On Monday, we covered some of the hidden indicators that seem to do a good job of projecting a team’s performance going forward. We also penned a love letter to the implementation of statistics in the conclusion of that article, but the reality is that stats simply don’t go far enough in 2012 to justify using them in the same way that they’re used in baseball or even basketball. They’re more useful on a team level than on an individual one.
The good news is that things are getting better. New opportunities for analysis and information are being opened up by new forms of technology and changes surrounding the game, and while those new concepts might not always take the form of traditional statistics, they’re going to help us navigate that void of darkness and misinformation that statistics try to quantify.
Specifically, there are three things happening right now that should dramatically change the way that we, as fans, interpret and understand professional football. They’ll morph into things that we can’t even imagine today,1 but we’ve gazed into our crystal ball and tried to find places where stats don’t currently go that could end up being the source of new forms of information, both for fans and for the league itself.
Video Player Tracking
You might have heard about the SportVU camera system that’s becoming more and more ubiquitous around the NBA. For the uninitiated (and the too lazy to click the link), SportVU hangs a number of cameras at the top of a number of NBA arenas to track and gather meaning from the movements of players. That data gets split in seemingly infinite ways; traditional video allows teams to figure out how effective their point guard is on pick-and-rolls, but SportVU allows them to figure out how effective that same guard is off of a pick-and-roll when a defender is within two feet of the ball handler. Even better, the system gathers all that information and processes it within about one minute.
Obviously, this sort of technology holds all kinds of possibilities for football, too. SportVU would require far more cameras to cover a full 100-yard field, but imagine all the things this sort of equipment could track! For one, we would be able to find out who the fastest player in football truly is. No more relying on years-old 40-yard-dash times from guys in shorts. There would also be more functional information coming out of the system, including:
1. Range and Motion Capture Let’s say a team wanted to sign a safety to play center field in their defense. Cameras would be able to track the functional range of that safety when he had played center field in the past, and to tell the team how long it took the receiver to break on a thrown ball and how much ground they could reasonably expect him to cover. We would get clear images of just how crisp certain receivers’ routes are and then, using context-based statistics, tie that information into how effective those receivers actually are. Oh, and there’s one more thing that SportVU currently tracks at NBA games
2. Deeper Quarterback Analysis the ball itself. Apply that same tracking ability to the actual pigskin being tossed around and there’s all kinds of new insights available for examining quarterbacks. Yes, we could get an idea of which quarterback has the strongest arm or made the longest throw. More important, we could use the information from the video to start considering bigger-picture questions. Are there certain quarterbacks who hit their receivers in stride more frequently than others? Do some force their receivers to do more work than it might otherwise seem? And, then, can we tie that into a stat like yards after catch? How many YAC does a quarterback produce by hitting his receiver in stride over the middle as opposed to throwing a pass behind him?
3. Examining the Quarterback-Receiver Relationship It’s still virtually impossible to extricate the performance of a receiver from the work done by his quarterback, a fact that colors our perceptions of those players. That goes for basic metrics like receptions and receiving yards (which don’t account for usage, which is like measuring a player by his hits as opposed to his batting average) and even advanced metrics like player DVOA (which don’t do a great job of considering a player’s role within his specific offensive scheme). With video, we might be able to separate the quarterback’s contributions from that of the receiver. How often does a given receiver catch a pass while nobody is within two feet of him? How often does a quarterback “miss” said open receiver?
If these questions just seem like theoretical exercises designed to keep a statistically inclined football writer in a job for the next few decades, you’re not totally off, but there’s more to it than that. Take the Jaguars, who just signed Laurent Robinson to a big contract this offseason after he put up impressive numbers with the Cowboys as their fourth option in the passing game. Some of Robinson’s performance came as a starter in place of the injured Miles Austin, but how did Robinson’s style of play change when he was in the slot? Was he just uncovered more often than not? Did he overwhelm inferior nickel and dime cornerbacks? How did he play differently from his time in St. Louis, where he was used more frequently as a primary or secondary target? How did Tony Romo’s abilities gel with Robinson’s? And how likely is that chemistry to reoccur with Blaine Gabbert in Jacksonville? None of those questions are enough to serve as the final say on whether paying Robinson was a good idea, but if the Jaguars could answer all those questions with data, they would have a far better idea of whether they would have been making the right move by signing Robinson.
4. Figuring Out Which Running Backs Do the “Little Things” For ballcarriers, statistical bureaus already track events like yards after first contact. That’s a step in the right direction, but there are far more interesting things to wonder about. Which ballcarriers, for one, get extra yards by churning and driving after last contact? Which ones lean forward and get an extra half-yard when they’re inches away from the ground? And we can finally start addressing that famed Jim Brown maxim — are there certain running backs who go out of bounds more often than others? Do those backs stay healthier than the guys who stick inbounds and take a hit?
5. Learning the Playbook Think about it. With 22 guys running around having their every twitch tracked by a number of cameras, wouldn’t it be easy to create a facsimile playbook for every team in the league? Variations and adjustments made at the line would limit the effectiveness of having that information, but we could begin to get an idea of what sorts of route combinations and blitz packages coaches rely upon when they need a first down. (Further reading required: Chris Brown’s piece from yesterday on “packaged plays”.)
The NFL’s decision to open up the All-22 film to public scrutiny won’t turn every fan into an X‘s and O‘s expert or make every aspect of player performance quantifiable, but it will provide something very valuable for players at a number of positions: a clear view. With coaches’ film, we should be able to reliably track which players are actually on the field, producing player participation data for the first time. While the traditional sideline angle bunches up the offensive linemen and only shows safeties downfield when they screw up, the end zone angle will allow us to watch these players perform their craft from before the ball is snapped to the end of the play. That alone is valuable, and it’s going to create new opportunities for analysis. Like what?
6. Measures That Get Beyond Sacks for Pass Rushers Top pass rushers like DeMarcus Ware or Julius Peppers will be on the field for about 950 snaps a year if they stay healthy for all 16 games. They force the offense to account for them on virtually every one of those snaps, regardless of whether it’s a designed pass play or a run, merely by being on the field. They create opportunities for inferior players around them to look good, in the same way that a Calvin Johnson draws coverage and opens up opportunities for Titus Young, but it’s impossible to track and rarely gets mentioned. Even if they have great seasons against the run and create havoc at the line of scrimmage, the difference between a “great” season for Ware and Peppers and a “disappointing” one can come down to seven steps. Those seven steps might be the ones that force a quarterback to panic, tuck the ball, and take a sack, and could serve as the difference between a 15-sack campaign (adoration) and an eight-sack one (clownfrauderation).
Charting groups have attempted to improve on this lack of information by adding details on quarterback knockdowns and hurries, but one man’s hurry is another man’s day off; totals tend to differ from group to group and from person to person within those groups, and there’s no clear indicator of what a hurry actually is. Instead, the All-22 will allow for new modes of analysis. How often does a player beat an offensive lineman who doesn’t get visible help? How often is he chipped by a tight end or running back when they line up on his side of the ball? How often is he visibly double-teamed?
7. Samesies for Offensive Linemen There will never be an all-in-one metric that encompasses every aspect of offensive line play for every player in the league. Without knowledge of the play call and the player’s intent, you can’t judge the player’s process properly. You can count events, though, and that’s where the new opportunities will arise. Which tackles get the most visible help in pass coverage, and which spend the most time locked up one-on-one against a pass rusher? Which linemen get the farthest downfield or pull the quickest?
8. Development of Unit Statistics Basketball researchers have found meaning in the successes and failures of certain five-man units, noting that certain lineups and groups of players tend to play better than others. It’s a concept that analytics-friendly teams like the Mavericks take to heart. Groupings of football players might not produce information that’s quite as strong, but it’s a topic that will at least be explorable with this new data. Is there a certain five-person set of running backs, receivers, and tight ends for a particular team that outperforms all the others? Are there certain archetypes within those five-man groupings that work better across all teams?2 Are there certain four-man defensive lines that do better than other rotations? This is all information worth considering, even if it doesn’t eventually prove revelatory.
Improvements in Personal Technology
The Browns once lined up Donte’ Stallworth to play across from Braylon Edwards at wide receiver, a move I thought made no sense. They’re both downfield receivers who struggled with intermediate routes and would be playing out of position as possession receivers. You can argue that it failed because the Browns had Derek Anderson at quarterback, but Anderson and Edwards had put together good years during the previous campaign with a consummate possession receiver — Joe Jurevicius — starting at wideout. I suspect that the ideal combination is something like the 2007 Patriots, who had two deep threats (Stallworth and Randy Moss) mixed in with a great slot receiver (Wes Welker), a possession receiver (Jabar Gaffney), and an excellent dumpoff back (Kevin Faulk). Point is: That’s something testable with different player types if we know which players are actually on the field.
The NFL isn’t exactly a league that’s amenable to advancements in technology, but that’s begun to change in recent years. The aforementioned coaches’ film is now available on the Internet for $60. Physical playbooks are being replaced by iPads, as D.J. Williams showed us this May. What’s next?
9. Better Stuff on the Sideline For all those advancements, we still live in a world where NFL teams are taking still photos of plays as they happen in real time, printing them out upstairs (since printers aren’t allowed on the sidelines), and sprinting them down to the coaches in binders so they can go over them with players. In 2012. That’s not going to be the case much longer. It’s absurd that coaches don’t have access to computers on the sidelines to go over that sort of information with players; most dorm rooms have better views of what happened on a play, and to be honest, it’s not like there’s an NFL team out there that can’t afford the expense of buying a few extra computers or tablets. Nobody’s going to be getting a significant competitive advantage if they have access to film in real time on the sidelines.
That opens up other opportunities. What if coaches had the ability to use a tablet as opposed to a laminated play chart? Or if teams were able to build their own program that told the coach exactly what to do on an upcoming fourth-and-short? And then what would happen when it rained and those same coaches had to go back to the laminated chart? Would that be a competitive advantage? Sure, but so what? It’s no different than a bad-weather team having the advantage of playing at home.
10. Improved Refereeing Let’s finish with one that’s rather topical these days. The replacement referees have been the biggest story of this preseason, but the possibilities here aren’t built around replacing referees.3 If we’re improving the on-field technology available to people on that field in real time, how can that improve refereeing? Judgment calls on things like pass interference won’t be any better, but what if we could actually track where the ball was at the moment a player was tackled? That could turn the dozens of bad spots we see into good ones that could turn a season.
Just one note about the replacement referees. Yes, they’re bad, but merely counting the bad calls they make or noting their mistakes isn’t enough. You have to compare them to the actual referees to get an idea of how much worse they actually are, and you know what? The replacement refs haven’t done this yet.
In fact, let’s just go to the next level. Why doesn’t the end zone have LED lights that change color when the football has actually crossed the plane of the end zone? Why don’t we have a tiny camera in the ball that actually allows us to see what goes on in the pile after a fumble? Why isn’t the “unofficial” first-down marker that we see on television actually projectable onto the field? In 10 years, let’s hope that the technology is there and the NFL’s actually using it.