Using Advanced Analytics in 1988- The Box

On December 19, 2018 the NFL made the following announcement

“The National Football League announced the inaugural NFL Big Data Bowl, a football analytics competition affording college students and professionals the opportunity to utilize historical data sets of the same player tracking data used by teams and suggest innovations about how football is played and coached.

Finalists will have the unique opportunity to present to league and club personnel at the 2019 Scouting Combine in Indianapolis, accessing NFL player tracking data used by teams to analyze league trends and develop on-field strategies.

….

Historical sets of both traditional and player tracking stats are available to participants, who are challenged to rethink player evaluation tools and innovate the way football is played and coached. NFL player tracking, also known as Next Gen Stats, is the capture of real-time location data, speed and acceleration for every player on every play on every inch of the field. Sensors throughout the stadium track tags placed on players' shoulder pads, charting individual movements within inches.”

In 2019, the hottest job in well, really any market, is that of a data scientist or a data analyst.  With the advances in technology and ability to collect reams of data, the key for any organization is to look at this data, and pull from it trends and learnings that can be used to inform decisions and produce the best return on investment (ROI).  Bottom line, how can you use data to get an edge.

It’s no surprise that sports have latched on to this as well.  Baseball was far ahead of the curve here.  Now, it also helps that baseball is very much a numbers game, decisions are made based upon statistics.  They went a long way from pitchers matching up against batters to try to get a lefty to pitch against a lefty batter.  Instead, they started to know more about pitchers groundball efficiency, strikeout efficiency, matchups against batters based on how they hit fastballs, etc.  Sabermetrics went from a curiosity that Bill James pushed and the Red Sox first paid attention to, to now you have baseball GMs who are no longer former players, but rather former scouts, and in some cases, pure data statisticians.  In other cases, such as the Yankees, Brian Cashman has at his disposal a huge amount of data analysts looking at reams of information to identify players throughout the Yankees system and across other systems (ie- Luke Voit trade from the Cardinals).

Other sports have gone along with this as well, including the NBA and the Golden State Warriors used advanced metrics to help build their roster into a dynasty.  But the NFL has always seemingly lagged behind.  In this case, the NFL is actually being smart in putting this invite out to the best and brightest to get them involved.  They will attract some good, young talent.  Who probably don’t quite understand that they can earn a lot more money going to work for a Hedge fund or other financial institution, but they can figure that out in their late 20s.  You are young, it’s more fun to be in the NFL doing this.

Several organizations have been credited for being well ahead of others.  The Philadelphia Eagles gave a great deal of credit for their 2017 Super Bowl title (it kills me to type that) to their analytics department working hand in hand with their coaches and scouts.  Now, the NFL has had their own Bill James who has been in and around teams for over 40 years, his name is Ernie Adams.  Adams spent time with the Patriots in the mid 1970s.  Adams joined the Giants in 1979, working for Ray Perkins, where he was an offensive assistant and recommended the Giants hire a young Bill Belichick as Special Teams coach.  Adams moved from a coach to the Giants front office as the pro personnel director, before he left to make more money on Wall Street.  Adams would resurface with Browns when Belichick became the head man in 1991 and returned to finance after Belichick was fired in 1996.  But when Little Bill landed with the Patriots, Adams rejoined him as “Football Research Director.”  Adams role has never been clear, but he has been involved in all aspects of the organization, from scouting, to reviewing that statistics around coaching decisions (ie- when to punt or go for it, going for 2 point conversions, etc.)  He looks at film, he breaks down players and tendencies.  He does it all.  But no one really know what he does.  Former Cleveland Browns owner Art Modell once said “I’ll pay anyone here $10,000 if you tell me what Ernie Adams does.”  What Ernie Adams does, to put it simply, is play the role of a roving data scientist in a football organization and is in lockstep with their Hall of Fame head coach in Belichick, who is a believer in the same attention to detail which helps him make decisions and win championships.  He’s been playing a role that only now has gone mainstream across several industries, and now the NFL has put the “bat signal” out so to speak for the best and brightest to find not the next Ernie Adams but rather 100s of Ernie Adamses.

But the reality is, the NFL has always been data hungry and always looked for more information to get an edge in evaluating players and making a decision.  But technology wasn’t as readily available and the entire sports science industry was nacient at best.  Which leads me to this amazing segment that I found from an old “Inside the NFL” tape I had from 1988.  I’ve written here before that I love HBO’s Inside the NFL and I would love to get my hands on as many old episodes as I can.  I used to tape them in the mid 1980s and was able to find several tapes, but not all of them.  Some were probably taped over.  Others thrown out.  Every now and then one pops up on youtube and I grab it to keep for myself.  In each hour long episode, there were your highlights, with the masterful NFL Films music with the Harry Kalas voice over.  You had your banter between Len and Nick.  You had your pick segment (just winners and losers, no spreads for you dirty gamblers).  And you had a feature segment.  In this case, they focused on the latest and greatest in technology used to evaluate players, the black box, or simply known as “the Box” in an investigative report provided by former Dallas Cowboy WR Drew Pearson.  I’ve posted the whole segment below for your viewing pleasure with a few thoughts:

Ok, where to start

• Drew Pearson starts off by saying “every year teams from around the NFL spend hours compiling information.  Information which might get them that extra edge on draft day.”  Hours?  Ummm..Hours?  I know this was 1988, but I’d like to think back then scouts spent more than a long weekend doing their homework.  Odds are we are talking in terms of months.  For now, certainly we know that’s the case.  And there is a whole job for TV analysts who are scouts who look at film.  Former Giants safety Mike Mayock became a leading voice on the NFL Network as a scout looking at film and was a staple on draft day.  He became so good at it, he was hired by the Raiders as their GM. 

• When Pearson discusses the players speed, specifically speed in the 40 yard dash, that’s something we all, now in the year 2019 can recognize.  Everyone knows about the players speed in the 40.  But they show a 60 second stopwatch with the hand timer, not even digital.  When the world of the difference now between a 4.45 and a 4.65 is literally millions of dollars on draft day, we had the old school stopwatch.  

• Drew is reading “High Technology” magazine.  Outstanding.  Full disclosure, I spent some of my early career in publications, working with magazines and their eventual  migration on to the internet.  Until the internet essentially put magazines out of business.  High Technology was started in 1982 by Bernard A. Goldhirsh, a former engineer at MIT in 1982. He sold it to Infotechnology in 1987.  It apparently closed in 1990. 

• Now we get to the good stuff, a machine was created “in Denver, by an Engineer that would scientifically measure a players agility and aide teams in selecting players.  Players who, in the past, may have been overlooked...it’s referred to as “the Box.”  Okay, now we are cooking with gas.  Some Bronco fan, I guess, came up with this neat little device that can find the diamond in the rough.  Someone who won’t stand out on film, but has other measurables that can help a team draft him.  Sound somewhat familiar in 2019 with data/ sports science.

• The Box costs $15,000 and “only a few teams in the NFL admit to owning one”.  This line cracks me up.  Why are teams reluctant to say that they don’t own one.  If you have a secret weapon, don’t you want other teams to get one.  The fact that, as we find out, the Cleveland Browns admit to having one, wouldn’t you say that you have one too, even if you don’t?  And $15k for a machine that can find players and win Super Bowls?  This guy in Denver really undersold himself.  Ok, but this is 1988, so with inflation, it has to be a ton of money in 2019, right?  Well…$32,140.91.  Maybe NFL teams were just really cheap in spending on scouting.

• OK, now comes the best part of this, hands down.  Mike Lombardi.  This is just wonderful on so many levels.  Lombardi would go on to have a nice long run in the NFL, over the course of 30+ years.  Lombardi would spend time as a scout, and up in the front office as a personnel director to a VP/ GM role.  He would spend time with the 49ers, Browns, Eagles, Raiders, and Patriots, with a couple of stints with Belichick (remember..Ernie Adams and wanting data).  Lombardi has since moved on to the media side, as a writer and podcaster.  But here, he’s 29 years old, he’s an area scout, and he is rocking quite the cop mustache and shades.  Adding in the Mission Impossible music as he strolls in with “the Box” in a suitcase into some high school gym is pure 1980s glory.

• “Sometimes in film it’s hard to really see a guy play athletically.  Sometimes schemes coaches use tend to hinder a kid athletically and it’s hard to get an evaluation on them properly athletically.  All we’re trying to do with the Box is get one more piece of the puzzle and try to find out how athletic a player is and make a determination and evaluation from that point.”  Now this is an outstanding quote from Lombardi and resonates today.  The yearning to get this information, that doesn’t come across on film, is where a scout makes their money.  This is what you heard in 1968 and 1988 and right now we hear it in 2019.  Front offices 30 years ago were trying to answer a question that is still being asked today.

• The guinea pig for us to see is Browns safety Brian Washington.  Washington was a 10th round pick out of Nebraska, playing on a loaded team with NFL talent around him on their defense (Broderick Thomas and Neil Smith in particular).  Washington had decent size (6’1”, 205 pounds).  And he ended up having a nice career.  He spent 2 years with the Browns, would go on to play 5 years with the Jets and finish his final 2 seasons in Kansas City.  In all, a 9 years NFL career, collecting 27 interceptions and scoring 4 TDs.  Any scout would sign up for a player to have that career.

• Next we are treated to seeing Washington put through the paces by the latest and greatest in technology.  And we also see how technology doesn’t necessarily age well at all.  The mystical box, essentially is a small computer with a stopwatch, with touch pad on the floor that based upon someone stepping on it, triggers the circuit to go on or off depending on the drill and tell you the time.  It’s at that point for the tester (in this case Lombardi) to write these times down for their data points.  If you ever watched Mythbusters, they usually built instruments somewhat similar to the Box (but with more advanced technology as that show came on 15 years later).  When you look at what Washington is doing, you actually see drills that are very similar to what we see at the NFL combine and the individual unit drills.  Tracking reaction time.  Doing lateral movement drills which essentially were 3 cone drills.  They also had a string on him, which was stay with him 10 yards and pop off (which would track his 10 yard time) and then a touchpad for him to step on at 20 yards, thus giving the scouts his 10 yard and 20 yard burst time.  When Tom Coughlin was the head coach of the Giants, during the offensive line drills, he used to sit at the 10 yard spot with his own watch and time them.  Why?  Simple.  Offensive lineman are not going to be running 40 yards down field, unless something really good his happening (a teammate has a long run underway) or something really bad (a defensive player is taking the ball back the other way, a very long way, and the OL his hauling ass trying to stop him).  Otherwise, offensive lineman, probably 90% of the snaps, will never see 10 yards down the field.  40 time is irrelevant.  The Box was simply giving a more accurate reading for the 10 and 20 yard accelerations.  This is what teams are now using thanks to chips that are in helmets and pads that track this and even show mph.  Same stuff they were looking for in 1988.  Just, the Box was the best they could do.

• At the end of all these tests, the scout (again Lombardi) takes all this data, plugs it in and uses it as part of their evaluation.  Same thing as today.  Again, everyone is wowed by the number crunching in 2019 and the eggheads involved in the NFL.  They were there 30 years ago too.

• When Lombardi was asked who could benefit from the Box:  “The kind of athlete who possibly didn’t run a good 40 when he was in college.  Ernie Byner did not run a good 40 at East Carolina but he’s been an explosively quick athlete who can accelerate and show a burst.  These kinds of players are what the game is looking for now, guys that can make a quick, explosive start, kind of guys like yourself (talking to Pearson).”  Ok, this is again the key to being a scout.  This is what sets them apart from your average football watcher.  Dave Gettleman said when he drafted Saquon Barkley “My mother could scout him”.  Meaning, he was so good, so dominant, he was obviously a star.  You didn’t need any special evaluation.  No brainer.  But let’s take Ernest Byner, as Lombardi mentioned.  Lombardi said Byner ran a bad 40 time, and I believe him for a couple of reasons.  First, I can’t find his 40 time.  Back in the 1980s, the scouting combine wasn’t nearly what it would become, so not much data was available.  Second, Byner was drafted in the 10th round, so that’s a long way to wait and there had to be a reason.  So much so that if this was 2019, Byner would have been an undrafted free agent.  But Byner was short but powerful (5’10’, 215).  He was smart.  He could catch.  He could block.  Unfortunately for Browns fans, he’s known for his fumble in Denver in the 1987 AFC Championship Game, as he was heading into the end zone to put the Browns in the lead and into the Super Bowl.  But looking at his career, for a guy who the 40 said couldn’t play, he ended up playing 14 years for the Browns, Redskins, and Ravens. He rushed for over 1000 yards 3 times (and had 998 one other year).  He made the Pro Bowl twice.  And in his career he put up almost 13000 all purpose yards and 71 TDs.  Scouts are paid to find Ernest Byners, not Saquon Barkleys.

• We end the piece watching Pearson run the drills and Lombardi basically tells him not to quit his day job.  But a few points came from their end discussion with Len and Nick.  Pearson made a good point “when it’s used in an overall scheme of things, to judge and rate future prospects, the box does have merit, but you have to include it with the 40 yard dash time, the strength and conditioning test, and what a player has done in his college career.  Evaluating all that, and then throwing in “the box”, then yes it has merit.”  It seems obvious, but this is the entire key of the discussion today of use of analytics.  There are aspects of football scouting which will never change, but using the right data, the right way to find players is what all this is about.  It’s the exact goal of the Box, if in a very rudimentary way.  

• “A lot more teams will use the box if they see teams like the Cleveland Browns and players like Brian Washington being discovered because of the Box.”  Again, get rid of “the Box” here and imagine we are talking about analytics in the front office.  Teams that might be more resistant, will start to fall in line because it’s a competitive edge.  The Browns were clearly doing it in the 1980s.  Ernie Adams has been doing it for years.  Now everyone wants in, even if they don’t really know how to use it.

• I enjoyed the final discussion about Steve Pelluer as the Dallas QB, and how Tom Landry didn’t have much patience with him and all the mistakes he’s making that is costing the Cowboys games.  Of course, as is the case with almost every former Cowboy who joins the media, they can never let go of that Star and still will cheerlead for Dallas.  Pearson said “Dallas will be a team to be reckoned with” once they stop making mistakes.  I love watching Dallas fail.  Always and forever and in the past.  This Inside the NFL was taped in Week 5 of the 1988 season.  The Cowboys had lost on Monday Night to the Saints to fall to 2-3.  Pearson seemed to say that Dallas was young and up and coming.  The after the Saints loss, the would drop their next 9 games, so a 10 game losing streak in all.  They finished 1988 at 3-13 as the wheels had come off the Cowboy bandwaggon completely.  Jerry Jones would buy the Cowboys, he’d fire Landry, push out Tex Schram, and put in Jimmy Johnson.  The Cowboys would go 1-15 in 1989.  But he traded Herschel Walker to the Vikings and then…

• Lastly, one thing to keep in mind about this clip.  Dave Gettleman, the Giants GM has gotten a great deal of heat in the media and from fans in how he seems to turn his nose up at analytics.  He shows the signs of the old man yelling at these data analysts to get off his lawn and teams will run circles around him.  He’s going off gut feels and talking about a QB having “instinct”.  All things you can’t measure.  Well, keep this in mind.  The 1988 Browns were on the forefront of this and Lombardi, again, was a young scout in their system.  The GM of those 1988 Browns who Lombardi ultimately worked for?  Ernie Accorsi.  Accorsi would be the Browns GM until he was fired after the 1992 season, and he would end up with the Giants, working under George Young.  Young would leave after the 1997 season and Accorsi would take over as the GM of the Giants, and underneath him he’d groom 2 future NFL GMs.  Jerry Reese.  And Dave Gettleman.