Race Note Taking - Read, Record & Learn

Second Of Two Parts

In the first installment we spent a good deal of time discussing the reasons for taking certain measurements, how to take these measurements, and what measurements you should take notes on in order to document your race day activities. That is the first step in a three-step approach to developing better documentation of your racing program. The other two steps are shop notes and the analysis of the race day notes.

The goal is to understand what variables really drive the performance of your race car. The note-taking process helps define and separate the significant few from the insignificant many. Without notes, you are depending on memory to tune the car. This is not always a completely bad thing, but it is far easier to make very costly errors when your racing process is dependent on memory for the analysis of data used to adjust the car.

Developing notes from a shop perspective better prepares your team to establish baseline setups for the car as you go forward. Establishing a baseline setup is critical. You need to understand the car's status at the start of the race so you can track the changes. Not having a baseline setting is much like going on a road trip without a map to a place you have never visited. You have to have a point of origin. If you do not know where the journey begins, how can you get to where you want to go, or know when you get there? Race cars are no different; knowing where you are starting from a setup perspective facilitates decision making when it comes to tuning. If you race at multiple tracks, you may find the baseline setup needs to be different from track to track. Keeping shop notes makes you a better prepared racer.

Keeping detailed notes on your shop activities helps you keep track of maintenance activities as well. In addition to shop notes, it is a good idea to develop a checklist of standard activities that take place between races. This checklist is used to define the maintenance process. Some activities take place between each race, and other activities take place on a more infrequent basis. Routine maintenance on the car needs to take place in the shop. The racetrack is not the place to discover damaged brake lines or a cracked shock mount. Keeping shop notes enables you to spend your day at the races tuning for speed, not working on the car just to make it run or repairing damaged or broken parts.

The third step in the process is using your notes to look for trends in the data. This is a critical step to really understanding if your tuning activity is working. What you learn this week or two weeks earlier may be a key to winning this week. Let's spend a little time talking about data and how it's presented.

When we are tuning a race car, we use data to answer questions. Therefore, it is necessary to employ critical thinking and ask the right questions. As we use data to answer our questions, we need to understand the forms in which it is presented. Data comes in two basic forms: continuous and attribute. Continuous data is expressed as a number. Lap times are an example of continuous data. Attribute data is what its name implies-an attribute, such as good, bad, fast, or slow.

Attribute data has a place in racing, but it is much more subjective. Racers have a bad habit of taking a string of continuous data and turning it into an attribute data set. For example, some racers look at a string of lap times and arbitrarily say it was a bad set of laps. The problem is the opportunity to review this data is lost if the racer just arbitrarily calls the whole data set bad. It may not be a bad set of laps for driver "A," but for driver "B" it may be a stellar set of laps. We need to look at the whole picture and the context in which the data was gathered.

Attribute data is not without merit; we just need to convert it into something we can use. We can develop a Likert scale in which we take attribute data and convert it into a more useable continuous data set. Using our good and bad data set example, we can assign a number to the data set with 1 being the best and 5 being the worst. A set of laps may represent 4 on our Likert scale. While not a completely qualitative measure, it gives us a quantitative number to use for comparison purposes. Try not to arbitrarily throw out data. The generation of data is very expensive, so try to use all the data you gather.

The analysis of data can be defined in three ways: practical, graphical, and analytical. If the analysis makes sense, and the answer is practical, and you understand the answer, you are probably on the right path. If you have graphical analysis of the data (i.e., the numbers arranged into graphs and charts, allowing you to better visualize what the data is saying), then you are on the right path. At the analytical step, if it is necessary to do some "icky math" and/or use esoteric statistical modeling tools, we may be asking the wrong questions or we may be collecting the wrong data. In fact, resorting to complex statistical analysis to answer questions is very rare. We must not overcomplicate the data.

Another thing we shouldn't do is drive the data toward any preconceived notion or position. If the data does not say what you want it to, don't keep reforming or reformatting the data in an attempt to make it support your position. If you just spent a bucket of money on a new product, and the car isn't going faster, don't blame the data. Keep the data pure and unbiased; otherwise, the exercise is worthless.

Data analysis is not a complex process, nor does it require advanced computer skills or expensive software. You do not have to have a computer at all. Just the use of pen and paper gets you through 90 percent of the analysis. That said, it is very handy to have a computer and a copy of Excel to help develop graphs faster than you would by hand. If you do not have the computer skills, then graph paper, a pencil, and a ruler can go a long way toward helping you reach your goals.