Pick 32 - Patriots select N’Keal Harry (WR) - 6 career AV. Best available by consensus Jawaan Taylor (OT) - 20 career AV -> minus 14 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus DK Metcalf (WR) - 30 career AV -> minus 27 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chauncey Gardner-Johnson (CB) - 12 career AV -> minus 2 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Hakeem Butler (WR) – 0 career AV -> plus 13 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Amani Oruwariye (CB) - 10 career AV -> minus 8 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus David Long (CB) - 9 career AV -> minus 9 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Deionte Thompson (S) - 4 career AV -> minus 3 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Kelvin Harmon (WR) – 3 career AV -> plus 6 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Emanuel Hall (WR) - 0 career AV -> plus 6 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Gerald Willis (DL) - 0 career AV -> plus 3 AV

2020 - plus 4 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Kristian Fulton (CB) - 5 career AV -> plus 4 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Josh Jones (OT) - 9 career AV -> minus 6 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Bryce Hall (CB) - 7 career AV -> minus 4 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 0 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Jacob Eason (QB) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Curtis Weaver (EDGE) - 0 career AV -> plus 0 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Donovan Peoples-Jones (WR) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Hunter Bryant (TE) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Eno Benjamin (RB) - 1 career AV -> plus 0 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Thaddeus Moss (TE) - 0 career AV -> plus 0 AV

2021 - minus 9 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Jeremiah Owusu-Karamoah (LB) - 4 career AV -> minus 1 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Jabril Cox (LB) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Daviyon Nixon (DL) – 0 career AV -> plus 5 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Trey Smith (G) - 10 career AV -> minus 10 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Quincy Roche (EDGE) - 2 career AV -> minus 2 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Marvin Wilson (DL) - 0 career AV -> plus 0 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Dylan Moses (LB) – 0 career AV -> plus 0 AV

So that's 35 AV lost over a period of five drafts just by going "BPA by the big board" instead of the guy they ended up taking. Now, I'm fully aware that this isn't a complete picture and that measuring AV of 2020 and 2021 draft picks especially is a little foolish considering many of those guys could still develop/get injured and the equation could change rapidly in the coming years. So they basically ended up pretty much near what you'd expect in every season, aside from 2019, which was a bloodbath (kinda tracks with my overall perception too). In essence: four meh drafts, one terrible draft.

After I finished this up, though, I began thinking this was an incomplete picture. Just going BPA makes little sense. The 2019 BPA draft had them taking 3 CBs and 4 WRs, for instance, which is insane. So I decided to refine the comparison a little and find out the AV gained/lost by the Patriots had they simply gone BPA at the position they ended up picking over the past five drafts. Here are the results.

2017 = plus 22 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Antonio Garcia (OT) - 0 career AV -> plus 0 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Joe Mathis (DE) - 0 career AV -> plus 22 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by Conor McDermott - 3 career AV -> plus 0 AV

2018 = plus 19 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Connor Williams (OT) - 23 career AV -> minus 7 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 19 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Isaiah Oliver (CB) - 11 career AV -> minus 8 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Genard Avery (LB) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam (LB) - 0 career AV. Best available by consensus Jack Cichy (LB) - 0 career AV -> plus 0 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> plus 4 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Kurt Benkert (QB) - 0 career AV -> plus 0 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Holton Hill (CB) - 4 career AV -> plus 1 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Marcus Baugh (TE) - 0 career AV -> plus 1 AV

2019 = minus 31 AV

Pick 32 – Patriots select N’Keal Harry (WR) - 6 career AV. Best available by DK Metcalf (WR) - 30 career AV -> minus 24 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus Greedy Williams (CB) - 7 career AV -> minus 4 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chase Winovich (DE) - 10 career AV -> plus 0 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Damien Harris (RB) - 13 career AV -> plus 0 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Yodney Cajuste (OT) - 2 career AV -> plus 0 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus Michael Jordan (G) - 11 career AV -> minus 11 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Jarrett Stidham (QB) - 1 career AV -> plus 0 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Charles Omenihu (DL) – 4 career AV -> plus 5 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Mitch Wishnowsky (P) - 6 career AV -> plus 0 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Saivion Smith (CB) - 0 career AV -> plus 3 AV

2020 = plus 3 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Grant Delpit (S) - 3 career AV -> plus 6 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Terrell Lewis (LB) - 3 career AV -> plus 0 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Jonathan Greenard (LB) - 6 career AV -> minus 3 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Adam Trautman (TE) – 4 career AV -> minus 4 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Brycen Hopkins (TE) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Rodrigo Blankenship (K) - 3 career AV -> minus 3 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Josh Simpson (OL) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Markus Bailey (LB) - 5 career AV -> minus 3 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Darryl Williams (IOL) - 0 career AV -> plus 0 AV

2021 = minus 2 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Christian Barmore (DT) – 3 career AV -> plus 0 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Carlos Basham (DE) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Kenneth Gainwell (RB) – 5 career AV -> plus 0 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Dylan Moses (LB) - 0 career AV -> plus 0 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Ar’Darius Washington (S) - 0 career AV -> plus 0 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Stone Forsythe (OL) - 1 career AV -> minus 1 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Cade Johnson (WR) – 0 career AV -> plus 0 AV

The Pats have gone from negative 35 AV to plus 12 AV over the past five years with the positional adjustment. The 2019 draft is still god awful, but now 2017 and 2018 look much better! But in what is the best representation of how weird this exercise is, the surplus value that accounted for this reversal came mostly from the selections of Sony Michel and Deatrich Wise Jr., two players I don't think most Pats fans would rank as particularly great draft picks.

So there you have it. What can we take from this? Who knows. Drafting is hard, mostly. And even role players that aren't stars or entrenched starters are generally valuable even when picked reasonably high.

For funsies, I thought it'd be fair to go back and look at how the 2016 Patriots draft (which I consider to be a very good draft) would have fared under the "what if they had simply gone BPA" parameters and the results are simply ridiculous:

2016 = plus 114 AV

Pick 60 – Patriots select Cyrus Jones (CB) - 4 career AV. Best available by consensus Andrew Billings (DT) 14 career AV -> minus 10 AV

Pick 78 – Patriots select Joe Thuney (G) - 48 career AV. Best available by consensus Jalen Mills (S) 26 career AV -> plus 22 AV

Pick 91 – Patriots select Jacoby Brissett (QB) - 24 career AV. Best available by consensus Kevin Hogan (QB) - 2 career AV -> plus 22 AV

Pick 96 – Patriots select Vincent Valentine (DT) - 2 career AV. Best available by consensus Joshua Perry (LB) - 1 career AV -> plus 1 AV

Pick 112 – Patriots select Malcolm Mitchell - 4 career AV. Best available by consensus Jeremy Cash (S) - 0 career AV -> plus 4 AV

Pick 208 – Patriots select Kamu Grugier-Hill (LB) - 18 career AV. Best available by consensus Charome Peake (WR) - 1 career AV -> plus 17 AV

Pick 214 – Patriots select Elandon Roberts (LB) - 34 career AV. Best available by consensus Scooby Wright III (LB) - 1 career AV -> plus 33 AV

Pick 221 – Patriots select Ted Karras (G) - 25 career AV. Best available by consensus Kenny Lawler (WR) - 0 career AV -> plus 25 AV

Pick 225 – Patriots select Devin Lucien (WR) - 0 career AV. Best available by consensus Bryce Williams (TE) - 0 career AV -> plus 0 AV

That's a whopping surplus of 114 AV in a single draft. They basically got Tony Romo's entire career (115 AV) just on added value when compared to the players they would have drafted by going big board BPA.
This was awesome, thank you.

One completely unrelated question, how has a Guard racked up 10 career AV in one year? He has been only slightly less valuable than Mac? Maybe I don't understand how AV works.

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Trey Smith (G) - 10 career AV -> minus 10 AV
 

rodderick

Member
SoSH Member
Apr 24, 2009
12,751
Belo Horizonte - Brazil
This was awesome, thank you.

One completely unrelated question, how has a Guard racked up 10 career AV in one year? He has been only slightly less valuable than Mac? Maybe I don't understand how AV works.

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Trey Smith (G) - 10 career AV -> minus 10 AV
AV isn't really all that scientific of a way to measure value, to be sure, especially when utilized in the manner I just did: trying to determine individual player contributions. It's just the only tool I could find to give a quick value figure without going into the more subjective side of player analysis for my experiment. From PFR's own explanation of it:

"AV is not meant to be a be-all end-all metric. Football stat lines just do not come close to capturing all the contributions of a player the way they do in baseball and basketball. If one player is a 16 and another is a 14, we can't be very confident that the 16AV player actually had a better season than the 14AV player. But I am pretty confident that the collection of all players with 16AV played better, as an entire group, than the collection of all players with 14AV."

"Essentially, AV is a substitute for --- and a significant improvement upon, in my opinion --- metrics like 'number of seasons as a starter' or 'number of times making the pro bowl' or the like. You should think of it as being essentially like those two metrics, but with interpolation in between. That is, 'number of seasons as a starter' is a reasonable starting point if you're trying to measure, say, how good a particular draft class is, or what kind of player you can expect to get with the #13 pick in the draft. But obviously some starters are better than others. Starters on good teams are, as a group, better than starters on bad teams. Starting WRs who had lots of receiving yards are, as a group, better than starting WRs who did not have many receiving yards. Starters who made the pro bowl are, as a group, better than starters who didn't, and so on. And non-starters aren't worthless, so they get some points too.
"
 

RetractableRoof

tolerates intolerance
Lifetime Member
SoSH Member
Dec 1, 2003
3,836
Quincy, MA
That teams aren't doing things that seem like obviously good ideas to me when the rewards for doing them should be high is a huge problem for my thinking. Logically it should therefore be much more likely that I'm wrong than they are. I believed this to be the case for a long time.

Then the financial crisis happened and I got to see the way people at the very top were thinking and oh boy no, sometimes people who you'd think were (a) smart (b) informed (c) heavily incentivised to do smart things are still doing really really dumb shit. So now I consider it possible that teams are just doing dumb shit too.
Panic / survival instinct are stronger than calmness / logical thought - especially in the moment. Sometimes it is as simple as that. It's why I asked about the 3 day market - it's got all kinds of psychological factors built in - FOMO being the most obvious.
 
Panic / survival instinct are stronger than calmness / logical thought - especially in the moment. Sometimes it is as simple as that. It's why I asked about the 3 day market - it's got all kinds of psychological factors built in - FOMO being the most obvious.
No question. But the people I was referring to were not panicking or trying to survive. They were just shockingly unaware of what they didn't know/understand and how important some of those things might be. But those are stories for another board.
 

Shelterdog

Well-Known Member
Lifetime Member
SoSH Member
Feb 19, 2002
15,375
New York City
[/QUOTE]
There are a lot of people making the point that it only takes one othe team to be looking at "your" player to mean you need to take him earlier (or in a couple of cases with the Pats this year much earlier) than consensus.

But this seems like classic winner's curse territory to me. If consensus is that a player should go in the 70s but one other team would take him at say 32 so you need to take him at 30 to get him....well you may well be better off letting the other guys overpay instead. It's not a case of take-him-now-or-lose-him-forever, it's take-him-now-or-you-know-take-some-other-guy-instead. It is surely more likely in such case that you're wrong about the player than the 30 teams who have him ranked much lower?

One of my professors at b-school described the ethos of the investment firm he worked for as essentially "we don't believe anyone knows anything. We don't think we're smarter, but we don't believe the guy who really wants to buy a particular stock knows anything either. So we're happy to let him pay extra for the stocks he really wants and take the other side of that trade". It seems to me the draft is like that too.

There's also a lot of anecdotal evidence being point to here - it's important to understand that the value of the consensus big board is the consensus part. Wisdom of crowds stuff is a real thing.

So what makes me happy is when Bill (a) trades down because some other team absolutely has to have their guy (b) trades into next year because discount rates in the NFL are absolutely insane and any GM with job security should be exploiting the hell out of that (c) doesn't pick people way higher than they are expected to go.

There's a huge amount of noise when it comes to drafting so there will always be numerous counter-examples someone can point to. But why wouldn't you want to shift the percentages in your favor when you can?
As others have noted, the passive management analogy doesn't work here. In passive management you're saying I'm going to ride along with all the investors, the winners and losers alike, let them sort it out, and win on fees--and then you're using a benchmarket where the active investors show their commitment to their views by actually investing money. But in the NFL draft the consenus we're talking about isn't the actual consensus view of NFL teams, it's more like the stock picks in Barron's or Kiplingers--info put together by journalists that to some extent reflects what active managers are doing but also reflects what particularly amateurs think, what professional wants to sell into the market, what the companies themselves are promoting, etc.

I suspect that BB has a pretty good idea of the actual consensus--he is a football information vacuum and has a lot of relationships with agents, college coaches, etc---but perhaps is more willing than others to pull the trigger on a Strange or Thornton than others are because he is unusually secure in is position.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
Isn't this more the area where Caserio and Groh assemble together their scouting reports and present them to Bill for approval? How much credit do we attribute to Bill and how much to Caserio/Groh?
(Yes, I know Caserio left, l meant that particular position, whatever it's called that's not actually a GM position).
On most teams, the coaching staff gets involved with the draft to varying degrees after the season ends. The amount of input and authority is going to vary. Mike Lombardi on his last GM Shuffle podcast estimated that Belichick personally writes up 500 prospects every year, so he is not passive here.

One of my professors at b-school described the ethos of the investment firm he worked for as essentially "we don't believe anyone knows anything. We don't think we're smarter, but we don't believe the guy who really wants to buy a particular stock knows anything either. So we're happy to let him pay extra for the stocks he really wants and take the other side of that trade". It seems to me the draft is like that too.
You can always elect not to trade. In the draft, eventually you have to pick players. If you're on the clock and you don't have any good tradeback offers and Cole Strange is number one on your board, what do you do?

But you might not. There seems to be an assumption here that the consensus big board is not a good estimator of where other teams rate players. That could be true. But it seems to me that all the evidence we have points to the opposite - that the consensus actually is a decent estimator of how teams value players. Not perfect, far from it. But then teams are obviously not perfect either, far from it. After all, the correlation between consensus and actual draft position is high, notwithstanding all the noise.
In the aggregate, the consensus boards do a good job, but that doesn't mean they do a good job with all players. Media boards are going to be much more plugged in to the league consensus on, say, a three-year starter at Clemson than a small school player like Cole Strange. And teams are going to have better information on this than the media.
 

DavidTai

Member
SoSH Member
Dec 18, 2003
1,242
Herndon, VA
Doesn't analyzing the draft results get complicated by that some players are picked that are bad fits for one team and better for another, so their value goes up -late- but they have lost a lot of value being with the wrong team? (Thinking of Kyle Van Noy, here.)
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,054
Hingham, MA
Doesn't analyzing the draft results get complicated by that some players are picked that are bad fits for one team and better for another, so their value goes up -late- but they have lost a lot of value being with the wrong team? (Thinking of Kyle Van Noy, here.)
I would argue that teams should get credit for drafting the player, even if they prosper elsewhere. After all, the team identified the talent; developing the talent is on the coaching staff, not the front office.

So I would give the Pats credit for drafting guys like Berrios and Darius Butler, even if all their AV has come from elsewhere. And similarly I would give credit to the Lions for drafting Van Noy.

The goal is to identify the talent in the draft. I realize the goal is also to turn that talent into production, but it's not a 1:1 relationship IMO.
 

Shelterdog

Well-Known Member
Lifetime Member
SoSH Member
Feb 19, 2002
15,375
New York City
In the aggregate, the consensus boards do a good job, but that doesn't mean they do a good job with all players. Media boards are going to be much more plugged in to the league consensus on, say, a three-year starter at Clemson than a small school player like Cole Strange. And teams are going to have better information on this than the media.
The media boards are more or less good but also get influenced by agents, whichever area scouts are talkative with the press, etc. Anyone remember the story from years ago about Mel Kiper helping to recruit a college defensive lineman Willi Howard from Stanford for a particular agent? https://www.ruleoftree.com/2010/10/13/1749056/confessions-of-an-agent-featuring-willie-howard
 
Last edited:

DavidTai

Member
SoSH Member
Dec 18, 2003
1,242
Herndon, VA
So I would give the Pats credit for drafting guys like Berrios and Darius Butler, even if all their AV has come from elsewhere. And similarly I would give credit to the Lions for drafting Van Noy.

The goal is to identify the talent in the draft. I realize the goal is also to turn that talent into production, but it's not a 1:1 relationship IMO.
My thought was that there are multiple players whose careers return little or no value because they were misused and then their careers ended before they could shift those AV, so even if the front office correctly identified them, they might still not show up or be shown as 'little value'.

If, say, Van Noy hadn't gone to the Patriots from Detroit, but had gone to Cleveland, and gained no value there, and retired, his actual value would have slipped through the cracks. So I'd give more credit towards the organization that -gained- those AV values, but how?

Treat it separately from the draft? Sort of like a "Free Agency/Trade AV gain" to measure the effectiveness of an organization overall?
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,054
Hingham, MA
My thought was that there are multiple players whose careers return little or no value because they were misused and then their careers ended before they could shift those AV, so even if the front office correctly identified them, they might still not show up or be shown as 'little value'.

If, say, Van Noy hadn't gone to the Patriots from Detroit, but had gone to Cleveland, and gained no value there, and retired, his actual value would have slipped through the cracks. So I'd give more credit towards the organization that -gained- those AV values, but how?

Treat it separately from the draft? Sort of like a "Free Agency/Trade AV gain" to measure the effectiveness of an organization overall?
This is why it's all very very difficult to analyze. Fit and team matters a ton. Just a ton. If Mahomes had been drafted by the Bears maybe he'd be out of the league right now. We have zero way of knowing.
 

Cousin Walter

New Member
Jun 26, 2006
168
Basement
I would argue that teams should get credit for drafting the player, even if they prosper elsewhere. After all, the team identified the talent; developing the talent is on the coaching staff, not the front office.

So I would give the Pats credit for drafting guys like Berrios and Darius Butler, even if all their AV has come from elsewhere. And similarly I would give credit to the Lions for drafting Van Noy.

The goal is to identify the talent in the draft. I realize the goal is also to turn that talent into production, but it's not a 1:1 relationship IMO.
What good is evaluating a player well if there is no value added? Take Ted Larsen, a center drafted by the Pats in the sixth round in 2010. He didn't even make the team out of training camp but went on to have a decent enough career. Yes, he was properly identified but he took reps and a roster spot away from another guy. The team got no trade value from him and could have traded the pick for a useful player or into a future year. Yes, sixth-rounders are crapshoots but I would argue that selecting a guy that is then cut makes the team worse off than just forfeiting the pick altogether.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,054
Hingham, MA
What good is evaluating a player well if there is no value added? Take Ted Larsen, a center drafted by the Pats in the sixth round in 2010. He didn't even make the team out of training camp but went on to have a decent enough career. Yes, he was properly identified but he took reps and a roster spot away from another guy. The team got no trade value from him and could have traded the pick for a useful player or into a future year. Yes, sixth-rounders are crapshoots but I would argue that selecting a guy that is then cut makes the team worse off than just forfeiting the pick altogether.
There are only so many roster spots.
 

E5 Yaz

Transcends message boarding
Lifetime Member
SoSH Member
Apr 25, 2002
90,014
Oregon
Yes, sixth-rounders are crapshoots but I would argue that selecting a guy that is then cut makes the team worse off than just forfeiting the pick altogether.
"In the sixth round of the 2010 NFL Draft, the New England Patriots have decided not to select anyone ... and will just forfeit the pick."

Kiper: Excellent value in this spot
 
Last edited:
As others have noted, the passive management analogy doesn't work here. In passive management you're saying I'm going to ride along with all the investors, the winners and losers alike, let them sort it out, and win on fees--and then you're using a benchmarket where the active investors show their commitment to their views by actually investing money. But in the NFL draft the consenus we're talking about isn't the actual consensus view of NFL teams, it's more like the stock picks in Barron's or Kiplingers--info put together by journalists that to some extent reflects what active managers are doing but also reflects what particularly amateurs think, what professional wants to sell into the market, what the companies themselves are promoting, etc.
This is true - the "consensus big board" isn't the NFL GM consensus (which obviously can't be known). But it's also not an entirely unrelated, uninformed or random set of views. It's highly correlated with draft order. It may be better correlated with draft order than the Pats board is, for example.

Maybe you think it's obvious that taking the best player available on your board is a better strategy than taking the best player available on the consensus board. I'm happy to acknowledge that might be true. I just don't find it obviously true (Clearly in practice positional considerations/fit are important too).

As such, I'm more comfortable when Bill takes players that the consensus thought were going to be good than I am when he takes players that the consensus did not think were going to be good. So I'm not as ready to think this is going to be a good Pats draft because Bill got the players he wanted as some people on this board are.

You can always elect not to trade. In the draft, eventually you have to pick players. If you're on the clock and you don't have any good tradeback offers and Cole Strange is number one on your board, what do you do?
Sure, you absolutely don't get what you want from a draft because you're competing with 31 other teams. So you do the best you can do. In your example, that's probably to take Strange. Fine.

The issue is more - how much effort do you make to trade down in advance of such a situation? How much adjustment do you make to your own evaluation of a player if you think the market values him very differently? My observation is that generally speaking, on average, making this claim as bland as I possibly can to avoid riling people unnecessarily....when someone thinks they know something the market doesn't, it's more likely that the someone is wrong than that the market is wrong.

How about this situation. You're picking #30. There's a player you have #20 on your board available that you think consensus has #50 and a player you have #25 that the consensus also has #25. My impression is that most people commenting here think taking the #20 guy is the right thing to do, and that some people think doing anything other than taking that guy is ridiculous. I'm open to the idea that taking the #25 guy is actually the better plan, let alone a plan worth considering.

In the aggregate, the consensus boards do a good job, but that doesn't mean they do a good job with all players. Media boards are going to be much more plugged in to the league consensus on, say, a three-year starter at Clemson than a small school player like Cole Strange. And teams are going to have better information on this than the media.
This seems likely to be true, and hopefully will be in this case.

To be clear, I'm happy to trust Bill to do what he thinks is best and live with the results. I'd just like to measure those results when he does unconventional stuff so in the future I could have more confidence that in these particular instances the unconventional has actually helped. Because it's not clear to me that it has.
 

Shelterdog

Well-Known Member
Lifetime Member
SoSH Member
Feb 19, 2002
15,375
New York City
To be clear, I'm happy to trust Bill to do what he thinks is best and live with the results. I'd just like to measure those results when he does unconventional stuff so in the future I could have more confidence that in these particular instances the unconventional has actually helped. Because it's not clear to me that it has.
Just based on memory his least conventional picks --- Mankins Vollmer Wilson Jordan Richards Vincent Valentine Duke Dawson -have probably worked out better on average than his conventional ones; Mankins and Vollmer were just such incredibly good picks that they obscure that Jordan Richards was about best an average late second round pick and Dawson and Valentine bad picks..
 

mulluysavage

Member
SoSH Member
Jul 19, 2005
714
Reads threads backwards
Certainly the Pats consider needs at positions andZappeZappe
pects at positions.

But I think more than other teams, they say "this guy is a really interesting football player and/or athlete, I wonder what he could do for us," at their position - or another one. This is at odds a bit with a logic of "of best at a position=sooner."

In past drafts, see examples Edelman, McCourty, Ebner or UDFAs Neal, Olszewski

2022 see Marcus Jones, Zappe

Pundits may say "wait, who?!?" Or "reach," but the Pats are drafting by a different logic than theirs.
 

GB5

New Member
Aug 26, 2013
675
I think this is going to be a really easy draft for outsiders to crush, with not much upside for reversal at least in the first year:

1. Strange: If he plays well, you will never hear his name. Guards also dont get a let of notoriety, unless they get beaten.
2. Thornton: He has a chance to be a standout in this class. Rookie..deep threat in a quick throw offense..
3. M. Jones: Has a chance to shine
4. J. Jones: Wouldnt stun me if he gets a Wade like development/maturation year.
5. Strong: If White is injured, he may get in to work in the passing game, otherwise does he get the Vereen/White redshirt year?
6. Zappe: If he sees game action this year, its because something has gone dramatically wrong
7. Harris: See Strong. Coming of back surgery, I could see an IR year.
8. Roberts: Undersized DT. Roster spot is there for the taking, but maybe a year of strength and development?
9. Hines: Developmental guard. Again, if you hit a home run, nobody mentions his name.
10. Steuber: See Hines.
 

BigJimEd

Member
SoSH Member
Jan 4, 2002
4,432
But it's also not an entirely unrelated, uninformed or random set of views. It's highly correlated with draft order.
Has there been any study on this? I mean it is certainly not unrelated. I believe it is fairly highly correlated early but curious just how highly and how much that changes as you get to 25th, 50th, 75th ranked prospect. I know someone posted the 82% of the media top 100 or so were taken in the first 3 rounds (something like that) but that includes guys like Strange, Willis, Dean, Corral, etc. that most in the media would call a steal/reach so that doesn't seem like a very good metric.


You mention fit and I do think that is huge factor that likely creates a much greater deviation among NFL boards than media boards. Teams have different needs based on scheme, roster and host of other factors. Those things are reflected in each teams boards but obviously are not reflected in media lists. Just because a player might be 35 on one board and 90 on another team's board doesn't mean that either team has misevaluated that player. some team9s) might not even have that player on their board at all.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
This is true - the "consensus big board" isn't the NFL GM consensus (which obviously can't be known). But it's also not an entirely unrelated, uninformed or random set of views. It's highly correlated with draft order. It may be better correlated with draft order than the Pats board is, for example.
Well yes, because there is more groupthink outside the NFL than within it. Teams in general are more inconsistent than the media.

Maybe you think it's obvious that taking the best player available on your board is a better strategy than taking the best player available on the consensus board. I'm happy to acknowledge that might be true. I just don't find it obviously true (Clearly in practice positional considerations/fit are important too).
The parenthetical here is really important. The Patriots literally have written definitions for their standards at positions and scout to those. They have a small board because they are very focused on finding players that fit. That's why you rarely see them make a mistake like Detroit did with Kyle Van Noy, where they drafted a good player who just did not fit their system at all.

The issue is more - how much effort do you make to trade down in advance of such a situation? How much adjustment do you make to your own evaluation of a player if you think the market values him very differently? My observation is that generally speaking, on average, making this claim as bland as I possibly can to avoid riling people unnecessarily....when someone thinks they know something the market doesn't, it's more likely that the someone is wrong than that the market is wrong.
It sounds like you're assuming that the "consensus big boards" reflect the market. They may not, and certainly do not in specific cases. They are probably the best information we have on the outside about what the market looks like, but teams have better information than we do.

How about this situation. You're picking #30. There's a player you have #20 on your board available that you think consensus has #50 and a player you have #25 that the consensus also has #25. My impression is that most people commenting here think taking the #20 guy is the right thing to do, and that some people think doing anything other than taking that guy is ridiculous. I'm open to the idea that taking the #25 guy is actually the better plan, let alone a plan worth considering.
Maybe the right thing to do here is trade back a few spots and take your guy ... which is basically what the Patriots did in this draft, trading back from 21 to 29 and then drafting Strange.
 

SMU_Sox

queer eye for the next pats guy
SoSH Member
Jul 20, 2009
8,878
Dallas
I don’t really care that much about going LG at 29 or Strange. Is it ideal value? No. But if you have a pocket passer with limited mobility I would argue there is more value in protecting him well vs a more mobile QB. They also needed a guy who could pull well since, you know, they run GAP schemes.

I’m not sold on Marcus Jones as a slot corner who can cover anyone but I can buy his status as an elite returner and a matchup guy against smaller twitchier athletes. It might work out and maybe he can neutralize the incredibly fast WRs in the AFC East.

Jack Jones and the RBs and OL later on are all fine picks. Zappe isn't so bad either. It's a backup QB in the 4th round in a draft where kickers and punters started going off the board.

The one that I’m not sure about is Tyquan Thornton. He has 2 potentially elite traits with his speed and being able to beat press off the line (still developing but could be elite). But like Mark said on his pod he is more of a vertical receiver. Yes, in some respects their X receivers run a lot of vertical stems so his horizontal breaks aren't as important but I worry about both him as a prospect and if he can translate to the NFL and his fit here. I would have preferred someone who is is a more of a big catch radius type who can get to the right spot at the right time with a more complete route tree. I am torn on this pick because I like going with someone who beats press and is fast but on the other hand he's a difficult eval and has some trapdoors. Is he a one trick pony? Is Mac even going to develop into a better deep ball thrower? Lot of risk here to me. Maybe it works out. If it doesn't the narrative about BB not being able to draft receivers is going to continue to be insufferable. Although... this time, unlike with Harry, he is going against the grain taking Thornton in front of Pickens, Skyy Moore, and Pierce. It is likely 2/4 of them bust.
 

DavidTai

Member
SoSH Member
Dec 18, 2003
1,242
Herndon, VA
It sounds like you're assuming that the "consensus big boards" reflect the market. They may not, and certainly do not in specific cases. They are probably the best information we have on the outside about what the market looks like, but teams have better information than we do.
To add to this, I would suggest this is a lot more like a blind auction. It's not 30 Gms you have to be concerned about to just buy a player.

From the bidder's point of view, you don't care about the people who value a player as a third rounder. You care a whole LOT more about the people who value the player the same or higher as you do.

So I don't give a flying fig about the 'consensus big board' because I'm a whole lot more concerned about, say, Tampa Bay or Houston wanting to take an OG in the second round than I am about the rest of the market who wants to take the OG in the third.
 

Cellar-Door

Member
SoSH Member
Aug 1, 2006
34,459
So I was curious given the talk about the 2019 draft class....

Mack Wilson who we traded for was considered a STEAL in that class, drafted 155th, he was 59th on the consensus big board.
 
Apr 24, 2019
1,278
Okay, so I had the slowest of days at work yesterday and embarked on a little project. With all the talk about the "consensus" and how often the Patriots zig when other teams zag, I set out on a quest to figure one thing out: using the consensus big board at nflmockdraftdatabase, how much AV would the Patriots have gained/lost in the past five drafts had they picked the highest ranked available player by that measure instead of the guy they actually selected? It took me a while, but the results are now in! These will be spoilered because otherwise the post would be unbearably long.

2017 - plus 7 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Carl Lawson (EDGE) 11 career AV -> minus 11 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Caleb Brantley (DT) - 2 career AV -> plus 20 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by consensus Isaiah Ford (WR) - 5 career AV -> minus 2 AV

2018 - minus 2 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Harold Landry (EDGE) - 30 career AV -> minus 14 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Josh Jackson (CB) 7 career AV -> plus 14 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 1 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Tyrell Crosby (OT) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam - 0 career AV. Best available by consensus Equanimeous St. Brown (WR) - 4 career AV -> minus 4 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Quenton Meeks (CB) - 1 career AV -> plus 8 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> minus 5 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Allen Lazard (WR) - 14 career AV -> minus 9 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Desmond Harrison (OT) - 3 career AV -> minus 2 AV

2019 - Minus 35 AV

Pick 32 - Patriots select N’Keal Harry (WR) - 6 career AV. Best available by consensus Jawaan Taylor (OT) - 20 career AV -> minus 14 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus DK Metcalf (WR) - 30 career AV -> minus 27 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chauncey Gardner-Johnson (CB) - 12 career AV -> minus 2 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Hakeem Butler (WR) – 0 career AV -> plus 13 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Amani Oruwariye (CB) - 10 career AV -> minus 8 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus David Long (CB) - 9 career AV -> minus 9 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Deionte Thompson (S) - 4 career AV -> minus 3 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Kelvin Harmon (WR) – 3 career AV -> plus 6 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Emanuel Hall (WR) - 0 career AV -> plus 6 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Gerald Willis (DL) - 0 career AV -> plus 3 AV

2020 - plus 4 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Kristian Fulton (CB) - 5 career AV -> plus 4 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Josh Jones (OT) - 9 career AV -> minus 6 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Bryce Hall (CB) - 7 career AV -> minus 4 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 0 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Jacob Eason (QB) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Curtis Weaver (EDGE) - 0 career AV -> plus 0 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Donovan Peoples-Jones (WR) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Hunter Bryant (TE) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Eno Benjamin (RB) - 1 career AV -> plus 0 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Thaddeus Moss (TE) - 0 career AV -> plus 0 AV

2021 - minus 9 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Jeremiah Owusu-Karamoah (LB) - 4 career AV -> minus 1 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Jabril Cox (LB) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Daviyon Nixon (DL) – 0 career AV -> plus 5 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Trey Smith (G) - 10 career AV -> minus 10 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Quincy Roche (EDGE) - 2 career AV -> minus 2 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Marvin Wilson (DL) - 0 career AV -> plus 0 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Dylan Moses (LB) – 0 career AV -> plus 0 AV

So that's 35 AV lost over a period of five drafts just by going "BPA by the big board" instead of the guy they ended up taking. Now, I'm fully aware that this isn't a complete picture and that measuring AV of 2020 and 2021 draft picks especially is a little foolish considering many of those guys could still develop/get injured and the equation could change rapidly in the coming years. So they basically ended up pretty much near what you'd expect in every season, aside from 2019, which was a bloodbath (kinda tracks with my overall perception too). In essence: four meh drafts, one terrible draft.

After I finished this up, though, I began thinking this was an incomplete picture. Just going BPA makes little sense. The 2019 BPA draft had them taking 3 CBs and 4 WRs, for instance, which is insane. So I decided to refine the comparison a little and find out the AV gained/lost by the Patriots had they simply gone BPA at the position they ended up picking over the past five drafts. Here are the results.

2017 = plus 22 AV

Pick 83 – Patriots select Derek Rivers (DE) - 4 career AV. Best available by consensus Derek Rivers (DE) 4 career AV -> plus 0 AV

Pick 85 – Patriots select Antonio Garcia (OT) - 0 career AV. Best available by consensus Antonio Garcia (OT) - 0 career AV -> plus 0 AV

Pick 131 – Patriots select Deatrich Wise Jr. (DE) - 22 career AV. Best available by consensus Joe Mathis (DE) - 0 career AV -> plus 22 AV

Pick 211 – Patriots select Conor McDermott - 3 career AV. Best available by Conor McDermott - 3 career AV -> plus 0 AV

2018 = plus 19 AV

Pick 23 – Patriots select Isaiah Wynn (OT) - 16 career AV. Best available by consensus Connor Williams (OT) - 23 career AV -> minus 7 AV

Pick 31 – Patriots select Sony Michel (RB) - 21 career AV. Best available by consensus Derrius Guice (RB) - 2 career AV -> plus 19 AV

Pick 56 – Patriots select Duke Dawson (CB) - 3 career AV. Best available by consensus Isaiah Oliver (CB) - 11 career AV -> minus 8 AV

Pick 143 – Patriots select Ja’Whaun Bentley (LB) - 20 career AV. Best available by consensus Genard Avery (LB) - 11 career AV -> plus 9 AV

Pick 178 – Patriots select Christian Sam (LB) - 0 career AV. Best available by consensus Jack Cichy (LB) - 0 career AV -> plus 0 AV

Pick 210 – Patriots select Braxton Berrios (WR) - 9 career AV. Best available by consensus Auden Tate (WR) - 5 career AV -> plus 4 AV

Pick 219 – Patriots select Danny Etling (QB) - 0 career AV. Best available by consensus Kurt Benkert (QB) - 0 career AV -> plus 0 AV

Pick 243 – Patriots select Keion Crossen (CB) - 5 career AV. Best available by consensus Holton Hill (CB) - 4 career AV -> plus 1 AV

Pick 250 – Patriots select Ryan Izzo (TE) - 1 career AV. Best available by consensus Marcus Baugh (TE) - 0 career AV -> plus 1 AV

2019 = minus 31 AV

Pick 32 – Patriots select N’Keal Harry (WR) - 6 career AV. Best available by DK Metcalf (WR) - 30 career AV -> minus 24 AV

Pick 45 – Patriots select Joejuan Williams – 3 career AV. Best available by consensus Greedy Williams (CB) - 7 career AV -> minus 4 AV

Pick 77 – Patriots select Chase Winovich (DE) - 10 career AV. Best available by consensus Chase Winovich (DE) - 10 career AV -> plus 0 AV

Pick 87 – Patriots select Damien Harris (RB) - 13 career AV. Best available by consensus Damien Harris (RB) - 13 career AV -> plus 0 AV

Pick 101 – Patriots select Yodney Cajuste (OT) - 2 career AV. Best available by consensus Yodney Cajuste (OT) - 2 career AV -> plus 0 AV

Pick 118 – Patriots select Hjalte Froholdt - 0 career AV. Best available by consensus Michael Jordan (G) - 11 career AV -> minus 11 AV

Pick 133 – Patriots select Jarrett Stidham (QB) - 1 career AV. Best available by consensus Jarrett Stidham (QB) - 1 career AV -> plus 0 AV

Pick 159 – Patriots select Byron Cowart (DT) - 9 career AV. Best available by consensus Charles Omenihu (DL) – 4 career AV -> plus 5 AV

Pick 163 – Patriots select Jake Bailey (P) - 6 career AV. Best available by consensus Mitch Wishnowsky (P) - 6 career AV -> plus 0 AV

Pick 252 – Patriots select Ken Webster (CB) - 3 career AV. Best available by consensus Saivion Smith (CB) - 0 career AV -> plus 3 AV

2020 = plus 3 AV

Pick 37 – Patriots select Kyle Dugger (S) - 9 career AV. Best available by consensus Grant Delpit (S) - 3 career AV -> plus 6 AV

Pick 60 – Patriots select Josh Uche (LB) – 3 career AV. Best available by consensus Terrell Lewis (LB) - 3 career AV -> plus 0 AV

Pick 87 – Patriots select Anfernee Jennings (LB) - 3 career AV. Best available by consensus Jonathan Greenard (LB) - 6 career AV -> minus 3 AV

Pick 91 – Patriots select Devin Asiasi (TE) - 0 career AV. Best available by consensus Adam Trautman (TE) – 4 career AV -> minus 4 AV

Pick 101 – Patriots select Dalton Keene (TE) - 0 career AV. Best available by consensus Brycen Hopkins (TE) - 0 career AV -> plus 0 AV

Pick 159 – Patriots select Justin Rohrwasser (K) - 0 career AV. Best available by consensus Rodrigo Blankenship (K) - 3 career AV -> minus 3 AV

Pick 182 – Patriots select Michael Onwenu (OL) - 12 career AV. Best available by consensus Josh Simpson (OL) - 8 career AV -> plus 4 AV

Pick 195 – Patriots select Justin Herron (OT) - 6 career AV. Best available by consensus Prince Tega Wanogho (OT) – 0 career AV -> plus 6 AV

Pick 204 – Patriots select Cassh Maluia (LB) - 1 career AV. Best available by consensus Markus Bailey (LB) - 4 career AV -> minus 3 AV

Pick 230 – Patriots select Dustin Woodward (C) - 0 career AV. Best available by consensus Darryl Williams (IOL) - 0 career AV -> plus 0 AV

2021 = minus 2 AV

Pick 15 – Patriots select Mac Jones (QB) - 14 career AV. Best available by consensus Mac Jones (QB) - 14 career AV -> plus 0 AV

Pick 38 – Patriots select Christian Barmore (DT) – 3 career AV. Best available by consensus Christian Barmore (DT) – 3 career AV -> plus 0 AV

Pick 96 – Patriots select Ronnie Perkins (DE) - 0 career AV. Best available by consensus Carlos Basham (DE) - 1 career AV -> minus 1 AV

Pick 120 – Patriots select Rhamondre Stevenson (RB) - 5 career AV. Best available by consensus Kenneth Gainwell (RB) – 5 career AV -> plus 0 AV

Pick 177 – Patriots select Cameron McGrone (LB) - 0 career AV. Best available by consensus Dylan Moses (LB) - 0 career AV -> plus 0 AV

Pick 188 – Patriots select Joshua Bledsoe (S) - 0 career AV. Best available by consensus Ar’Darius Washington (S) - 0 career AV -> plus 0 AV

Pick 197 – Patriots select William Sherman (OL) - 0 career AV. Best available by consensus Stone Forsythe (OL) - 1 career AV -> minus 1 AV

Pick 242 – Patriots select Tre Nixon (WR) - 0 career AV. Best available by consensus Cade Johnson (WR) – 0 career AV -> plus 0 AV

The Pats have gone from negative 35 AV to plus 12 AV over the past five years with the positional adjustment. The 2019 draft is still god awful, but now 2017 and 2018 look much better! But in what is the best representation of how weird this exercise is, the surplus value that accounted for this reversal came mostly from the selections of Sony Michel and Deatrich Wise Jr., two players I don't think most Pats fans would rank as particularly great draft picks.

So there you have it. What can we take from this? Who knows. Drafting is hard, mostly. And even role players that aren't stars or entrenched starters are generally valuable even when picked reasonably high.

For funsies, I thought it'd be fair to go back and look at how the 2016 Patriots draft (which I consider to be a very good draft) would have fared under the "what if they had simply gone BPA" parameters and the results are simply ridiculous:

2016 = plus 114 AV

Pick 60 – Patriots select Cyrus Jones (CB) - 4 career AV. Best available by consensus Andrew Billings (DT) 14 career AV -> minus 10 AV

Pick 78 – Patriots select Joe Thuney (G) - 48 career AV. Best available by consensus Jalen Mills (S) 26 career AV -> plus 22 AV

Pick 91 – Patriots select Jacoby Brissett (QB) - 24 career AV. Best available by consensus Kevin Hogan (QB) - 2 career AV -> plus 22 AV

Pick 96 – Patriots select Vincent Valentine (DT) - 2 career AV. Best available by consensus Joshua Perry (LB) - 1 career AV -> plus 1 AV

Pick 112 – Patriots select Malcolm Mitchell - 4 career AV. Best available by consensus Jeremy Cash (S) - 0 career AV -> plus 4 AV

Pick 208 – Patriots select Kamu Grugier-Hill (LB) - 18 career AV. Best available by consensus Charome Peake (WR) - 1 career AV -> plus 17 AV

Pick 214 – Patriots select Elandon Roberts (LB) - 34 career AV. Best available by consensus Scooby Wright III (LB) - 1 career AV -> plus 33 AV

Pick 221 – Patriots select Ted Karras (G) - 25 career AV. Best available by consensus Kenny Lawler (WR) - 0 career AV -> plus 25 AV

Pick 225 – Patriots select Devin Lucien (WR) - 0 career AV. Best available by consensus Bryce Williams (TE) - 0 career AV -> plus 0 AV

That's a whopping surplus of 114 AV in a single draft. They basically got Tony Romo's entire career (115 AV) just on added value when compared to the players they would have drafted by going big board BPA.
This was a ton of work and very interesting for lesser draftniks like me. I really appreciate the time and energy that went into it. Thanks!
 

Niastri

Member
SoSH Member
No one really believes the pure efficient-market hypothesis is true. Obviously information isn't fully and equally known. Yet passive management still works.

It doesn't require perfect knowledge, or anything close to it, for it to be right to go with consensus over your own opinion that consensus is wrong.
Another point is that in the stock market consensus CREATES the correct answer.

If Tesla is worthless, but everybody is it out of their mind to buy Tesla, the first people into the Ponzi scheme are proven correct, as the price of their valueless stock increases and they get rich. Bitcoin is another example of this. It doesn't matter what the security is worth in the immediate term, because people are convinced by the consensus that they should buy. (Of course in the long run the market is a weighing machine, not a voting machine, but tell that to Wall Street Bets)

On the other hand, if consensus is that a guy is a great football player, and he isn't, drafting him early doesn't make him any better. The goal is to win football games, not draft night.

Following consensus in drafting and the stock market are very different situations.
 

SMU_Sox

queer eye for the next pats guy
SoSH Member
Jul 20, 2009
8,878
Dallas
@rodderick that was awesome. Thank you for doing that!

The Consensus Board is good at predicting, usually, where guys go in the draft. And where guys go in the draft is correlated to success. Some/most of that is because the guys drafted higher are the ones who get the opportunity. Over time the board is good at predicting value vs an individuals board. In years where there is no consensus on players, like this year, the board is not as helpful.

Marcus Jones is a good example of the lack of consensus. Some boards had him as early as the low 50s while others had him in the 190s.

I feel pretty confident that Strange works out FWIW and the OL, corners, and RBs can contribute. Thornton… we shall see.
 

rodderick

Member
SoSH Member
Apr 24, 2009
12,751
Belo Horizonte - Brazil
@rodderick that was awesome. Thank you for doing that!

The Consensus Board is good at predicting, usually, where guys go in the draft. And where guys go in the draft is correlated to success. Some/most of that is because the guys drafted higher are the ones who get the opportunity. Over time the board is good at predicting value vs an individuals board. In years where there is no consensus on players, like this year, the board is not as helpful.

Marcus Jones is a good example of the lack of consensus. Some boards had him as early as the low 50s while others had him in the 190s.

I feel pretty confident that Strange works out FWIW and the OL, corners, and RBs can contribute. Thornton… we shall see.
And they generally do a good job of predicting the overall range a player will be drafted! 82% of this draft's first 150 picks were on the consensus big board top 150, which to me is a great result.
 

BigJimEd

Member
SoSH Member
Jan 4, 2002
4,432
And they generally do a good job of predicting the overall range a player will be drafted! 82% of this draft's first 150 picks were on the consensus big board top 150, which to me is a great result.
I asked this earlier but is it? Or more to the point is there a better metric? Willis, Dean, Strange, Corral and others were drafted in the top 150 but most would agree that media projections for them were not very accurate.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,054
Hingham, MA
I asked this earlier but is it? Or more to the point is there a better metric? Willis, Dean, Strange, Corral and others were drafted in the top 150 but most would agree that media projections for them were not very accurate.
There has to be a better metric. I'd love to see what % of picks were picked with +/- 5 slots of general consensus, 10 slots, 20 slots, etc. It'd be fascinating to look at.
 

Phil Plantier

Member
SoSH Member
Mar 7, 2002
3,419
On the other hand, if consensus is that a guy is a great football player, and he isn't, drafting him early doesn't make him any better. The goal is to win football games, not draft night.
I think this is a little more complicated: the earlier someone is drafted, the more opportunities they would normally get. This would definitely lead to increased (overall) production, and might be correlated with improvements due to experience.
 

Saints Rest

Well-Known Member
Lifetime Member
SoSH Member
Doesn't analyzing the draft results get complicated by that some players are picked that are bad fits for one team and better for another, so their value goes up -late- but they have lost a lot of value being with the wrong team? (Thinking of Kyle Van Noy, here.)
This is exactly right and precisely why the stock market analogy carries no weight as a comp. In the stock market, money is money; you don't need to know if your Apple stock will be enhanced by being in your portfolio next to your Coke stock. You can still get maximum value out of having both Coke stop and Pepsi stock. In the NFL, fit matters, not only in terms of your team on the field (IOW, it would do a team no good to have Mahomes AND Rodgers), but also in terms of your team in the locker room (could a team survive having 3 diva WRs at the same time??). Furthermore, the players have to fit YOUR schemes -- horses for courses and all that.

As such, not all teams share the same rating systems or methods. A reach is only real if a team chose to pick a player ahead of their own internal ranking system. Crowd-sourcing (and "consensus big boards" are exactly that as noted upthread) only has value where the needs/wants/desires of the entity doing the picking align with those of the crowd. It's why Oscars don't simply match up to Rotten Tomatoes scores.