Sunday, May 15, 2016

Measuring the NFL 2013 EDGE Rusher Class


We have heard NFL media sources state it is premature to judge any draft class until the players have had 3 seasons under their belts.  While the 2013 draft class as a whole will be open to scrutiny at the end of the upcoming NFL season since this will be their fourth season and expectations will expect to be met, this article will provide an update of the “EDGE” rusher class of that year.

The Top Prospects

After the 2012-2013 college football season, NFL.com provided grades for all prospect players.  Here are several of the DL/LB players who come out that year who were considered “EDGE” prospects:

  1. Dion Jordan, DE/OLB Oregon – 6’6” 248 lbs.  NFL Grade: 93.5.  After missing his high school senior season battling injuries from a home fire, Jordan converted from TE to DE his sophomore year at OU.  Injuries led to him missing the Senior Bowl.  Upside based on his physical tools led to him being projected a Top 5 pick.   
  2. Ziggy Ansah, DE BYU – 6’5” 271, lbs.  NFL Grade: 93.8.  Another late bloomer, Ansah originally tried out for the basketball and track teams before attempting football for the first time ever in his sophomore year.  Despite many obstacles, Ansah worked at learning the game and bulking up to earn a starting job in his final year. Clearly, lack of experience was one of many concerns for NFL teams. 
  3. Barkevious Mingo, DE LSU – 6’4”, 241 lbs.  NFL Grade: 92.8.  A high school standout at DE, Mingo worked his way up the talented LSU depth chart to have a solid career.
  4. Jarvis Jones , OLB Georgia -  6’2” 245 lbs.  NFL Grade: 88.8.  Despite transfer from USC and a season ending neck injury, Jones was a candidate for every major college football defensive category for which he qualified. 
  5. Bjoern Werner, DE FSU –  6’3’, 266 lbs.  NFL Grade: 88.2.  Coming to America as a high school exchange student with club football experience from his native Germany, Werner eventually committed to FSU where he had a productive career.  He passed on his senior year and entered the draft early.

Other top prospects included:
  • Datone Jones, DE UCLA – 6’4”, 283 lbs. NFL Grade: 85.6 
  • Damontre Moore, DE TAMU – 6’4”, 250 lbs.  NFL Grade: 87.9 
  • Margus Hunt, DE SMU – 6’8”, 277 lbs.  NFL Grade: 75.4 
  • Cornelius Washington, OLB Georgia  6’4”, 265 lbs. NFL Grade: 69.4  
  • Tank Carradine, DE FSU 6’4” 276 lbs.  NFL Grade: 85.4 
  • William Gholston, DE Michigan State 6’6” 281 lbs.  NFL Grade: 74.1 
  • Alex Okafor, DE Texas 6’4” 264 lbs.  NFL Grade: 85.2


The Combine

Combine measurable numbers play a significant role in how NFL teams rank players and construct their draft boards.   Of course, because player evaluation metrics are totally subjective, there is no way to know what one team values more than the next.  Here are the numbers for the prospects included in the defensive end group at the 2013 NFL Combine (“Pos” represents the position listed for each player on NFL.com):
(Note:  Because there is no Combine data for Jarvis Jones, who was drafted #17 overall by Pittsburgh, his data is not included in the analyses contained in this article.)
Ratios Used for Analysis

From the above data, the following measures can be calculated:
  • Density calculated as weight in lbs/height in inches.  This permits us to normalize the differences in height and weight that naturally exist between players.  This number is best used to compare a player to his position average. Obviously, there will be players outside the average range (within 1 standard deviation) who are impactful players despite being heavier or lighter than the average.
  • Vertical Jump to Height ratio (also, “Vert%).  The use of raw vertical jump results can be potentially misleading as the point of the vertical is to measure how much power one can generate, not so much how high one can jump off the ground.  This measure gauges the percentage of a player’s height he is able to jump off the ground, again, attempting to normalize.  The higher the better within that player’s position group.   
  • Broad Jump Index (also, “BJI”).  This measure attempts to measure the “work” a player exerts during the drill to measure power generation (explosiveness).  Because a heavier man has a natural disadvantage, the broad jump is multiplied by the player’s Density, resulting in an index number.  Again, the higher the better within that player’s position group.   

Vertical to Height ratio and Broad Jump Index are graded to the group average along with 3 Cone, bench and 40 yard sprint times and players are awarded points as follows:



The total points earned are measured against potential points (some players did not perform all drills) for an overall score.  For example, Cornelius Washington had a 40 time of 4.55, (compared to 4.63 which is 1 standard deviation from the average) cranked out 36 reps on the bench (compared to 27.7), had the top Vert% of 51.32% and top BJI of 446.32 resulting in a perfect 12 points out of a possible 12 (I had no 3 cone number for him, so his max was 12, not 15 points).   Below are the scores for each Combine participant based on the performance metrics:


The combine performers with “passing grades” greater than 1 SD above average were Washington, Mingo, Powell, Hunt and Stuart.  While other factors go into the GMs’ player rankings, we would expect to see some correlation with draft position and combine performance.  Here is the table above sorted by draft position:



(Note: Walt Stewart was omitted from the list as he went undrafted due to medical issues.)
It is interesting to see that of the 5 first rounders analyzed, only one earned an “Exceptional” overall rating from our Combine analysis.  Two earned above average grades while two were below average.  Our top performer, Washington, didn’t get drafted until the sixth round. 

Fast Forward – 2015 Performance


So, ranking of the EDGE rusher group from the 2013 NFL Player Combine based on our analysis of participant performance did NOT match up very well with where the players actually went.  Athletic achievement at the combine didn’t translate into the high draft positions (and the higher salaries) these players had hoped they would receive.  Certainly, the NFL execs, armed with untold resources and scouting staffs of invaluable experience, have a leg up on fans and wanna-be scouts alike in this draft thing and knew something we didn’t.  The devil is in the details so there must be some method of interpreting the Combine data to which we average yokels will remain ignorant, horded by the NFL GMs.  Despite our unsophistication, let us attempt to close the loop on our exercise by adding player performance through the end of the 2015 season to the mix.   The list below was sorted by “Career  Sacks” for the sake of presentation (that is, there is no particular emphasis on Sacks for this exercise).
(Note:  Players with less than 3 career sacks were omitted from the list.)
Relative Productivity Metric

From a raw numbers standpoint, it would appear the Detroit Lions hit the jackpot with Ziggy Ansah.  He’s produced solid numbers anyway you look at them and is becoming a household name among the NFL faithful.  His numbers standout but how do we compare him to the rest of the field.   And how do we judge players who may have had a “redshirt” rookie year or have dealt with injuries (Washington )?  By comparing the above results to each player’s participation based on the total snaps played in 2015 (thanks to http://www.footballoutsiders.com/stats/snapcounts), we get a clear picture relative productivity.


Key:
  • Tackles are calculated at 1 tackle for each solo and 0.5 for each assist.
  • Snaps/Sack = The number of snaps played to generate one sack.
  • S/S% = The percentage of player Snaps/Sack ratio compared to the group average.
  • Snaps/T  = The number of snaps played to generate one tackle.
  • S/T% = The percentage of player Snaps/Tackle ratio compared to the group average.
  • Relative Prdctvty = Relative Productivity as the average of S/S% and S/T%.


From the table above, Armonty Bryant is clearly trending to be the most productive EDGE rusher selected in the 2013 NFL Draft.   His one sack every 35.27 snaps is nearly 10 times that of Okafor’s rate of one every 659 snaps.  Also, he produced one tackle every 5.71 snaps, nearly 12 times the rate of “Exceptional” combine performer and #6 overall pick Mingo. 
Here’s a final table showing our Combine grades, the draft positions and Relative Productivity side by side:

It must be acknowledged that there may be any number of unknown factors that may have impacted the final results of this analysis (players not reporting injuries at the Combine that hampered performance, depth chart issues limiting snaps, possible impact on how a player is used within a scheme or sub packages which could have provided some boost to stats, etc.).  Notwithstanding, the numbers would indicate the three player drafted lowest among the sample had the biggest impact despite only one having what could be considered an “Above Average” combine performance.

Outperforming Players
  • Armonty Bryant (#1 Relative Productivity; Below Average Combine Performance, drafted 217th overall), who attended DII East Central University in Ada, Oklahoma (anyone who can reply and identify what the East Central mascot is, without looking it up, will win a prize of subjective value!) , seems like the steal of the draft for the Browns.  But, (sorry CLE fans) this is the Browns we’re talking about. Bryant will miss the first 4 games of the season for violating league rules. It should be noted that back in college, Bryant sold drugs to an undercover cop and was subsequently arrested during practice. (http://www.silverandblackpride.com/2013/4/7/4192900/ecus-de-armonty-bryant-the-true-diamond-in-the-rough)   If it’s any consolation, despite being suspended for EIGHT games that year, he racked up 10.5 sacks and 54 tackles.
  • Devin Taylor (#2 Relative Productivity; Above Average Combine Performance, drafted 132th overall) played at the real USC where his impressive resume littered with team leading stats didn’t seem to make much noise to many, likely because  a kid named Jadeveon Clowney was the talk of the town.  Looking back at Taylor, he’s been a leader and winner at every level.  He and Ansah will be causing chaos for the Lions for years to come.
  • David Bass (#3 Relative Productivity; Out of Range Combine Performance, drafted 233rd overall) produced the highest tackle rate and the fourth highest sack rate despite playing the fewest snaps in the group (only 154 snaps).  He attended Missouri Western State (dreaded rivals of Kansas Eastern, I can only assume).  It’s clear why he wasn’t recruited out of school.  He was only a  three sport HS star and named to the All  Metro team in the competitive St. Louis area.  Who was elected captain of the football team.  And the basketball team.  Oh, did I mention he was captain of both teams his Junior AND Senior years?  He had 7 Passes Defended in his final college season as a Defensive End.  That’s as many as first round DB’s Eli Apple and more than Alexander Mackensie(5), Artie Burns (5) and Vernon Hargreaves (4).   Drafted by the Raiders, Bass is on his third team (he racked up 3 sacks in 8 games of action during the 2014 season for the Bears before he was cut), he will continue to provide depth to the Titans defensive line.


With some players moving to new teams and others looking to breakout and make names for themselves (not the least of which being the three named above), it will be interesting to see how the 2013 EDGE draftees perform in the upcoming 2016 season.    

Tuesday, May 3, 2016

2016 NFL Draft for Value Rankings

In every aspect of our lives, we're looking for value. 

Getting more for less. In the NFL Draft, it's all about taking advantage of the chaotic environment to find players other teams may have overlooked and avoiding the temptation to draft a player earlier than the market would dictate. This analysis attempts to use the expected position salaries from the first overall pick to the final man selected to assess the cost of each player compared to the potential value he may have to the franchise.

Assuming 2014 rookie salary figures for the recently drafted players (that is, Goff’s salary will be the same as the first player picked in 2014, Wentz's would be the same as the second player selected in 2014 and so on),  we can assume the value of a player to a franchise from a financial commitment level perspective.  This allows us to assume certain cost comparison values. Accordingly, Goff is 38.9x more costly than the last player drafted (please see the post for the Player Comparison Tool, which will allow you to compare players based cost comparison value).  Based on our reasoning, he is also 3.43x more valuable to his franchise than newly drafted Bronco Paxton Lynch.   So, if  Goff does not provide 3.43x more impact to his team, we could say the Broncos got a tremendous value in Lynch. Of course, the subjectivity of what is "impact" is material for another day, but this exercise is designed to help quantify an immeasurable value.  Remember, all figures relate to the #1 overall spot.  The higher the number, the lower the cost and the higher the potential for value. Using this comparative value concept as a framework for the rest of our analysis, we come up with the following three measures:

Average Overall Value/Overall #1 Pick – This measures the average value of each player’s comparable value to the Overall #1 pick.  Essentially, the higher the number,the lower in the draft the average player on that team was selected.  This number alone does not give us an idea of how well the Eagles selected because draft pick position is not an indicator of quality.  The Eagles were the top team with average picks at a cheap 27.9x less cost than the #1 pick. The Saints had the costliest picks, an average of 16.16x less than Goff.

Average Position Value/#1 – This measures each position player by their cost in relation to the #1 drafted player in their position.   For example, Joey Bosa was the #1 DE drafted so he has a value of 1.00x.  Shaq Lawson was the third DE to come off the board and, based on value to the franchise, Bosa is 2.86x more valuable than Lawson. If Bosa’s impact over the course of their rookie contracts is not 2.86x greater, the Bills got a tremendous value in Lawson. Going through each teams picks and comparing them to the player picked #1 at each position, the Niners got the best position value at 19.33x, while the Saints picked players early once their position “opened” after the #1 player was drafted, paying 4.17x

Average Sequential Pick Value – How soon after the last player  at a certain position was drafted did a team draft the same position?  In Rounds 2-7, if you select a position after it had just been selected, your value is going to be closer to 1.00x since salary values don’t decline as rapidly in these rounds as in the first. Big values can be found drafting outside of the current round.   For example, Kamalei Correa at #42 and Jihad Ward  at #44 are of nearly identical value coming out of the second round, but both are 2.02x more valuable than Carl Nassib, drafted in the top of the 3rd round.  If Nassib performs on par with the other two, he was a major steel. The biggest gap (rightfully so in some minds) was at running back where Zeke Elliott is 6.55x more valuable than Derrick Henry.   The Titans got nice value with an avg of 1.67x when choosing the next position player while the Steelers seemed to be copy cats with an avg. sequential pick value of 1.03x.

Aggregate Draft Value – This calculated the average rank (1-32) in each of the three other categories. The Top 3 in terms of picking for value are the Niners, Eagles and Vikings.  The bottom three are the Saints (#32), Buccaneers (#31) and Panthers (#30).  To get the entire picture, we have to assess the quality of players they got at those perspective values.

Average Player Grade - Player grades were assigned based on the NFL.com DraftTracker  That Can Be Seen Here .  The teams with the highest average player grade were the Jaguars, Titans and Bears.  The lowest were the Patriots (#32), Chiefs (#31) and Eagles (30).

Best Draft for Value – Averaging the Aggregate Draft Value with the Average Player Grade give us the final Value Rank for each team.  The results are here:  Google Drive 2016 NFL Team Draft Value Comparison Final value grade is in the last column on the right. 

Art of the Deal – The four teams pulling off block buster trades had varied results.
  • Titans – Traded out of #1 position had the 4th best value based draft.
  • Ram – Traded into the #1 spot and was 24th on the list.
  • Browns – Trade out of the #2 position and had the 28th rated draft based on value.
  • Eagles – Moved up to #2 and came away with the 14th rated draft from a value perspective.

Rising in the East - The AFC East occupied the top 3 spots.
Southern Blues – Teams from the NFC South occupied 3 of the 4 bottom spots.

Most Picks
  • CLE (14) – 28th                    
  • BAL (11) - 15th
  • SF (11) – 7th


Fewest Picks
  • CAR (5) -32nd
  • NO (5) -25th       
Despite all the numbers and rationale, we won't know who "won" the draft until they line up to play in September.  As we go over this draft in three years, we will see who really got the best value out of this draft.  Usually, we'll know by which GMs and head coaches are still employed.  

Monday, May 2, 2016

2016 NFL Draft Comparison tool

Hello children,

I'm sure everyone enjoyed the NFL draft this weekend.  Talk about drama!  I'm working on a piece detailing who "won" the draft based on the actual "value" of each player to the franchise based on projected salaries. As a financial guy, I'm always measuring things in terms of the value of this compared to the value of that.  In a nutshell, if the first overall player selected is awarded salary "X" and you as a 3 round pick are awarded 1/10 X as a salary, then the #1 player has 10x more value to his franchise than you to yours.   On a team basis, this average Position Value figure is measured against the average player grade (using NFL.com for now).  As time goes on, we will be able to really judge the value of one player to another in general terms.

So, to whet the appetite, here is a spreadsheet that will allow you to compare the value of any one drafted player to another based on projected salary.  For example, Ezekiel Elliott is 6.55x more valuable than Derrick Henry.  That's a huge number especially for consecutively selected men playing the same position.

Here is the link to the spreadsheet:

 Boombear's 2016 NFL Draft Player Value Comparison Tool

Have fun with it and let me know if it bugs out on you.

Thanks.