I just want to say, I really appreciate your style of writing. It's very conversational and personal. Very enjoyable to read.
Photography Credit: David S. Wallens
A few years ago, we recommitted ourselves to properly testing new cars. This meant testing them on track, what with us being a track-focused media outlet and all.
Fortunately, we have an exceptional test facility nearby in the form of the Florida International Rally & Motorsport Park, whose partnership makes this part of the job realistic. Every relevant new car that we can test gets a proper track workout, complete with data capture and lap times from yours truly.
[The Grassroots Motorsports ultimate guide to track car lap times]
Because we’re pretty much the only media outlet doing consistent track testing of relevant new models, I get more than a few questions on how we test and what the results really mean. So I thought I’d dump some thoughts into this issue’s column.
1. Lap times are an interesting metric, but they can be deceiving. Because we test cars in as-delivered condition–manufacturers would rather we didn’t sneak some camber or toe into their machines–none of the cars we test can truly be considered “optimized” for track use. Add to this the fact that we test on OEM tires, probably the single biggest variable in producing lap times.
So while lap times are a great yardstick for message board bragging rights, they don’t tell the whole story. Instead of focusing on lap times, take a harder look at the data traces, which can provide insight into how the car actually drove. Did it put down power well out of corners or was I constantly adjusting speed post-apex? Did it trail-brake well into corners, or did the braking traces have inconsistent lower peaks? How much speed was lost on turn-in? These spots on the plot can give you more insight than raw numbers ever could.
2. No, I’m not “gaming” the hot laps, and the test is of the car, not me. Seasoned time trial competitors know that taking an extra-deep run into the final corner and building additional speed across the start/finish line is a great way to pick up some fractions of a second, but it doesn’t make for a fair comparison across months, years or different cars.
Also, I’m there to set a time for the car, not to shatter my personal records or to wad up some nice machinery and have to make a very awkward phone call. So my hot laps tend to be closer to 99% than 101%. That said, I’d like to think that my 99% is a pretty well-seasoned and skilled 99%, so the data you get is relevant.
But that’s another complication: With each test day, I gain more experience and knowledge of the FIRM and its tendencies. I’m certainly more comfortable now, after thousands of laps there, than I was on test lap 14. This isn’t something I can really account for in testing, other than to say that my mission with each car is to produce the best lap it’s capable of while still giving it back in good shape.
Looking over my historical data–find the link on our home page–I think I feel comfortable saying that the best lap times probably haven’t changed that much, but my ability to generate them in fewer total laps has. So when you look at older tests, you can assume those lap times and data charts are still pretty relevant.
3. Yes, it’s fun, but it’s still work. Our colleagues at another magazine who I won’t name (but if I were to, I’d say it was Road & Track) are charging like $250 to go hang out at one of their track test days. Now, I don’t know if our sales department is working on something similar right now, but honestly, if someone wants to pay anything more than $0 to watch me fill a balaclava with head sweat for 3 hours, they need to rethink their lives.
Yes, driving cool cars at the limit is fun, but my true track love is competition, and testing means I need to keep my headspace as far from a competition mindset as possible. The guy you might run into if you visit the FIRM on a Friday we happen to be there is definitely not the same guy you’ll bump into at a TT event. Yes, he’s still nice and will hand you a sticker and hang out and chat, but he’s nowhere near $250 worth of interesting.
4. Cars are soooooo good today. Sure, everything we drive can be criticized. Some heat the oil up a bit, some fade the brakes, but the overall percentage of sporty cars today capable of producing a satisfying track experience is astounding. And many of the “fast” cars from the OEMs are downright dialed, even before you start optimizing with alignment and better pads and tires. It’s a good time to be a car enthusiast.
And it’s a good time to set some good times. I’m loading up the trailer with our latest press loaner to do that as soon as I’m done with this column. If you’re nearby, stop by and watch me sweat.
I just want to say, I really appreciate your style of writing. It's very conversational and personal. Very enjoyable to read.
I'll second what Toyman! said.
Glad you mentioned OEM tires. A number of reviews of the MK7 and 7.5 GTi complained about the all-season tires VW had on the car that gave up way too early and spoiled a lot about what was a very good chassis. I agreed. The Bridgestone Potenza RE97 AS that came on my car had so little cornering grip -especially in the wet- that I researched them to see what was "wrong" with them.
Turns out, nothing was technically wrong with them, and the simple answer was that they were developed almost solely for delivering good gas mileage, which yeah, they delivered on.
When the MK8 GTi was delivered for reviews, many people remarked about the serious upgrade in tires this time around, a few outlets even using words like "cheater tires." The tires no doubt helped the various lap times tumble, probably more than the usual bump in horsepower and torque, and some chassis development...I have no doubt it all helped, but the big takeaway seemed to be the rubber that hit the road.
I appreciate the way y'all approach the topic of testing and the data logging, as it does provide interesting insight.
This post is long enough, but you mention a magazine that may or may not go by the initials of R&T. I remember reading that mag back in the early 1980's when ABS was a new thing and there was "enthusiast" pushback about safety nanny intervention, it interferes with driver involvement, and everyone was a Driving God who was simply better at braking than some artificial ABS system would ever be. (as an aside, aren't you glad enthusiasts got over those types of tired arguments long ago?
)
Anyway, to *prove* humans were better than computers, that magazine had two guys test a car with and without ABS. And yes, those two dudes were able to threshold brake about 2 ft. shorter than the car with ABS did.
That mag proved its point once and for all, case closed.
Oh, I'll just mention something even though it's likely of no importance, really, but those two dudes were named Paul Frere and Phil Hill. Just your usual "Everyman" drivers who happened to win a lot of sportscar, endurance, and F1 races, and stuff. But I doubt that impacted the test in any way. 
Excellent article, and we really do appreciate the method.
There are so many issues with a lot of outlets, from ringer drivers to ringer cars to ringer tires to rollout when measuring 0-60. We appreciate that you do none of those things.
My kids are discovering Top Gear for the first time. I hear something, and point out "Hey kiddo, you know that this is wildly entertaining, but all scripted, right?" and they are so surprised. That's Top Gear. Imagine discovering R&T for the first time and not knowing this.
When I was doing a round of magazine tests (I think it was for the Westfield), I was chatting with some of the major magazine editors about 0-60 testing. They all have different ways of interpreting the data, all with the intent of making it more consistent. IIRC two of them would have come up with a 0.3s difference in the calculated result of the very same 0-60 run.
But having been there for instrumented testing - they're not trying to game anything. I've never seen an attempt to try to come up with a desired result, only the most accurate.
An aside, speaking of manufacturer ringers: I was at a shoot for a TV show (Proving Grounds) a few years back. BMW had sent along their latest sporty compact - some M2 variant, I think - and the Veloster N was also there. The BMW burned up its back tires doing drifty corner exits before the actual test. BMW offered a new set of tires, but someone at the track would have to go pick them up in Las Vegas, a 500+ mile round trip. That wasn't going to work. Eventually the team managed to swap some front tires from a new Mustang on to the BMW wheels and the car did its timed run with a mismatched set. Also, that Mustang went back to Ford with a set of toasted front tires of a different make, which probably caused some entertainment.
Meanwhile, Hyundai had provided at least four full sets of mounted and balanced tires for the Veloster because, well, of course they're going to get used up, and making a good showing on a TV show is important.
I also had a spare set of rubber on hand for the V8 Miata I had brought :)
The whole R&T "pay to watch us test" is just an attempt to get a few bucks to cover costs. It will likely fail because unless you're just there to hear the latest exotic at full chat, it's a pretty boring way to spend a day even for free.
Glad to hear you all enjoyed the piece. I enjoyed it as well.
And, yeah, we have debated the tire issue: OE tires or an aftermarket model?
But if going aftermarket, which one? And what if doesn’t come in all of our sizes? Or what if that size is on back order? Or it gets discontinued or updated?
OE sizes or a common plus-zero upgrade?
Fresh tires for each test? And where to store all those tires?
It got complicated quickly.
In reply to David S. Wallens :
It feels like a constant battle against variables. In the end, there's likely just too many to account for.
Putting these tests together and organizing the results is a lot of work, but I'm glad you all are enjoying it.
I think testing on the OE tires is the way to go. That's how most people will drive the car for the first few tens of thousands of miles, and it is how the manufacturer chose to spec the car. They're often not off-the-shelf rubber. If they've prioritized fuel economy over grip, well, they didn't really think track performance was the main selling point of the car.
We bought a Canadian-market 1991 Passat new back in the day. It came with some sort of gumball summer tire from the factory because it was a German Sports Wagon. They wore like crazy and were completely hopeless in the snow. In Canada. Didn't take long before VW was offering a very good deal on snow tires for all Passat wagon owners.
I forgot a fun story circa 2002 or so. I worked for a tier 1 supplier as a co-op, basically a fly on the wall. We were making TPMS for [redacted] who decided to have their car prepared for a magazine shootout of new [redacted]. They called in a panic, the TPMS system was going off everywhere, everything was showing 0 psi. Now infant mortality in automotive was near-zero, and all four on one vehicle is insane, so we sent a guy there over the weekend. I heard the story the week afterwards from him.
The manufacturer sent the vehicles to be prepped. As part of the prep, they replaced the air in the tires with helium. Helium. I still don't ever understand why. Maybe the shop had a really poor understanding of mass and buoyancy? Anyway helium, being a very very small molecule, goes right through rubber pretty quickly, and while it takes a while to get through a radial tire, it goes through a diaphragm in a chip near immediately.
Nothing was wrong with anything other than the guy, clearly seeking some performance advantage by being a complete idiot. It stands to reason that sometimes that guy isn't a complete idiot.
Keith Tanner said:I think testing on the OE tires is the way to go. That's how most people will drive the car for the first few tens of thousands of miles, and it is how the manufacturer chose to spec the car. They're often not off-the-shelf rubber. If they've prioritized fuel economy over grip, well, they didn't really think track performance was the main selling point of the car.
We bought a Canadian-market 1991 Passat new back in the day. It came with some sort of gumball summer tire from the factory because it was a German Sports Wagon. They wore like crazy and were completely hopeless in the snow. In Canada. Didn't take long before VW was offering a very good deal on snow tires for all Passat wagon owners.
I agree, though I offer that it depends. Perhaps both are required. OEM tires are clearly the right answer for an SUV or a crossover to test with, but if we're talking about a car likely to be tracked, it might help to show what a trackday tire would do to the car, but only with clear explanation and clarification that you have done so, with no comparisons to other cars without that change as well.
I think GRM does that the right way - test the car with the OE tires (not OEM, that would be like putting Michelin Pilot Sport Cups on a Miata), then test the tires on some of the most popular track day cars. That doesn't tell you exactly how tire X will work on car Y, but short of testing every possible setup that's not going to happen. Also, GRM's tire tests are some of the most popular articles they publish so they're certainly not going to want to stop doing them :)
Displaying 1-10 of 23 commentsView all comments on the GRM forums
You'll need to log in to post.
