ATLAS TEAM F1


Going Public with Unused Data
by Bill Doolittle

Its Friday morning of a grand prix weekend. Cars are frantically taking to the track, making a lap or two or three, and then darting back into their garages. While parked, the drivers do not dare vacate their seats as there are adjustments to be made and precious few moments to make them. Cars must be back on the track as quickly as possible in order for the drivers to learn the feel of the track and for the cars to collect, through their mazes of electronic circuitry, data about how each car and its driver are performing.

While in the garage, engineers plug their laptop computers into ports located near their car's cockpit. One, plugs-in a monitor and sets it on the cowling in front of the driver. The data collected by the on-board system are downloaded and certain team members inspect information about the previous lap(s). Adjustments are made to the car. The driver reevaluates his strategy, and then shoots back to the track. The whole cycle is repeated several times until the practice session comes to an end.

Many Formula One fans have longed for details about the data that are downloaded, how they appear, how they are interpreted, and how they are used to make adjustments. This curiosity is fueled largely by fans' high level of computer literacy. Most fans use computers with capabilities not unlike those used by the various teams. Curiosity is piqued by the teams tendency to be secretive and not openly show what types of information they have and how they use it. To be sure, most fans, principally because of their high level of electronic and mechanical knowledge, have a pretty good idea about the general nature of the data and how they are analyzed, but when it comes to specific details, they are left in the dark. This is unfortunate for fans and teams alike.

Although we haven't been shown much about on-board data collection in the past, we are about to be shown a tantalyzing bit in the very near future. In the forthcoming edition of his book, The Art of Motor Racing, written with the help of Gordon Kirby, Emerson Fittipaldi devotes an entire chapter to data gathering and analysis. The text is beautifully written and the illustrations are superb. The latter include photographs of "little black boxes", sensors, and instrumentation. They also include charts and diagrams as reproduced directly from the screens of computer monitors. A wealth of information exists in only a few pages! All motor racing fans interested in the electronic side of the sport will surely want to see this chapter.

But, the chapter does not say everything about car and driver data, and for some very good reasons. One of the reasons why some data are not discussed is quite simple--the teams do NOT use the them! According to Fittipaldi, "There are a million things you can measure, and the top Indy Car and F1 teams do measure everything; but we tend not to use the data unless there's a problem. We measure all the suspension travel all the time, for example. We measure ride height to the ground all the time; but we don't really look at the data unless there's a reason to look at it. We record tons of information, but it's only there as a reference, or in case there's a problem we need to troubleshoot."

Given the competitive nature of the highest form of motor racing, it seems obvious that teams would not be willing to share information with other teams. But upon reflection, some things may not be all that "obvious," and they may not be as real as they initially "seem." Myths are eventually seen for exactly what they are. And, this leads to an interesting question: If teams aren't even using certain data, why don't they make them public after a certain period of time?

Now, I'm not calling for an immediate posting of all data, available to every team, as soon as they are downloaded. But, even if I were, what difference would it make? Teams are too busy with their own cars and their own data to invest in watching others during practice and qualifying. Furthermore, the teams who would benefit the most by monitoring the performance of others, are the teams with slow cars and limited financial and human resources. As much as they might like to learn from the Benettons and the Williams, the Arrows and the Minardis of this world simply cannot afford it. Conversely, it is hard to imagine Ferrari monitoring Forti, although they can certainly afford it.

What I have in mind is something different. I would like to see each team put certain of its unused data on-line within a week of their collection. The data to be posted could be left to the discretion of the data which owns them. Presumably, each team would post the data it deems to be the least important or most unlikely to benefit competitors.

Now, you ask, what is the value of posting such data? Well, there are at least two advantages, both of which accrue to the respective teams. First, it would build a stronger sense of community and belonging between teams and their fans. Almost every F1 fan has his or her favorite team, and many of these fans would just love to play around with some of "their" team's data. Second, given the large number of highly skilled and knowledgable fans, someone every so often just might find something that could prove to be helpful and important. As we all know, some of the world's greatest discoveries were made by obscure people who scrutinized seemingly meaningless things that numerous other people had "looked at" but had not "seen." Granted, these discoveries won't be of a timely nature, but they may help in long term development. And, which fan wouldn't like being responsible for helping his or her team?

In order for publically available data to be of value, each team will not only have to post data on-line, but receive and respond to messages from fans as well. This may seem like a lot to ask, but all it will really involve is for one or more team members to act as laison, reviewing fans' analyses and passing the more insightful ones on to the engineers and drivers, who will use and reject as they see fit. The laison person(s) should acknowledge each analysis whether or not it is used. Doing so would not be all that expensive or time-consuming, and it would certainly be a good public relations move--something the sponsors will appreciate. Whatever the cost of making data public, it will more than be offset by the benefits.


Bill Doolittle
Send comments to:dolitl@mail.utexas.edu