The Process
In this article, I'm going to outline a simple, easy-to-use process that I often use to approach my setup and racing line development. The beauty of this process is in the structure; you cut out the "circle logic", chasing your tail, and the confusing results you get when you chase a setup instead of chasing your approach as a driver. Being able to tell whether your driving the chassis incorrectly or you're driving a chassis properly that is set up incorrectly is one of the vital skills you must learn in order to produce a setup that works in multiple places, time and again, for a variety of people. In this team, we have so many people capable of running a properly tuned car to a great result, but in order to capitalize on this opportunity you must also have the properly tuned car. In addition, you cannot waste time chasing a setup back and forth making changes at random. This is time that could be used honing the racing line, working on finding chunks of time here and there, and improving your race craft. The complete package ultimately produces the best result. Dividing these different areas of your race performance up will help to isolate issues with the car and driver, and this article should help solve the problems you come across when dealing with a flaw in the driver or the chassis, or both!
Telemetry is a wonderful tool. Not only can you nerd out over the gathered data (Who doesn't love to watch tire temps rise and fall, compare steering angle through a hairpin from one lap to the next, look at how another driver solves the racing-line on a circuit, and all the other fantastic things you can investigate through telemetry?), you can actually make it useful... Imagine that. From determining a possible setup adjustment to finding out where you lose grip and what the hell to do about it, telemetry will inevitably play in important role in improving your game in the sim world and real life should you choose to make the step from Sunday Stroller to Weekend Warrior.
This is an in-depth analysis of my "method" when I attack a new track. And as a word of warning and encouragement. I have drawn this process out and explained it as intricately as possible, for your benefit as well as mine. The process will be followed by an example. I will take a car and a track and go through the entire process. Since I wrote this article initially, I've lost the graphics that went with the original data. In order to keep with the times at BAM, I am now updating this article and should end up with the same result; a big improvement in both the driver performance, and the setup.
Step One - Benchmarking
The concept is simple - you can't tell how far you've come unless you know where you started. I often find myself jumping on track, running about two laps, and then tweaking shit on the car, in the line, how I drive, blah blah blah, ALL AT ONCE and it really showed. What I consider "consistent" could vary so much from session to session, day to day, lap to lap that it was hard to tell whether I was getting faster or slower. More often than not I actually felt like I was slowly getting worse!
The first step in my method is to develop a benchmark. Load up a basic setup. Read: Baseline or the go-to standardized settings. Some cars automatically like a few pounds of pressure let out of the tires, or the brake bias moved forward a fuzz, etc, and those inevitable adjustments are the only ones that should take place before you benchmark. Making those adjustments halfway into the process will clearly drop lap times (probably enough to look into new gear selections for corners and substantial modification of your racing line) and basically put you back to square one of the process. With your basic, no-nonsense setup, go out and run a few sighting laps. Don't push the car; not a squawk from the tires. Seriously. Two or three laps idling around the track sounds lame and excessive; you're a great driver, it doesn't apply to you, and you can just skip this step and blah blah blah NOPE. DO NOT. END THAT SHIT. Anyway... on your sighter laps, just feel out the line and use some powerful observation. The more you know about the subtleties of the track, the further your benchmark will get you. Notice braking zones and where a conservative braking marker may be. Categorize your corners from the Racer's Bible guides. What kind of radius? Is it a Type 1, 2, or 3? Observe some possible late turn-in points. Make note of camber changes, bumps, dips, changes in these characteristics if you use a different line (think Sunset bend at Sebring). While you swing through the corner your second time around, mock up a conservative late apex point. Conservative, but not so conservative as to outright kill your pace through the corner. It should be safe, and leave you with run off room for your track out point before you get to the edge of the available track surface but it shouldn't end up with you exiting the corner like your terrified of your own existence. This gives you room to build on a little at a time.
With these sighter laps complete, you should have a good picture of what to expect when you up the pace. Transfer your mentality from running sighter laps to brisk pacing laps. Don't crack open a can of kick ass and run a balls-out qualifying session. The key here is to build up slow. Take this with the same mentality as the Racer's Bible. Before you build on the the sighter pace, be sure you know what to expect; what you expect to happen when you push a little harder. Then pick up the pace a little bit and see how you did choosing your three points for the racing line through each corner. Again, we're doing this right the first time; spend a few of these brisk pace laps hitting the marks you set in your sighter laps. If they're wrong, hit them anyway. Analyze why you picked wrong. Jot it down in your head that you could actually turn way later in T3 than you thought you could but ended up setting your turn in point for T5 way too late; because T3 sort of swoops into the apex with a very helpful camber change and you expected to hit an apex for T5 during your sighters that was just too late and you end up squaring the turn off and not carrying near as much speed through it as you could with a more optimal line. Make a mental note of the changes you should make to this line, and then one corner at a time make those changes. Do not try and change them again; if your new points are worse than your original points, go back to the originals and that's that. Towards the end of this process, you should have as many laps as you wish in for seat time. I've been trying to get a good 10-12 laps in total after I've tried out one round of changes, depending on the track. One thing I will say: Running too many sighter and pace laps will never hurt. Not running enough, not so much. You should note that at this point we're not concerned about logging data, looking at sector times or lap speed, or making an actual measurement of performance. We're using some common sense here and just seeing what works, and what doesn't.
Once you have those figured out, get off the track. Open up your note book. I'm dead serious; if you don't have one, buy one. Open up a word document. I don't care how you do it (I don't care THAT you do it, either. Or any of this. Because if you don't and I do, I win lol), just do it, and you can thank me later. Back on topic... Write down your Benchmark.
For EVERY corner of the circuit, even if it's just a kink you take flat out:
- Assess your corner with the type discriptions from article #2 of the Racer's Bible: I.E. Increasing Radius Type 2
- Write down the three points of your racing line for the corner and include the brake point; I.E. Brake Marker, Turn-in, Apex, and Track-out. Four points in total. Be psychotically specific here. You may find yourself changing the points of your racing line through a corner in inches, to gain tenths. Writing down a very descriptive explanation will help you be very consistent.
- A solid summary of your thoughts on the corner is a must. Note how you brake, if it's just a quick brush or a lift of the throttle, or if you threshold brake. Add in where you return to throttle, how, and why you decided on that. Any ideas on how the car feels? Does it push mid corner? Is there a bump towards track out you should consider avoiding? How do you feel you're doing in this one? Think you're getting through it pretty well or do you know already that this is a corner you can make a lot of improvement in? Make note of the corner's importance. Is this a good place to pass/allow a car through from a different class? Why? Does it lead onto the longest straight, i.e. is this a corner where you may stand to gain the most advantageous exit or shed the most time? Is this an area of the track you can go two wide through and not lose much time, or do you have to be single file? How would you let a faster class through at this section if forced? How would you overtake a slower class or car through here, or would you 100% back off and follow them through?
Do that for one corner, and then move onto the next until you've written down your entire Benchmark for the circuit.
Now, this is the part where you can put the hammer down and go for it. But this is important; you're not trying to improve on these initial points. You're not going to change your racing line at all. What you're going to do is hot lap. This is where it is imperative that you can demonstrate your understanding of the basics outlined in the Racer's Bible. Spend time getting the line you have chosen down perfectly. You haven't gotten your baseline down until you can turn laps that are identical. Analyzing data from your run should allow you to compare one lap to the next more efficiently. Most people look at the lap time; that's it. Their sole comparison of the lap. "I was a second slower on the second one, then I picked up seven tenths, then I spun out on that one, then I picked up another four tenths, and then I was three tenths slower." No. That's not a useful metric. At least watch the replay and make a mental note of why things happened. Use data analysis software such as Motec, iSpeed, even VRS etc. Save your replays. You can definitely write down some things while you're becoming more consistent in your ability to hit your marks. If you didn't hit your marks, and you picked up two tenths in that corner, then you've just saved yourself some guesswork and you have something to try for when you make the move to improving on your benchmark!
If, by chance, you find in an area of the track that these initial points are causing you problems and adjusting them slightly will allow you to hit your marks more consistently, it is okay to do so. Use some common sense. While the emphasis here is trying to get a repeatable result, it will be much easier to do so with a repeatable line. A novice driver will likely not have to worry about this much, as the consistently alone in this step should provide for quite the undertaking. As you progress in your skill and proficiency, you will be able to string longer sessions together and immediately pick up on trouble spots and adjust them instinctively; this is a good skill, and I do not want to deny it. But, make note of it and try and think of it like this: Fix the problem spots, but don't try and improve on areas that aren't providing you difficulty in repeatability. The improvements will come in at Step 2.
You should end up on a plateau; that's when you know you've succeeded in developing an accurate benchmark. You can hit your marks within inches every time. Breaking down each lap (via splits on iSpeed, individual turns and straights on MoTeC, etc.) should reveal identical numbers of each piece of track from lap to lap to lap to lap. At this point, your plateau is the point at which you have used all the tools available to run your racing line absolutely flawlessly, every time. And your lap times should show this. As you gain consistency in the execution and performance your Benchmark line, it will reflect upon a theoretical maximum potential and you have your plateau. A consistent, fairly solid "Optimal time". The optimal time in iRacing, iSpeed whatever you use may not be extremely close to this time, but it should be close if you performed a clean and consistent execution of your racing line, working from those slow pace laps to your eventual current optimal speed and likewise lap time. Write this Optimal Time down: Mind you, not the one that it says on iRacing. The optimal time here being the average consistent lap time at which you plateau. It's even better to go to further lengths of detail here. Try and reach these consistent "working optimum" times not only in your overall lap time, but your sector times and even the time spent in each corner and straight (Using my MoTeC project might be the best way to get down to this level of commitment, and the most profitable). If you haven't gotten to the point where you are executing the line just right and hitting lap times that are within a few tenths of each other with sector times being considerable closer, you're not doing it right. Go back a few paragraphs. Otherwise, yes. It's time to advance from this benchmark to developing a better line. It's worth noting here that if you're strugging with running a fairly good pace but your iSpeed or iRacing optimal is considerably faster, this shows an inconsistency between your line and something you've done in the past in a particular part of the track. If you haven't been using the steps in this article and are often in this situation, it's a good indicator that you've pushed your limits very early in your own process, and have lost these accidental improvements to your line without taking note of HOW you gained that time in a particular sector or on a particular lap. The process outlined here takes advantage of slow incremental steps and constant observation and analysis to spot these areas out, find out what changed, and use them to add to your solution of the racing line. That will bring your "working optimal" consistent times much closer to your measured iSpeed or iRacing optimal time.
Step 2 - Optimize Your Line
You've developed a consistent benchmark for your performance in execution of a proper racing line. While this might not be the best one, it's a start. In this step, you can start moving your markers around to see if find some time.
The key to this step is to know what to focus on. After benchmarking, you should possess two important observations. Firstly, you should know what corners you should emphasize due to their importance; such as a corner that runs onto the longest straight, and especially if that straight runs into a corner whose braking zone provides the best opportunity to pass around that track! Having a more optimal line through that corner than your competitors will put you at a great advantage. It'll be easier to pass others and getting around you will be much more difficult. The second observation is using your understanding of the racing line to spot flaws in your benchmark, and what to do about them. Maybe you know by now that your apex and throttle tip-in point can be moved a little earlier for X corner, because you can tell that your tires at both ends of the car are below their potential through the exit of the corner and as a result, the track-out feels just too easy.
You can now start playing with your line. At no point have I talked about changing your setup. DO NOT. NOPE. Not in this step. That's another day's work, and it can only happen when you have reached the chassis' limit for performance through each and every corner, each and every lap (or damn close - we're not all aliens!). Do not apply all of your changes at once. Consider this; any time you change one of the points in your racing line through a turn, it will produce the butterfly effect. Ripples will be felt in the next few areas of track. Maybe your change means that you've picked up a little speed through the corner. Fantastic! But, because you chose to work on a Type 3 corner first, which is a corner that leads into another corner, you've screwed up your benchmark line through the section as a whole and you haven't picked up any time overall. It would have been a good idea in this example to remember that your first priorities are the corners that lead onto straights. Because the corner leading onto the straight can most likely be modified much easier than changing the first turn corner in the series, it is wise to start there. Work one corner at a time, and have a system (i.e. work on the corner that leads onto the longest straight first, second longest straight second, then any corners that are linked into the run up to those corners, then ideal passing zones, etc).
Approach this like you did before. Make a note of what you have chosen to change first. Then, make the change in testing, and become consistent in lapping the circuit with that change. Observe the effects through video replay and telemetry analysis. Did your minimum speed through the corner increase? Did you pull more g-load through the corner? Did your time through the corner increase instead of decrease? How does the car feel? Write it all down, and work slowly and methodically. Once you've assessed the performance of your change, keep making changes in the same manner for that corner until you feel as though you don't see room for improvement. Or, move on when you get the feeling that further analysis would only drop a negligible amount off your lap time, whereas moving onto the next corner will probably net you tenths. At that point it's okay to leave this corner how it is and improve the next section of track.
You could spend all the time in the world on this, and it will never be perfect. That's good. Nobody can run a lap they'll never be able to improve on. Not even the best in the business. At no point during the race weekend does Scott Pruett think "Nope, I did that lap perfectly, and a 1:29.432 is exactly how fast this car can get around the track. I'll just do that from now on". Nope. However! Eventually, you will keep working on corners and improving your line until you are throwing down some seriously quick times. And you should be doing so with much more consistency. Instead of hitting a PB lap that you'll never come within tenths of again and an optimal time from the data that's a second faster still from your PB, it will be quite the opposite. Your personal best should be a few tenths at most from your optimal lap, and every lap you put down rivals your personal best not only in overall time, but in comparison corner-by-corner. This is now your optimal racing line so far, and you should break out the notebook again and write it down in the same manner. By this time, you should have a lot written down in there. Notes from the creation of your initial racing line. Changes you made, and then your Benchmark racing line. Changes you made, and then your Optimal racing line. This should provide you with plenty of information for future use; be it Step three, learning a new track by comparing the qualities of the corners to the track you're on right now, learning a new car by being able to look up the racing line and data you have for a previous/similar car, etc.
It is at this point that you are free to make the jump from Step 2 to Step 3. And it's your choice when. Just remember that Step 3 must always wait until the absolute completion of Step 2, and that you can return to Step 2 for a reassessment to see if the setup changes you've been making have actually made a potential improvement to your driving performance.
Step 3 - Optimize Your Chassis
This is a process all on its own. And I have decided that I am going to write that process down separately, and at a later time. But in regards to going faster, there are a few things that must be included in this section and made clear.
First off, you can get carried away with setup changes. Everyone can. And getting ready for a race serious enough to warrant going through this process of going faster means that you need to plan far enough ahead of time to get that whole "getting carried away" thing out of your system. It happens. So, while it's not the usual guidance you'll find regarding how to set up a chassis, I highly suggest this (Especially if you're like me, and your fascination with vehicle setups can spill over into the situations where you ACTUALLY need to focus on improvement and not experimentation); go fuck with stuff. Honestly, it's the best first thing you can do. If you have ideas about what might be helpful, go frickin do it. Change your rollbars drastically, run heavy springs, run light springs, run lots of caster, run less camber, change your tire pressures, play with the damper settings, get it all out of your system, and as ALWAYS make note of what happened when you changed what you did. Once you've gotten that out of your system, on to the usual rant.
Setting up a chassis is complex. Look at it like this. You have basically two problems. Oversteer. Understeer. That is it. You have upwards of 50-75 areas you can make adjustments to the chassis. Within those areas of adjustment, you're given anywhere from 2 or 3, to another 100 individual settings; be it firm/soft/no rollbar, or 100 different settings of ride height at each corner of the car or 40 different choices of wing angle. You can see where I'm going with this. Two handling characteristics. Hundreds and fifties of millions of possible combinations. I suggest learning the setup AND racing line process on a car whose adjustments are few, simple, and straightforward. The Skip Barber, the Mustang, the MX5... great options here. At any rate, you must be able to assess what is causing the handling characteristic you are trying to improve. If you adjust something that makes sense to adjust but was not out of adjustment, you have now created another problem. It logically follows that you have to have a system in play for being able to tell if an adjustment you make actually improves the performance of the car more than it hinders it.
Start out with all of your adjustments at the same settings you benchmarked with, of course. This is where Step 3 builds on Step 2; you must have the understanding that you are now switching your train of thought from assessing your driving performance, to assessing your vehicle performance. In the same way that the vehicle performance remained constant up until this point, the driver performance must now remain constant. If you are still picking up tenths and changing your racing line while making setup adjustments, you will get put in situations where you made a very negative setup adjustment, but because you found out that you could brake later in three important turns and pick up a later apex, you netted three tenths faster lap times for the overall process and your goal of improving the car is now blown to bits.
My Step 3 revolves around the ABA technique. If you haven't heard of it, it's pretty simple. Actually, it's almost similar to our benchmarking process.
"A" is your benchmark performance. "B" stands for a new performance assessment with one adjustment made to the setup of the vehicle. You assess the performance of the car both by observation personally, and numerically (Maximum cornering force or Lateral G's, increased minimum speed in a corner sector, higher maximum speed on a straight achieved by the ability to brake later, sector/corner/straight/lap times, etc... Metrics! Measurables!). Then, you return to the previous "A" benchmark adjustments, and compare it again. You will find more often than you think that making an adjustment from "A" to "B" will feel faster AND produce a numerical improvement, and then when you move the adjustment back from "B" to "A" the numbers stay the same. In many cases, the small adjustment made a section of track easier to manage, or it made one of your driving duties simpler. An example of this would be: While trying to fix understeer into the corner, a change to the front bump/compression shock settings allowed you to brake a little later into a few of the corners on track. The setting didn't fix your understeer. But it did distract you from your problem of understeer by boosting your confidence under braking and netting you a tenth. However, on your return from "B" to "A" for your reassessment, this confidence under braking is carried over and you are still a tenth faster. In this manner, your chassis should really remain at setting "A"; the adjustment did not have any effect on your problem, and that means that there is still an area of the setup that is out of adjustment.
Performing setup changes in this manner will allow you to not only make notes of areas where you can change your Optimum racing line (At which point it is smart to hold off on the chassis adjustments, go back to your Optimum racing line procedure, and make note of that change and do another once-over for anywhere else you can apply those changes), but it will also allow you to hone in on the adjustments that are causing fundamental problems. This is the irony in vehicle setup and tuning; it is difficult and complex because it is easy and simple. The simplicity; You have a maximum of two problems, solve them. The difficulty is in our desire to fix what ain't broke.
Further expanding on this theory leads to a much larger and in-depth article on chassis setup. Once the fundamental problems are solved, you can begin working on the more intricate parts of the setup. How would you like the car to perform? How can you achieve more stability? How about bumps, is the suspension not soaking them up very well? How do you make that work better? These are all questions you can answer in the chassis setup, as well as driving the car differently (read; more optimally). Keep in mind what we said previously though. All of these "optional" goals can be broken down into your two simple problems yet again. Because the physics at play have total control over the situations, it is the key metric by which you even have your opinion for how the car should drive. And it follows that your opinions will most likely have a lot to do with oversteer and understeer, albeit under the surface.
Credit to Travis Mihm of Blackadder Motorsport for the article!
