Use Numbers, Not Emotions, to Select Your Next Autocross Car

Andy
By Andy Hollis
Sep 25, 2020 | Autocross | Posted in Buyer's Guides | From the Feb. 2010 issue | Never miss an article

Photography credit: photosbyjuha.com

[Editor's note: This article originally ran in the February 2010 issue of Grassroots Motorsports]

When most people take up the sport of autocross, it’s merely an enthusiast activity that makes use of the car out in their driveway. As the racing bug bites harder, they spend a little time and money to further prep that car for whatever class it naturally falls into.

But with so many car models out there divided into so few classes, the odds of already owning the optimal model for any one class are pretty slim. For many people, their love for their cars is greater than their competitive aspirations, and they go on having fun regardless. Others, though, are willing to swap out their machines to move up the ranks.

So, how do you select the best car for any given class? One popular method is to simply buy the same make and model that’s been winning national events in the past few years. Those guys at the top must know what they’re doing, right?

This thought process leads to what is affectionately known as the lemming phenomenon. Need evidence? Just look at any national championship grid that sports rows of the same make and model. While not particularly scientific, buying the winning car isn’t a bad approach for people who prefer to be judged on driving skill alone. After all, it ensures that you’ll never be at a disadvantage due to car choice.

The lemming method has another benefit: groupthink on setup. With so many people running the same car, a large sample of testing data is available. Plus, autocrossers are usually happy to share their setups and ideas. Internet forums serve as think tanks, and results from all over the country are uploaded instantly for the benefit of all. If you’re the only one in your class running a particular car, however, you’re on your own; it may take you years to figure out what a class community could nail down in a couple of months.

However, all of this goes out the window when new car models are classed, rules are changed via class restructuring, or a completely new class comes along. Then what do you do?

Facing Change

Examples of these situations are happening right now, as the SCCA Solo Events Board has just completed a reorganization involving four classes: Super Stock, A Stock, B Stock and C Stock. The existing B Stock and C Stock classes were combined into one single group. That made room for a new “tweener” class that encompasses the slower cars from Super Stock. Then the existing A Stock cars were simply moved to B Stock.

For the newly combined C Stock, there is plenty of data out there from the existing results to suggest that the top dogs from either the old B Stock or C Stock can win on any given day. After that, it just depends on the course.

But what about that tweener class? No one has developed and run those cars for years, and they are about as different from each other as can be. The standouts appear to be the non-Z06 C5 Corvette, third-gen Mazda RX-7 and Porsche Cayman S. Since these eligible options are so diverse, how do you choose which one to campaign?

An even harder choice is currently being made by folks wishing to compete in another brand-new class, Street Touring Roadster. This one takes the highly popular Street Touring formula and applies it to a number of higher-powered, better-handling sports cars like the Honda S2000 and Mazda MX-5. Other legal cars include the Toyota MR2 Spyder and Nissan 350Z plus the BMW Z3 and Z4. 

Unlike the Stock category, where factory performance can be a good indicator of autocross potential, Street Touring allows bolt-on mods that can affect different cars in different ways. The result: A factory dud can sometimes be turned into an autocross rocket. The 1989 Honda Civic Si is a great example. It fell out of contention in Solo’s slowest Stock class, but the modifications allowed in the Street Touring ranks have made it pretty much unbeatable.

Two Approaches

There are two basic approaches to the selection process, and ideally drivers should employ both. Getting the same answer after following each thought pattern is a good sign that the car is a strong contender. As a comparison, these two methods are not unlike how valuations are done in the real estate business.

The first method—we’ll call it the empirical method—simply compares cars that are already prepped to similar standards. For example, to develop a car for the new STR class, it’s often a good starting point to look at results from recent Stock or Street Prepared contests. This data can then be altered when the allowable modifications are figured in. 

This approach has the benefit of using actual developed cars in a real competition environment. It’s also relatively quick and easy to do.

Let’s call the second method the theoretical approach. This one’s a bit more involved: Build the best example of each car on paper and compare the field on several key points. The devil is in the details, though, and the paper model won’t hold up to reality if the comparisons aren’t performed accurately.

Practical Examples

Using the empirical approach for the new Street Touring R class, we see that in Stock form the Honda S2000 is a front-running car for A Stock, while the MX-5 is a leader for the slower C Stock class. The Toyota MR2 Spyder is only mid-pack in C Stock. This same progression plays out in the Street Prepared ranks, where the S2000 is a mid-packer in B Street Prepared, while the MX-5 and MR2 Spyder run up front in the slightly slower C Street Prepared category. 

In Stock and Street Prepared, though, the extra weight of the S2000 is offset by its ability to put its extra power down through larger wheels and tires. Take those away with the Street Touring R tire width limit and you have a whole new ball game. The same thing happens to the 350Z and BMW roadsters, but to an even more significant degree. Seems like there are no clear winners—yet.

Now it’s time to apply the theoretical approach. Generally speaking, our first step is to identify the best year, model and option package combination for each car in question. 

This is pretty straightforward, as the sporty model is typically the golden ticket. These cars have the best suspensions, make the most power, and eschew many of the heavier luxury items found on the upscale versions. Also, the first year of a model generation is typically the lightest. Manufacturers tend to bloat out their cars in succeeding years, making them “better” by adding more gizmos that have nothing to do with performance. 

Wheel width is a critical factor when selecting an autocross car. Wider is almost always better, but the rules in many classes—including Street Touring—limit the maximum width of a wheel-and-tire package. Photography credit: Alex Groves

Occasionally, however, power gains, gearing changes and even special models pop up near the end of a run. Porsche is famous for this, and it drives the rule makers nuts.

As we do our virtual builds, we keep track of all relevant data in a spreadsheet for easy comparison. We’ll need the published specs for curb weight, engine power, torque and rev limits plus OE wheel and tire sizes. It’s also important to have the ratios for the final drive as well as the lower gears. Finally, we’ll also note things like alignment adjustment ranges and any tire clearance issues. (You can view the sheets below.)

Next, we “prepare” our cars. For machines in the Stock ranks, this prep is limited and its effect is fairly consistent across all cars. The easiest approach is simply to list the tire choices that will fit the OE wheels. 

For our Street Touring example, we’ll need to extrapolate further; this is where the art comes into play. For example, while each Street Touring class has a limit on tire and wheel width, not all cars can fit the max due to fender clearance. There’s a chance the hot tire will be unavailable in the optimal size, meaning a compromise would have to be made between width and gearing. 

We’ll also need to make some educated guesses as to how much power the allowed mods will add over the baselines. Not all of the legal modifications can be applied to every car, so careful research within the existing enthusiast community for the particular model can be beneficial. We’ll also look to see if fixes can be found for any basic OE deficiencies, such as alignment limitations. 

And finally, we’ll do some weight loss analysis. Does the car have a particularly heavy stock exhaust? Heavy power seats? Potentially significant savings should be noted in our build sheet.

Comparing Weapons

Now that we’ve done our virtual prep, we want to compare performance potential. The first data most people look at are the power-to-weight ratios, as carmakers have ingrained these numbers in our heads for years. 

More power equals better performance, right? Maybe at the drag strip, but not so much for autocross. Sure, it’s important data, but not nearly as much as handling characteristics. 

Unfortunately, handling is much harder to quantify. Magazine tests can be helpful, as they contain skidpad and slalom data on consistent courses, but variances in OE tire choices skew those results drastically. So, where to start?

A key predictor for autocross handling can be made by comparing weight to the wheel or tire width. Given an optimized suspension, it all comes down to how much rubber is on the road and how much load it has to support at each corner. 

Furthermore, the wheel width is a controlling factor in how much tire width can effectively be used. For street tires, here’s a good rule of thumb: Additional tire tread width provides gains up until it matches the wheel width. For example, anything more than 8 inches of tire tread on an 8-inch-wide wheel is mostly extra weight to rotate. That said, thanks to the super-stiff sidewalls of the R-compound tires allowed in the Stock classes, “overtiring” a rim can often provide a handling improvement, albeit with some loss of responsiveness and potentially a gearing hit. 

Class revisions and additions can bring once-unpopular cars to the fore. The 1989-'91 Honda Civic Si was a dud in Stock class competition, but these days Street Touring is jokingly referred to as Spec Civic. Photography credit: Alex Groves

For our Street Touring R spreadsheet, we’ll want a column that divides vehicle weight by tire tread width for all tire sizes under consideration. We normalized those numbers as a percentage of the best, then put it in a column labeled “grip factor.” 

Sometimes it pays to take this analysis a bit further. For example, how do you compare a front-wheel-drive car to a rear-drive car? Here, you’ll want to know the weight distribution of the vehicle so that you can compare tire loading on the heaviest end of the car. After all, that will be the limiting factor in lateral grip, assuming equal tires front and rear. 

In the case of the Street Touring Sports class, for example, both the Honda CRX and Mazda Miata wear the same tire size. However, the Miata is a full 100 pounds heavier. Advantage goes to the CRX, right? Not really. The Miata has a 50/50 weight distribution, while the CRX has 65 percent of its weight on the front tires—the same ones used for turning, braking and accelerating. Advantage: Miata.

Now that we have a decent read on basic handling, let’s return to forward motivation and add a column representing acceleration in our spreadsheet. For these figures, we’ll take the weight of the vehicle and divide it by the actual torque applied to the wheels on a typical autocross course. 

The latter piece of data is computed by multiplying the engine torque by the second gear ratio and final drive before dividing it by the tire diameter. These numbers have also been normalized to a scale of 100 and listed under the “thrust factor” heading. We also want to compute how fast this combination goes at redline—that info goes under the “speed” heading—as short gearing for better thrust may require too much shifting to be effective.

Another tip we’ve learned through the years: Slalom performance is primarily related to car width. Even street tires can spike amazing lateral g-loads, and putting them on a narrow Civic will have it outrunning a portly R-compound-equipped Z06 Corvette through a steady-state slalom. 

Wheelbase also plays a factor here, as a longer distance between the front and rear wheels can add stability in big offsets or fast 90-degree turns. Too much wheelbase, however, can also make the car harder to rotate in a slalom or sweeper.

A column should be reserved for notes on special circumstances. This is especially important in the Stock classes, but some of the items that cannot be fixed under the Street Touring rules should be included. Issues include limited camber adjustment, major toe or bumpsteer issues, geometry limitations on lowering, fender clearances, weird gearing, low rev limits and even parts failure issues.

Like what you're reading? We rely on your financial support. For as little as $3, you can support Grassroots Motorsports by becoming a Patron today. 

Join Free Join our community to easily find more Autocross articles.
Comments
View comments on the GRM forums
bentwrench
bentwrench SuperDork
9/25/20 9:43 a.m.

404s on the rest of the story and original

Colin Wood
Colin Wood Associate Editor
9/25/20 11:17 a.m.

In reply to bentwrench :

Looks like some wires got crossed on the back end, it should be fixed now.

nderwater
nderwater UltimaDork
9/25/20 11:23 a.m.

This is a solid article.  It's ten years later now but I bet Andy is still maintaining his spreadsheet.  I'd be curious to hear what his top picks are these days, based on the latest crop of cars and the current rule set.

maschinenbau (I live here)
maschinenbau (I live here) GRM+ Memberand SuperDork
9/25/20 11:29 a.m.

No

RadBarchetta
RadBarchetta New Reader
9/25/20 12:18 p.m.

First time I've ever seen SCCA Solo described as having "so few classes"!

bobzilla
bobzilla MegaDork
9/25/20 12:25 p.m.

I did this once and ended up with a Swift GT in STS. It sucked. I went back to emotionally picking my cars and it's been much more fun. 

The swift, numerically, made sense. It was narrower, shorter, lighter and made similar power to the reigning car in the class the 89 Civic Si and could scream to 9500 rpms chipped. What you DIDN'T know from the numbers was the chassis is a wet noodle, the gearbox has a mountain between 1-2, the suspension is terrible and you can barely fit a 195/50/15 under it. But damn did that G13B make people's ears bleed. 

Jordan Rimpela (FS)
Jordan Rimpela (FS) GRM+ Memberand Dork
9/25/20 12:33 p.m.

In reply to RadBarchetta :

Original story is from 2010, so grain of salt and whatnot. 

 

 

dps214
dps214 HalfDork
9/25/20 1:15 p.m.
bobzilla said:

I did this once and ended up with a Swift GT in STS. It sucked. I went back to emotionally picking my cars and it's been much more fun. 

The swift, numerically, made sense. It was narrower, shorter, lighter and made similar power to the reigning car in the class the 89 Civic Si and could scream to 9500 rpms chipped. What you DIDN'T know from the numbers was the chassis is a wet noodle, the gearbox has a mountain between 1-2, the suspension is terrible and you can barely fit a 195/50/15 under it. But damn did that G13B make people's ears bleed. 

I mean...the numbers should have told you some of those things. Gearing is almost more important than actual power/torque numbers for autocross, and area under the curve is more important than peak numbers for that matter. If you're not looking into that, you're doing it wrong. And if that info isn't available, it's a good sign that the chassis isn't well supported enough to make a good autocross car out of. Somewhere around here I still have the thrust calculation spreadsheet from when I was trying to numerically justify my emotional desire to run a STR Z3 and gaming out which model was the best starting point (luckily the ND was added shortly after I started looking into it and killed all hope of it ever being competitive and brought me to my senses).

RadBarchetta
RadBarchetta New Reader
9/25/20 1:18 p.m.

In reply to Jordan Rimpela (FS) :

I know. There are more classes now then there were then, but not by a large margin. There were still a lot then, too.

slowbird
slowbird SuperDork
9/25/20 1:40 p.m.

But autocrossing is more fun when I actually like the car I'm driving! Even if that's a Lincoln Town Car, flopping all over the cones! laugh

bobzilla
bobzilla MegaDork
9/25/20 1:58 p.m.
dps214 said:
bobzilla said:

I did this once and ended up with a Swift GT in STS. It sucked. I went back to emotionally picking my cars and it's been much more fun. 

The swift, numerically, made sense. It was narrower, shorter, lighter and made similar power to the reigning car in the class the 89 Civic Si and could scream to 9500 rpms chipped. What you DIDN'T know from the numbers was the chassis is a wet noodle, the gearbox has a mountain between 1-2, the suspension is terrible and you can barely fit a 195/50/15 under it. But damn did that G13B make people's ears bleed. 

I mean...the numbers should have told you some of those things. Gearing is almost more important than actual power/torque numbers for autocross, and area under the curve is more important than peak numbers for that matter. If you're not looking into that, you're doing it wrong. And if that info isn't available, it's a good sign that the chassis isn't well supported enough to make a good autocross car out of. Somewhere around here I still have the thrust calculation spreadsheet from when I was trying to numerically justify my emotional desire to run a STR Z3 and gaming out which model was the best starting point (luckily the ND was added shortly after I started looking into it and killed all hope of it ever being competitive and brought me to my senses).

there's just so many things though that you can't quantify. The steering rack for instance was fine for a road course, but tight slaloms it was just a little long and required a bit more angle than you could get out of your arms. So that benefit of the narrow track was lost in the slower steering rack. But it wasn't much, but just enough. Some slaloms it was fine. Others it wasn't. 2nd gear on it was great for autox. 35-65mph in the power, problem was 1st ran out at 20mph and you had this lag to get back into the power from a dig. Another issue was always wheelspin. Because it was so light (mine weighed in at 1770lbs with a full 9 gallons of fuel) inside wheelspin was hard to manage. An LSD would have made that car so much better, but wasn't (isn't) legal for ST(S). 

bobzilla
bobzilla MegaDork
9/25/20 1:59 p.m.
slowbird said:

But autocrossing is more fun when I actually like the car I'm driving! Even if that's a Lincoln Town Car, flopping all over the cones! laugh

I brought a farm truck to the second Peru Cam Challenge.... 

Ranger50
Ranger50 UltimaDork
9/25/20 2:07 p.m.

Pffft. Trucks suck at dodging cones. Do I care? Nope. I'll still try.

frenchyd
frenchyd GRM+ Memberand PowerDork
9/25/20 2:39 p.m.
bobzilla said:
slowbird said:

But autocrossing is more fun when I actually like the car I'm driving! Even if that's a Lincoln Town Car, flopping all over the cones! laugh

I brought a farm truck to the second Peru Cam Challenge.... 

Oops I forgot to preface the following statement with my Thought that yes race the car you love.  Emotions win because the effort to reward ratio is way off. No way can you justify logically the amount of effort to the level of reward. 

I really don't understand the whole concept of autocross. Probably because I don't like accounting. 
auto cross is technically exact.  Same car same track same repeat and repeat. Measure to the nanosecond. Upgrade your car to go faster and it will go faster.  
     Wheel to wheel is more about competition, man to man. ( OK person to person)    It's possible to draft a faster car and then under braking place yourself in a position to take his line away and finish ahead of a faster car. 
Yes it's more like chess than checkers. Yes it's often the newest most expensive car wins. But just often enough the driver makes the difference. 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
9/25/20 3:07 p.m.

The course is not the same track time after time, Frenchy. It's different at every event, unlike a racetrack. So there's a learning component to it. Watch a top level autocrosser and you'll see them exploring the course and making mistakes to see what works before laying down a fast run.

Autocrossers tend to learn how to drive on the track pretty quickly. It doesn't always go the other way.

frenchyd
frenchyd GRM+ Memberand PowerDork
9/25/20 5:31 p.m.

In reply to Keith Tanner :

Fair enough. Your point is valid.  But my "feelings" are that a person only competes with himself. He/she tries to put their best lap in and does so without outside interference. 
 There is no one on one.  Sort of the difference between playing basketball with yourself and playing against a competitor. 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
9/25/20 5:42 p.m.

There are a lot of sports where it's just you against the clock or some other measurement. The majority of the sports at the Olympics, for example. Or, if you like internal combustion - hillclimb, rally, drag racing, gymkhana, time trial, desert racing, Ultra4 racing or qualifying for a race :) The fact that nobody else can influence your performance doesn't make it any less valid. If anything, it's more pure. It's nothing but one on one.

I've noticed that motorsports where there isn't the chance for one competitor to interfere with another are a lot more congenial in the pits. You won't see a fist fight in a rally service area (well, maybe in Ireland on general principles) because it's the competitors vs the clock. In fact, competitors will help each other because they want to win fair and square. It's much happier environment.

Meanwhile, my 3 year old nephew idolizes Chick Hicks in Cars because he wins by crashing out all the other cars :)

Sidewayze
Sidewayze New Reader
9/25/20 6:17 p.m.

In reply to frenchyd :

Actually, when you are at an AX and in a battle for 10th's of a second with someone from run to run, it's actually really intense.

alfadriver (Forum Supporter)
alfadriver (Forum Supporter) MegaDork
9/25/20 7:42 p.m.

In reply to frenchyd :

You've made your point many times over the years. Why repeat yourself?  Just accept that autocrossing is popular, and move on. 
 

As for the original article, I've always brought up the point of copying the fast cars, as it saves a lot of time, efforts, and excuses. But we are also some who would rather race a car we love than one that is meaningless. 

frenchyd
frenchyd GRM+ Memberand PowerDork
9/25/20 9:23 p.m.
Keith Tanner said:

There are a lot of sports where it's just you against the clock or some other measurement. The majority of the sports at the Olympics, for example. Or, if you like internal combustion - hillclimb, rally, drag racing, gymkhana, time trial, desert racing, Ultra4 racing or qualifying for a race :) The fact that nobody else can influence your performance doesn't make it any less valid. If anything, it's more pure. It's nothing but one on one.

I've noticed that motorsports where there isn't the chance for one competitor to interfere with another are a lot more congenial in the pits. You won't see a fist fight in a rally service area (well, maybe in Ireland on general principles) because it's the competitors vs the clock. In fact, competitors will help each other because they want to win fair and square. It's much happier environment.

Meanwhile, my 3 year old nephew idolizes Chick Hicks in Cars because he wins by crashing out all the other cars :)

Your argument is well reasoned  and intelligently discussed.  
I grew up in the era of blood sports. Among the rewards of winning was remaining alive.  I started racing during that period. We were losing a major racer every week.  As enforced safety saved lives  attendance grew. Justified by increased safety reduced loss  attendance increased. Furthering greater attention to safety etc 

As racing grew safer and safer  and more amateurs   began participating.  Leading to safe and safer versions of racing. Hence the growth of Autocrossing. 

irish44j (Forum Supporter)
irish44j (Forum Supporter) MegaDork
9/25/20 10:42 p.m.

Motorsports would be super boring if everybody just drove the same car because it is "the ultimate" one for any given class or event. What makes motorsports fun are the crazy builds of people making something they love that may or may not win but will always be original.

I'd rather lose every motor sport event in a car I actually like and enjoy driving, rather than dump a bunch of money into a car that I may not really like just so I can get a plastic trophy (since I doubt there are very many people on here who are actually making a living winning motorsport events)

 

 

300zxfreak
300zxfreak Reader
9/26/20 1:57 p.m.

In reply to Keith Tanner :

Keith, I believe that fist fights are required in Irish rallying. 

irish44j (Forum Supporter)
irish44j (Forum Supporter) MegaDork
9/26/20 3:29 p.m.

In reply to 300zxfreak :

Ironically of all the motorsports I've participated in over the years, including tons of autocross and road racing, I find that the people at rally are by far the friendliest and least likely to get up in arms. 

The one exception to that rule: If Ryan Millen is competing there always seems to be a spat between him and one of the other top teams. The Irish guys tend to stay in their own little group. But almost all of them are at the front of the pack so their group is nowhere near mine lol.

I can't speak for Rallies actually in Ireland of course :)

Cooter
Cooter UberDork
9/26/20 5:45 p.m.

If everyone used numbers instead of emotions in motorsports, no one would be in motorsports.

spacecadet (Forum Supporter)
spacecadet (Forum Supporter) GRM+ Memberand SuperDork
9/26/20 6:37 p.m.

There's a lot of folks commenting here who fall outside the train of thought as someone like Andy...... or many of the other national champion and pointy end nationally competitive autocrossers I've met and competed with(mostly gotten destroyed by) over the past 5 years.

if you just wanna go out and do your best (as you see it) on that given day and have fun driving and seeing where the chips fall.. then absolutely let emotion drive your choice.

But those folks are a lot more rare at the pointiest end of the field for autocross, which is obviously where Andy is. 

if you want to develop your driving and car control in an autocross environment efficiently, and compete to win in SCCA national autocross.. I personally believe you have to think this way.

One of the best things I did to help myself really get better as an autocrosser was seat time.. but not only seat time and attending events.. but making a conscious choice to put myself in a car that was properly classed and can contend(aka properly prepped) well in the class when i'm getting that seat time. Then from event to event.. looking at how I was or wasn't improving using PAX as a tool... You can absolutely do this if you run the same car from event to event even if it's not the right car.. but the ceiling for your potential will be unknown since the prep level or potential speed of the car you're running will not properly be represented by PAX. The effectiveness of this is also directly affected by the depth of talent in your region.

that was a bit of a side tangent to the article.. but this way of thinking also helped me consult with my buddy back in 2018.. and get myself a codrive in the STH Audi TT last year at solo nats and I got my first nats trophy thanks to that opportunity..


 

Toebra
Toebra Dork
9/26/20 7:53 p.m.
300zxfreak said:

In reply to Keith Tanner :

Keith, I believe that fist fights are required in Irish rallying. 

and whiskey, I hear tell that if not for whiskey, the Irish would rule the world.

 

Oh, and basic principles do not change.  I like the way that Mr Hollis lays stuff out, so even I can understand.  I think the method by which he approaches a conundrum is what I find most instructive.

Our Preferred Partners
an3T4azLVMDYcgKGWEb2qpvmKWM3kf7nl0dAfaYnDaEGR6YbGEV3REgKBrPyIPv7