Interesting analysis of Pats drafting

[patchick]

Baltimore assistant GM Eric DeCosta agrees with you:



Baltimore Ravens | News | John Eisenberg: Ravens Cracked The Code

The article is a good read on the Ravens' drafting philsoophy, and is worth checking out.

It is a good article, and presumably much applies to the Pats too, since they run a similar scouting-service-free operation. It was particularly interesting how much they focused on the player's attitude and approach to the game, as opposed to skillset. (Ahoy, everybody who's still grousing over the Pats' passing on Trumaine Johnson!)

 

[borg]

Need clearly plays a part, but...doesn't the lack of impact players on the 2011 defense show that the Patriots had a clear need for defensive playmakers in the 2011 draft? As I recall, everybody here thought so at the time.

The NEED on D you refer to regarding the 2011 team and the 2011 draft was addressed by BB in the form of Free Agents....Carter, Ellis, Haynesworth, Anderson etc. Without a full training camp, BB chose the veteran route to fill holes. Unfortunately, he whiffed on two ....Haynesworth and Ellis....and Carter went down with injury.To compound this, his secondary plans collapsed during preseason when he blew up the safety position and Bodden. I file this as an unforeseen set back.

Examining the 2011 draft picks, with the exception of Dowling... Solder/Vareen/Ridley were all drafted to slide in after the final year of Light, Faulk, and BJGE's contracts. To me, BB approached the 2011 draft as a man comfortable with his offseason team build and used the draft to add a layer of depth , mostly on offense. What transpired after that....well...it wasn't pretty early....but got them to the big game.

 

[mayoclinic]

It is a good article, and presumably much applies to the Pats too, since they run a similar scouting-service-free operation. It was particularly interesting how much they focused on the player's attitude and approach to the game, as opposed to skillset. (Ahoy, everybody who's still grousing over the Pats' passing on Trumaine Johnson!)

For a discussion of what happens when you take the alternate approach, I've just posted a new thread on the draft forum, based on an Albert Breer article looking at the importance of character issues in light of the recent off-field problems of Detroit's top 3 picks from the 2011 draft:

http://www.patsfans.com/new-england-patriots/messageboard/13/919533-lessons-lions-2011-draft.html

The Breer article raises a lot of interesting points, and is a good read:

Detroit Lions' troubled 2011 draft class raises larger questions - NFL.com

 

[Deus Irae]

I.....1) More premium picks are better than fewer premium picks.....regardless of changing financial dynamics/contract lengths...

This is only potentially true, because hit rate drops along with draft position. For example, the Patriots have never had a first round bust, but they've had plenty of them in the second round. So, in order for the "more are better" argument to be accurate for the Patriots, the picks gotten from a drop to the second round have to produce more than 1.0 "hits" as an average when the trade downs occur.

As a league average, Casserly came up with (as I recall) 75% in round one and 50% in round two, so the league numbers would be different than the Patriots-specific numbers, but the general idea is the same.

Casserly cited on a 10-year study – which evaluated drafts after four years had passed – and listed the following numbers of success:

First round – 75 percent
Second round – 50 percent
Third round – 30 percent
Fourth round – 25 percent
Fifth round – 20 percent
Sixth round – 9 percent
Seventh round – 9 percent

Casserly, Billick talk Patriots - Reiss' Pieces - Boston.com

 

[patchick]

This is only potentially true, because hit rate drops along with draft position. For example, the Patriots have never had a first round bust, but they've had plenty of them in the second round. So, in order for the "more are better" argument to be accurate for the Patriots, the picks gotten from a drop to the second round have to produce more than 1.0 "hits" as an average when the trade downs occur.

As a league average, Casserly came up with (as I recall) 75% in round one and 50% in round two, so the league numbers would be different than the Patriots-specific numbers, but the general idea is the same.



Casserly, Billick talk Patriots - Reiss' Pieces - Boston.com

Given that the average 1st-round pick is worth, in trade, at least twice the average 2nd-round pick, and 2nd-rounders are worth twice as much as thirds, etc., playing those percentages could lead to some pretty major trading down! But of course there are more factors that could contribute to the 2nd being a "sweet spot":

- The numbers game. Realistically, you're just not going to be able to carry 8 4th-round rookies, giving them all enough snaps to let them develop and realize their potential, vs. 2-3 2nd rounders.

- Factoring in the chances of a "special talent," rather than just a "successful pick." My sense from an informal scan of the draft histories of the top teams is that the "special" rate drops off significantly in round 3.

 

[Deus Irae]

Given that the average 1st-round pick is worth, in trade, at least twice the average 2nd-round pick, and 2nd-rounders are worth twice as much as thirds, etc., playing those percentages could lead to some pretty major trading down! But of course there are more factors that could contribute to the 2nd being a "sweet spot":

- The numbers game. Realistically, you're just not going to be able to carry 8 4th-round rookies, giving them all enough snaps to let them develop and realize their potential, vs. 2-3 2nd rounders.

- Factoring in the chances of a "special talent," rather than just a "successful pick." My sense from an informal scan of the draft histories of the top teams is that the "special" rate drops off significantly in round 3.

I was trying to keep out the "special talent" argument, because it's not accounted for very well by Casserly. However, if you look at the BB Patriots draft history

2012 21 21 Chandler Jones DE Syracuse
2012 25 25 Dont'a Hightower LB Alabama
2011 17 17 Nate Solder T Colorado
2010 27 27 Devin McCourty DB Rutgers
2008 10 10 Jerod Mayo LB Tennessee
2007 24 24 Brandon Meriweather DB Miami (FL)
2006 21 21 Laurence Maroney RB Minnesota
2005 32 32 Logan Mankins G Fresno State
2004 21 21 Vince Wilfork DT Miami (FL)
2004 32 32 Ben Watson TE Georgia
2003 13 13 Ty Warren DT Texas A&M
2002 21 21 Dan Graham TE Colorado
2001 6 6 Richard Seymour DT Georgia

2012 16 48 Tavon Wilson DB Illinois
2011 1 33 Ras-I Dowling DB Virginia
2011 24 56 Shane Vereen RB California
2010 10 42 Rob Gronkowski TE Arizona
2010 21 53 Jermaine Cunningham LB Florida
2010 30 62 Brandon Spikes LB Florida
2009 2 34 Patrick Chung DB Oregon
2009 8 40 Ron Brace DT Boston College
2009 9 41 Darius Butler DB Connecticut
2009 26 58 Sebastian VollmerT Houston
2008 31 62 Terrence Wheatley DB Colorado
2006 4 36 Chad Jackson WR Florida
2004 31 63 Marquise Hill DE Louisiana State
2003 4 36 Eugene Wilson DB Illinois
2003 13 45 Bethel Johnson WR Texas A&M
2002 33 65 Deion Branch WR Louisville
2001 17 48 Matt Light T Purdue
2000 15 46 Adrian Klemm T Hawaii

I think we see that the quality of the "hits" is weighted towards the "hits" in round 1.

 

[ay-yo]

My 2 cents, the end of the first is not really a sweet spot:

21 - Wilfork
21 - Maroney
21 - Graham
24 - Meriweather
32 - Mankins
32 - Watson

Only 2 of 6 players got a 2nd contract with New England. The success of McCourty (27), Jones (21) and Hightower (25) could shift this range dramatically either way.

Still, you have to wonder just how sweet picks 21-32 really are. It changes every year depending on the players available but its a trend I have noticed.

Compare this to players selected in the top 20, where all picks earned a 2nd contract with us:
6 - Seymour
10 - Mayo
13 - Warren
(17 - Solder - TBD)

 

[Palm Beach Pats Fan]

Hit rate is nowhere near as important as the number of good players obtained.

Take two hypothetical teams, A and B, over 5 years

team A:
3 of 5 first rounders are a success (60%)
3 of 5 2nd rounders are a success (60%)
2 of 5 3rd rounders are a success (40%)

team B:
2 of 3 first rounders are a success (67%)
5 of 11 2nd rounders are a success (45%)
3 of 8 3rd rounders are a success (38%)

Team B has moved 1st round picks for multiple second & 3rd rounders, accepted a lower "hit rate" in the 2nd and 3rd rounds but amassed two more quality players than team A in the same time span. What if many of the trades gave extra picks in later rounds as well?

 

[Ring 6]

I just question whether or not if .500 is successful. Seems average.

If Tom Brady completes only half his passes, hes Mark Sanchez or Matt Cassel. A WR on the Pats isnt going to be around very long dropping 50% of his passes when Welker is catching 70%. FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | 2011 WIDE RECEIVERS

This has to officially be the most ingorant post of the year so far.

Corrollary to your theory. I am a good driver because only 20% of the time I get behind the wheel, I end up in a fatal accident, and that is much better than the 'average 50%'.

 

[Ring 6]

This is only potentially true, because hit rate drops along with draft position. For example, the Patriots have never had a first round bust, but they've had plenty of them in the second round. So, in order for the "more are better" argument to be accurate for the Patriots, the picks gotten from a drop to the second round have to produce more than 1.0 "hits" as an average when the trade downs occur.

As a league average, Casserly came up with (as I recall) 75% in round one and 50% in round two, so the league numbers would be different than the Patriots-specific numbers, but the general idea is the same.



Casserly, Billick talk Patriots - Reiss' Pieces - Boston.com

Except they almost never trade a first for seconds, they trade a first for another first the next year PLUS a 2nd. In the grand scheme of things the 2nd is simply a bonus pick, because a first taken in any season is the same as in any other season if you are looking at the long range prospects of the franchise.

 

[patchick]

I was trying to keep out the "special talent" argument, because it's not accounted for very well by Casserly. However, if you look at the BB Patriots draft history



I think we see that the quality of the "hits" is weighted towards the "hits" in round 1.

Deus, I'm not sure what you're getting at here...I see the "special talent" hit rate as about 50-60% in the 1st and 25-30% in the 2nd. What does that demonstrate?

 

[condon84]

That interactive graph is very well done.

 

[patchick]

Except they almost never trade a first for seconds, they trade a first for another first the next year PLUS a 2nd. In the grand scheme of things the 2nd is simply a bonus pick, because a first taken in any season is the same as in any other season if you are looking at the long range prospects of the franchise.

Yep, this seems to be a common misperception -- we remember all of the "trades down," but most were actually trades forward. The most common way they acquired extra 2nds was by trading a 3rd into the future and moving UP a round.

FWIW, the 1 occasion I've found when they did drop out of the 1st without getting a 1st back, the return was roughly two 2nds, a 3rd and a 7th.

 

[mgteich]

I strongly agree with almost ALL of tour analysis.

I agree that the patriots (and probably almost everyone else) has done better drafting 6-20 than at 21-32.
---------------------

1) signing a second contract with New England is NOT a measure of success. The question whether the player contributed and was valuable during his contract period. Many things affect who a player will sign with in his second contract.

Let's look at your six and see the "misses" (there weren't any). I think that most any team would consider this group of 9 picks to be a solid group of draftees; much better production that most teams have gotten with similar picks, with fewer disappointments.

WILFORK - an all-pro, and the best pick since Brady

MARONEY - started as a contender for offensive rookie of the year. If you choose to judge players on production after injury, that is your choice.

GRAHAM
I wanted a linebacker when we traded up for Graham. However, Graham played very. very well for us. We certainly got solid value for the pick.

MERIWEATHER
All but folks on pats message boards thought Meriweather was one of the very best values picked in his draft class. He started for three years, and played reasonable well. He certainly was not a bust by any means.

MANKINS
Another pro-bowler

WATSON
We didn't want a TE, so message board fans disliked Watkins. He played very well early. Watson had one of the very best plays in the history of the patriots. Watson signed for someone else. Like Graham, we received reasonable value in Watson (not as much as Graham).

My 2 cents, the end of the first is not really a sweet spot:

21 - Wilfork
21 - Maroney
21 - Graham
24 - Meriweather
32 - Mankins
32 - Watson

Only 2 of 6 players got a 2nd contract with New England. The success of McCourty (27), Jones (21) and Hightower (25) could shift this range dramatically either way.

Still, you have to wonder just how sweet picks 21-32 really are. It changes every year depending on the players available but its a trend I have noticed.

Compare this to players selected in the top 20, where all picks earned a 2nd contract with us:
6 - Seymour
10 - Mayo
13 - Warren
(17 - Solder - TBD)

 

[patchick]

My 2 cents, the end of the first is not really a sweet spot:

21 - Wilfork
21 - Maroney
21 - Graham
24 - Meriweather
32 - Mankins
32 - Watson

Only 2 of 6 players got a 2nd contract with New England. The success of McCourty (27), Jones (21) and Hightower (25) could shift this range dramatically either way.

Still, you have to wonder just how sweet picks 21-32 really are. It changes every year depending on the players available but its a trend I have noticed.

Compare this to players selected in the top 20, where all picks earned a 2nd contract with us:
6 - Seymour
10 - Mayo
13 - Warren
(17 - Solder - TBD)

I'm pretty sure that the author is talking about a leaguewide cost/benefit "sweet spot." Nobody's arguing that a #30 pick is more likely to succeed than a #10 pick!

 

[Triumph]

This has to officially be the most ingorant post of the year so far.

To be fair, you just started to post in this thread.

Corrollary to your theory. I am a good driver because only 20% of the time I get behind the wheel, I end up in a fatal accident, and that is much better than the 'average 50%'.

This board should be so lucky.

Yo, the discussion is about drafting. Nobody would argue that a .500 baseball hitter isnt the GOAT.

 

[patchick]

Yo, the discussion is about drafting.

...which is why your comparison to pass completion percentage seemed pretty random. :confused2:

 

[Triumph]

My apologies, lousy phrasing on my part! I said "regular starter" the first time, which is what I meant and what I should have said throughout. I considered Spikes a regular starter for those years, just as I consider Aaron Hernandez a regular starter for last year even though he didn't start every game (depending on matchups and formations). And most importantly, I applied the same standard to the other teams.

Not a problem.

IMO, its too early to tell if Spikes or Vollmer are "Hits." Spikes played very well at the end of the season, but as we have seen with McCourty, nothing is guaranteed in the next season. It looks promising for Spikes. Vollmer still has much to prove IMO.

 

[Ring 6]

To be fair, you just started to post in this thread.



This board should be so lucky.

Yo, the discussion is about drafting. Nobody would argue that a .500 baseball hitter isnt the GOAT.

Thats exactly the point.
Comparing the % draft success to completion percentage for a QB is no more valid than batting average in baseball or car accident percentage.
In your reply, you proved my point in addition to your humorous jabs.

 

[Deus Irae]

Deus, I'm not sure what you're getting at here...I see the "special talent" hit rate as about 50-60% in the 1st and 25-30% in the 2nd. What does that demonstrate?

It demonstrates that just using the "starter" basis for rating the picks isn't sufficient to determining the best strategy. For the Patriots, the first round is not just better for hits, it's significantly better for the subset of hit quality.

In other words, you have another dynamic layer when you are looking at the option of a trade down. Do you take the lesser hit rate, plus the lesser "quality hit" rate, or do you keep the pick? Looking at this from the Patriots drafting history, and using your numbers (I'll use the lower numbers of 50/25 because they're easier math ), here's the 1:1 rate comparison:

1st round hit rate = 100%
1st round quality hit rate = 50%

Results: Every 2 picks, you get 2 hits and one quality hit

2nd round hit rate = 50%
2nd round quality hit rate = 25%

Results: Every 2 picks, you get 1 hit and .25 quality hit

That's a 4:1 quality hit rate ratio.