Thursday Night 3rd and Long: Analyzing the Draft

Sorry for the delay- this article is coming out a bit late because the statistics I assembled took me quite a bit of time to compile.  At 1 a.m. this morning I had an unfinished article, a dayjob looming, and a decision to make.  So a morning article became a night article (just for this week).  Please stay with me even though this is long-winded, because I think the statistics I put together are really interesting.

Usually before I sit down on Wednesday night to write this column, I watch the prior weekend's Patriots game to remember anything fresh I may have missed or to analyze certain players performances more in-depth.  This week, I decided to skip that.

To be honest, I'm not sure I could have subjected myself to that torture again.

I think that everyone here at Pats Pulpit has done a great job of breaking down New England's struggles against the Steelers, so I have decided to do a statistical breakdown of the last eight years worth of Patriots draft to see if I can spot any trends in how New England values different positions in the draft and then also what the value added has been per position and per round over the last eight years.

Before people go on a tirade about how this system is flawed, I'll go over a few things:

A.    For the "perceived value," I created an arbitrary valuation system that I believe accurately shows how teams value certain rounds in the draft.  I went with decreasing values for rounds 1-7 that add up to a total value of one.  These values are respectively 0.4, 0.25, 0.15, 0.1, 0.06, 0.03, and 0.01, with 0.4 representing the perceived value of a first round pick and 0.01 being the perceived value of a seventh round pick. 

B.     In the second set of analysis, I decided to base what I felt a player's actual value was to the team in teams of two things, and two things only: games played of potential games (meaning games in which the player was actively employed by the team), and games started.  Production is an entirely different thing, but it's hard to have any concrete comparison between different positions based on production.  If we instead look at value as games played and especially games started, we can compare by round and by position how the Patriots draft according to need.

C.    In the upcoming weeks I will have charted similar data for at least one other team to compare the Patriots.  This particular set took me long enough that there was no way to get it accomplished in time to publish this.

First, lets examine draft picks from 2004-2011 (I figured 8 years would be an accurate snapshot).        

	1	2	3	4	5	6	7
2004	DL- Vince Wilfork	DL- Marquise Hill	S- Guss Scott	S- Dexter Reid	WR- P.K. Sam		CB- Christian Morton
	TE- Ben Watson			RB- Cedric Cobbs
2005	OL- Logan Mankins	-	CB- Ellis Hobbs	S- James Sanders	LB- Ryan Claridge		QB- Matt Cassel
			OL- Nick Kazcur				TE- Andy Stokes
2006	RB- Laurence Maroney	WR- Chad Jackson	TE- Dave Thomas	TE- Garrett Mills	OL- Ryan O'Callaghan	DL- Jeremy Mincey	CB- Willie Andrews
				K- Stephen Gostowski		OL- Dan Stevenson
						DL- Le Kevin Smith
2007	S- Brandon Meriweather			DL- Kareem Brown	OL- Clint Oldenburg	LB- Justin Rogers	LB- Oscar Lua
						CB- Mike Richardson	OL- Mike Elgin
						RB- Justise Hairston
						OL- Corey Hilliard
2008	LB- Jerod Mayo	CB- Terrence Wheatley	LB- Shawn Crable	CB- Jonathan Wilhite	WR- Matthew Slater	LB- Bo Ruud
			QB- Kevin O'Connell
2009	-	S- Pat Chung	WR- Brandon Tate	OL- Rich Ohrnberger	OL- George Bussey	LS- Jake Ingram	WR- Julian Edelman
		DL- Ron Brace	LB- Tyrone McKenzie			DL- Myron Pryor	DL- Darryl Richard
		CB- Darius Butler
		OL- Sebastian Vollmer
2010	CB- Devin McCourty	TE- Rob Gronkowski	WR- Taylor Price	TE- Aaron Hernandez	P- Zoltan Mesko	C- Ted Larson	OT- Thomas Welch
		OLB/DE- Jermaine Cunningham					DL- Brandon Deaderick
		LB- Brandon Spikes					DL- Kade Weston
							QB- Zac Robinson
2011	OT- Nate Solder	DB- Ras-I Dowling	RB- Stevan Ridley	-	OL- Marcus Cannon	LB/DE Markell Carter	DB- Malcolm Williams
		RB- Shane Vereen	QB- Ryan Mallett		TE- Lee Smith

Here is the breakdown of position by round and how many total draft picks were used in the time period at each position:

By Position	LB	DL	WR	RB	QB	S	OL	K, P, LS	DB	TE
1	1	1	0	1	0	1	2	0	1	1
2	1	3	1	1	0	1	1	0	3	1
3	2	0	2	1	1	1	1	0	1	1
4	0	1	0	1	0	2	1	1	1	2
5	1	0	2	0	0	0	4	1	0	1
6	2	4	0	1	0	0	3	1	1	0
7	1	3	1	0	2	0	2	0	2	1
Total	8	12	6	5	3	5	14	3	9	7

I take a few things from this initial count.  First- the parity of the first round selection.  The Patriots have used a first round pick on every position with the exception of WR, QB, and the combination of K,P, and LS in the last eight years, with a maximum of 2 picks used on the offensive line. I feel that this demonstrates a remarkable equality between positions in drafting strategy. 

Now, lets take the notion of "perceived value" and multiply it with the position drafted by round:

Perceived Value	LB	DL	WR	RB	QB	S	OL	K, P, LS	DB	TE
1	0.4	0.4	0	0.4	0	0.4	0.8	0	0.4	0.4
2	0.25	0.75	0.25	0.25	0	0.25	0.25	0	0.75	0.25
3	0.3	0	0.3	0.15	0.15	0.15	0.15	0	0.15	0.15
4	0	0.1	0	0.1	0	0.2	0.1	0.1	0.1	0.2
5	0.06	0	0.12	0	0	0	0.24	0.06	0	0.06
6	0.06	0.12	0	0.03	0	0	0.09	0.03	0.03	0
7	0.01	0.03	0.01	0	0.02	0	0.02	0	0.02	0.01
Total	1.08	1.4	0.68	0.93	0.17	1	1.65	0.19	1.45	1.07
Per Pick	0.135	0.117	0.113	0.186	0.057	0.200	0.118	0.063	0.161	0.153

This distribution tells a story of how much the Patriots are willing to "spend" in terms of perceived value per position.  The offensive line is obviously the area that stands out, but mostly because two first round picks were used on those players.  The next highest valued draft positions are the defensive line and defensive backs, which is interesting considering how poorly these two positions have performed as of recent. Not surprisingly, quarterback, wide receiver, and the combination of K, P, and LS (long snapper) are the least valued draft positions, at least according to this chart.  However, we must take into account that the former position has been occupied by Tom Brady for the last decade, so there was little need to utilize better draft picks for that position.  Also, before 2007 Randy Moss and Wes Welker were traded for, so that is most likely why the drafted value for wide receiver is so low.

I then took the average per position to see what the standard value associated with a specific position would be. Safety and running back stand out as the highest average, indicating that the Patriots were more willing to use higher draft picks on those positions.  Values of .2 and .186 would indicate that New England spends, on average, a pick between the 2^nd and 3^rd rounds on safeties and running backs.  Values of .161 and .153 would indicate that the Patriots average a third round pick on cornerbacks and tight ends.  The values for defensive lineman, wide receivers, and offensive lineman are all around .11, which would indicate that New England usually drafts these positions around round four, but occasionally higher. The value for quarterback is lowest, but the Patriots have also only selected 3 quarterbacks in the measured span of time, but average a fifth round pick on the position.

Moving on from perceived value, we next visit the added value of a player according to a simple formula.  First, I gathered information on how long each pick was with the team.  For this calculation, I'm ignoring any time spent on injured reserve or injured during the season.  Although injury may not be the fault of the player, it definitely subtracts from the value a player has to the team, at least in terms of starts.  So for this added value statistic, I'm using PG- potential games, GP- games played, and GS- games started.  The calcuation is simple: (PG/GP) + (GS/GP)= value.  Given this equation, the maximum value for any player is 2, which would imply that the player played and started in every possible game after they were drafted.

	1	PG	GP	GS	Value	2	PG	GP	GS	Value
2004	DL- Vince Wilfork	119	113	103	1.861	DL- Marquise Hill	48	12	0	0.250
	TE- Ben Watson	96	87	63	1.630
2005	OL- Logan Mankins	119	106	106	1.891	-
2006	RB- Laurence Maroney	64	45	14	1.014	WR- Chad Jackson	24	14	1	0.655
2007	S- Brandon Meriweather	64	64	40	1.625
2008	LB- Jerod Mayo	64	50	48	1.741	CB- Terrence Wheatley	32	11	1	0.435
2009	-					S- Pat Chung	39	36	20	1.479
						DL- Ron Brace	39	22	7	0.882
						CB- Darius Butler	32	29	8	1.182
						OL- Sebastian Vollmer	39	32	25	1.602
2010	CB- Devin McCourty	23	23	23	2.000	TE- Rob Gronkowski	23	23	18	1.783
						OLB/DE- Jermaine Cunningham	23	20	11	1.420
						LB- Brandon Spikes	23	18	11	1.394
2011	OT- Nate Solder	7	7	6	1.857	DB- Ras-I Dowling	7	2	2	1.286
						RB- Shane Vereen	7	3	0	0.429

3	PG	GP	GS	Value	4	PG	GP	GS	Value
S- Guss Scott	32	5	2	0.556	s- Dexter Reid	16	13	2	0.966
					RB- Cedric Cobbs	16	3	0	0.188
CB- Ellis Hobbs	64	63	49	1.762	S- James Sanders	96	84	50	1.470
OL- Nick Kazcur	90	70	64	1.692
TE- Dave Thomas	48	32	13	1.073	TE- Garrett Mills	16	0	0	0.000
					K- Stephen Gostowski	87	79	79	1.908
					DL- Kareem Brown	16	0	0	0.000
LB- Shawn Crable	18	6	0	0.333	CB- Jonathan Wilhite	48	39	13	1.146
QB- Kevin O'Connell	16	2	0	0.125
WR- Brandon Tate	32	18	11	1.174	OL- Rich Ohrnberger	23	5	0	0.217
LB- Tyrone McKenzie	16	0	0	0.000
WR- Taylor Price	23	3	0	0.130	TE- Aaron Hernandez	23	19	12	1.458
RB- Stevan Ridley	7	6	1	1.024	-
QB- Ryan Mallett	7	0	0	0.000

5	PG	GP	GS	Value	6	PG	GP	GS	Value	7	PG	GP	GS	Value
WR- P.K. Sam	16	2	0	0.125						CB- Christian Morton	0	0	0	0
LB- Ryan Claridge	16	0	0	0.000						QB- Matt Cassel	64	30	15	0.969
										TE- Andy Stokes	0	0	0	0.000
OL- Ryan O'Callaghan	32	26	7	1.082	DL- Jeremy Mincey	0	0	0	0	CB- Willie Andrews	32	30	0	0.938
					OL- Dan Stevenson	8	0	0	0
					DL- Le Kevin Smith	48	31	0	0.646
OL- Clint Oldenburg	0	0	0	0.000	LB- Justin Rogers	0	0	0	0	LB- Oscar Lua	16	0	0	0.000
					CB- Mike Richardson	16	10	0	0.625	OL- Mike Elgin	0	0	0	0.000
					RB- Justise Hairston	0	0	0	0
					OL- Corey Hilliard	0	0	0	0
WR- Matthew Slater	55	50	0	0.909	LB- Bo Ruud	0	0	0	0
OL- George Bussey	0	0	0	0.000	LS- Jake Ingram	24	24	24	2	WR- Julian Edelman	39	30	10	1.103
					DL- Myron Pryor	39	24	2	0.699	DL- Darryl Richard	16	0	0	0.000
P- Zoltan Mesko	23	23	23	2.000	C- Ted Larson	0	0	0	0	OT- Thomas Welch	0	0	0	0.000
										DL- Brandon Deaderick	23	11	4	0.842
										DL- Kade Weston	0	0	0	0.000
										QB- Zac Robinson	0	0	0	0.000
OL- Marcus Cannon	7	0	0	0.000	LB/DE Markell Carter	0	0	0	0	DB- Malcolm Williams	0	0	0	0.000
TE- Lee Smith	0	0	0	0.000

From these values, we can then take total value per round and then assess average value per player:

	1	2	3	4	5	6	7
Total	13.620	12.794	7.870	7.353	4.116	3.970	3.851
avg	1.702	1.066	0.715	0.817	0.457	0.331	0.296

The trend in total value is exactly what you would hope for; i.e., the Patriots are getting the maximum value out of their first round picks and that value decreases as the rounds progress.  However, we have to see that Stephen Gostkowski, Aaron Hernandez, and James Sanders all push the average value of the fourth round higher than the third.

This statistic shows one major thing- the Patriots don't appear to be having major busts when it counts.  They are getting great value (at least in terms of starts) from their first round picks, with the average per pick of 1.7 indicating that each player drafted in that round both plays and starts in almost every game while they're with the team.  Instant and prolonged contribution is exactly what a team is looking for out of a first round pick.

Then, we go to total and average value per position:

	LB	DL	WR	RB	QB	S	OL	K, P, LS	DB	TE
Total	3.468	5.180	4.095	2.654	1.094	6.096	8.341	5.908	8.087	5.944
avg	0.434	0.432	0.683	0.531	0.365	1.219	0.596	1.969	0.899	0.849

I take a few things from here (and for the few people that are still reading this, thank you kindly): We can really disregard the special positions category (K, P, LS) because they weren't actual starts, but I considered them so. 

-        Surprisingly, the defensive back position had the best average value and nearly the highest total value.  This obviously isn't taking into account production, as I said, but instead shows that the cornerbacks that are getting drafted are playing a good percentage of the time.

-        As offensive line was the highest "perceived value" in the draft, it has also become the most valuable position over the last eight years in terms of games played and started.  The average is low, but this is mostly because lineman are drafted for depth in the later rounds, and the turnover in the position is relatively low.

-        The third highest average value is for tight ends, which makes sense- Dave Thomas in his years here, Ben Watson, and recently Gronkowski and Hernandez.

-        Though most people will gripe about Brandon Meriweather's production, the fact is that in terms of games played and started, the Patriots are very efficient with drafting safeties.  This is evidenced by the fact that the average value is an insane 1.219, which means the Patriots drafted mostly only safeties that have an impact.

-        Linebacker and defensive end are unsurprisingly our weakest positions during this time period (quarterback not included).  As most people have cried for impact players in those positions for the years after New England's 3 Super Bowls, these numbers would definitely indicate that the Patriots are not getting very consistent starters with the linebackers and defensive linemen they draft.  Obviously, this doesn't take into account free agent acquisitions or players traded for.

These are all the immediate impressions I take from this data.  I'm sure I'll have a head slapping moment between this week and next, but I will obviously write about that in the future.  As I said, I will definitely have a comparison team broken down in the future- I'm thinking the Colts or the Packers, but post in the comments if you want another team looked at.

A few more quick notes from the team:

-        I'll be disappointed and revolted if the allegations regarding Julian Edelman's conduct are a reality, and he is found guilty.  Though he's been unproductive as of recent, I've always been a fan of Julian and if these allegations are true, I hope that Robert Kraft takes action.  I made my sentiments on Albert Haynesworth known before, but violence against women is absolutely unacceptable no matter one's occupation.

-        A thought popped into my head today- I'm not sure the reason for Leigh Bodden being released, and I'm not certain we will ever know.  Most have speculated that he was no happy in the "star" role.  It definitely wasn't a financial transactions because New England will likely have to pay his remaining salary.  Is it possible that the move had nothing to do with Bodden himself?  Leigh was the Patriots' oldest secondary member by far, and definitely the highest paid.  With Devin McCourty being named a captain, it's obvious that he represents what the Patriots are looking for in a leader.  I've heard Patrick Chung discussed in the same hard working and studious manner as Devin.  Is it possible that at his relatively high age, and level of financial security, that Bodden wasn't ready to study as hard or put in as much work?  Not as a slight to Bodden, but he wasn't playing for his first non-rookie contact.  Even if Leigh wasn't acting as a malcontent, maybe Belichick wanted his youthful secondary members to mold themselves in the vein of McCourty and Chung.

-        We might get to see more of Taylor Price this weekend.  Awesome!