Tom Brady had an incredible overtime at the Super Bowl last weekend.
After having come back from the biggest deficit in Super Bowl history, he completed passes for first 6, then 14, and then 16 yards in quick succession. After a loss of three yards, Brady completed again for 15 yards to put the Patriots deep into the Falcons half. After an interference call that had Atlanta fans cursing, Brady and his team lined up at the two-yard line. Two plays and an almost interception later and James White scored the game-winning touchdown that delivered Brady his fifth Super Bowl ring.
What a game: who could have predicted that sort of comeback? Nobody, right?
But there was a lot about the game, about the Patriots victory, and about Tom Brady’s passes that was predictable – family easily predictable.
For example, Tom Brady completed 43 out of 62 of the passes he threw, a Super Bowl record. How was it possible for him to hit his receiver so often? The computations in the huddle or on the sidelines light be complicated, but the physics are simple. A football thrown at a certain angle and velocity will reach a point in space at a specified time, and a receiver where that ball is going to be can arrange to be there to catch it. There isn’t a lot of physics being done on the field, of course, but hours of training drills guarantee that when Brady calls the play and releases the ball as planned, his receiver knows where that ball will be a few seconds later.
Other parts of the Super Bowl are less simple, however. Indeed, the Super Bowl is a fantastic example of multiple complex systems interacting in fascinating ways.
Take the ticket price, for example. Officially, seats at the game cost anywhere from $2,500 to $12,000 but prices vary per public demand, the teams who make the game, and availability of tickets on the resale market. In 2013 the average resale price for a ticket to the Ravens-49ers matchup was $2,173. In 2015 when the Patriots met the Seahawks the average re-sale price jumped to nearly $7,000. This year the average price hovered around $5,000. Predicting where Tom Brady is going to place the ball is easy but predicting what it will cost to watch him do so is hard.
Or consider how the fans in the stadium got to the game. In 2014 New Jersey’s MetLife Stadium played host to Super Bowl XLVIII between the Broncos and the Seahawks. It was billed as “the first mass transit Super Bowl” when almost all the 80,000 fans who arrived at the game did so by either rail or bus. In a European city, this might not make for much complexity; indeed, in my home city of Lyon almost all of the fans who voyage to the 60,000 seat Parc OL stadium do so by tram or bus, and the price of the transport is included in the price of the seat. However, in the case of the MetLife stadium, it took coordination between five different authorities to get everyone to the stadium on-time. New Jersey Transit, New York’s MTA, the PATH system, the interstate Amtrak system, and even New York Waterways had to align schedules, coordinate timing, deal with congestion, and anticipate traffic snafus in tight windows of only a couple of hours pre- and post-game. What’s more, they had to do all this while still serving the other 11 million mass transit passengers who travel in the region around the stadium each day.
At CoSMo we’ve worked with transportation networks and, trust me, if it sounds like a complex task you don’t know the half of it!
How can this complexity be understood and managed in the optimal manner?
Unlike a Tom Brady pass where the future position of the ball can be predicted and calculated for any angle, velocity, rate of spin, and even temperature with relative ease, there is no simple way to calculate the future position or trajectory of a complex system. It’s not that we don’t understand the mathematics, it’s just that understanding the mathematics doesn’t make any difference. We can’t do anything but sit back and wait for the complexity to play out in real time. Put another way, there are no shortcuts when it comes to predicting the future state of a complex system by doing the math.
But while we can’t take a mathematical shortcut to the future state of a complex system, we can model and simulate that system to determine what the future state will be. If we can model all the important systems and sub-systems precisely enough, if we identify the causes of change and the impacts of those changes on the systems and sub-systems, and if we bake in expert knowledge of the systems as well as the latest approaches in complexity science, and then run simulations on that model, well then we can reliably determine what the future state of the system will be.
This modeling and simulation approach is the only way to accurately predict the future of a complex system. Unlike trying to do the complex mathematics learning how a system evolves in real-time and without a shortcut, a simulation can be done in minutes and predict the future state of a complex system decades in advance. It’s the only way for companies, municipalities, and governments to understand how their complex systems will evolve, and the only way to determine which strategies to employ to take advantage of that complexity.