Records, as they say, are meant to be broken. I once blitzed across the country in 31 hours and 4 minutes. Ed Bolian beat that run. Earlier this year Carl Reese, Deena Mastracci and a team of their friends set an EV Cannonball record in a Tesla Model S. I have to say then when Team Polizei raced across the U.S. in 2006, I never imagined I’d be doing it again one day in an electric car that would drive itself 96 percent of the time.

For those still unaware, I joined Carl and Deena last week to set two new EV and autonomous Cannonball Records, driving from Los Angeles to New York City in 57 hours and 48 minutes.

Most of the time, their 2015 Tesla Model S was driving itself. The experience utterly transformed my views on the future of EVs, autonomous cars and the future of driving.

Like any record run, this one was a team effort, and not one that could be done by jumping in a car and hitting the road. Carl and Deena spent 11 months planning the journey and the route, and as soon as they knew Tesla’s Autopilot software was ready, they made sure to be ready too. And in May, they asked me to join them.


Every cross country drive I’ve ever embarked on was a big-data project. I don’t get in a car because I like driving fast. I drive as fast as I feel comfortable because of all the planning that goes into mitigating the risks. If clear weather and minimal traffic days converge on the calendar, that’s the only time I’ll consider exceeding the flow of traffic. Once you’ve spent three years planning how to cross the country in 31 hours and 4 minutes, casual speeding is no longer of interest.

What did the data tell me? Excessive speeding is inefficient. I now love bicycles and mass transit. I should get a motorcycle license. Nine years after setting the Transcontinental Driving record—since broken—I was convinced there was no more data to collect, no more lessons to learn. I was wrong.


I’ve never been a Tesla fanboy. In fact, I’m not a fan of anything made after 2000. I’m a “Fourth Car” guy. My garage contains a ‘73 Citroen SM, an ‘87 Targa, a ‘91 928 and a 2012 Morgan 3-Wheeler. The Morgan doesn’t count, for obvious reasons.

The most modern car I own is my E39 MunchenAutobahnPolizeiVerfolgungJagerGeshutz M5, which is German for best M5. Until two weeks ago, there was literally nothing about new cars that interested me. New cars are like watches. In my mind, they tell time just like old watches, only infinitesimally better.


Photo credit Deena Mastracci

But the drive changed my perspective entirely. The piece you are about to read is not intended as an analysis of Tesla the company. This is what I gleaned partially about EV’s and mostly about the Model S itself, seven days after driving one cross-country in 57 hours and 48 minutes, with less than 12 hours of charging and 96 percent of the driving done by Tesla’s Autopilot system.

Current EV technology doesn’t allow for a run against the full-on Cannonball Record of 28 hours and 50 minutes. Our driving average was approximately 65 mph. The point wasn’t speed. This was a technology demonstration. Here’s what I learned about the present and the future.


Lesson 1: Autopilot Is The Real Deal

Only by the hand of OttoNamus, the God of Self-Driving Cars, could I have been doing 90mph behind the wheel of a Model S on Autopilot when the first “Autopilot Fail” videos came out.

What a bunch of click-baiting trolls. These aren’t car people. These videos clearly weren’t intended to educate anyone about Tesla’s implementation—just one of several to be released in the near future. These people deliberately sought Autopilot’s limitations, found them, and called each shutoff threshold a defect.


These aren’t defects. These are features.

Tesla was very clear in their press, and is very clear on their in-car display, that Autopilot is in Beta. If this is their .9 release, then Tesla has absolutely hit semi-Autonomous Driving (AD) out of the park. How? By tossing the Wall-E 15 second limit common to lane-keeping systems like that of the current S-class. Finally, we have real AD. Get on the Interstate, engage Autopilot, and the car goes. It just goes. At or near the speed limit, it inspired confidence, allowing Carl, Deena and I to take turns in the driver’s seat - sometimes without touching the steering wheel for as long as 40 minutes - and casually watch 2,995 miles go by.


Approximately 96 percent of our drive was on Autopilot. Yes. Ninety-six percent, which translates to approximately 44 hours of driving without human intervention. I did about 1/3rd of the human driving, which translates to about half an hour, most of it in NYC traffic.

I’ve never felt so relaxed or alert during and after some 46 hours in a car. Although my teammates preferred to adhere to the speed limit, I couldn’t help but push the Autopilot all the way to its design limit of 90 mph. Given the right, safe road conditions, it worked perfectly.

I loved it.

Autopilot is the first truly revolutionary technology I’ve ever experienced in a car. Meaningful, real AD in the form of an improved driving experience is here. Could it be improved? Absolutely. But its flaws point a clear path forward toward the fun half of the Autonomous Future. The half that matters to car enthusiasts. As I recently wrote, the future of the car industry will see a split between Mobility and Driving brands.


The .9 release of Autopilot implemented in the Model S is a joyous example of what we can and should demand from every manufacturer who wants to remain a Driving brand. We, three experienced endurance drivers - once having learned the thresholds of the Autopilot’s capabilities - were able to maximize its use not merely to reduce our workload and fatigue, but to greatly improve our car’s performance in pursuit of our goal - driving cross-country as quickly as possible within the limits of rapidly improving technologies.

Had our goal merely been to get out of the LA basin for a spirited drive in the mountains, well, Tesla has just solved the problem of how to make the traffic and highway portions of that trip much, much easier.


Autopilot’s biggest flaw has to do with the least sexy part of our march toward broader adoption of semi-AD: Transitions.

Whereas every other manufacturer so far has been cautious about exposing drivers to potentially accident-inducing shutdowns of their autonomy, Tesla has thrown Autopilot onto the market with grossly insufficient transition alerts. When the first Autopilot accident occurs, it won’t be because Autopilot failed. It will be because it worked perfectly until the driver was forced to take over, and the driver wasn’t ready.

When thresholds are unclear, warnings are unclear, and when transitions are too short, the untrained the driver will pay the price. This is why allowing the car to cruise on Autopilot up to 90mph—regardless of the speed limit—is unwise.


I could write another 10,000 words deconstructing the .9 Autopilot’s logic and thresholds, but my enthusiasm is obvious, and Elon Musk has already hinted at upcoming upgrades specific to the analysis I can’t write fast enough.

In the meantime, I suggest restricting Autopilot to the speed limit, or the flow of traffic. I wouldn’t want to figure out the latter algorithm.

Lesson 2: Maintenance Needs A New Definition

Prior to my 31:04 record, I made a list of all the items that might fail on my then six-year-old M5: valves, seals, connecting rods, camshaft, camshaft position sensor, clutch, VANOS, fuel pump, and on and on. I took the car to two BMW dealers under warranty - both well aware (and secretly enthusiastic) about what I intended to do - and one of the best mechanics in the Northeast. Just to be sure. I spent thousands on preventative maintenance.


This was a perfectly maintained, 60,000 mile car, whose engine had been torn down and rebuilt twice. And still the fuel pump failed on the first attempt.

There is no greater endurance test on public roads than a cross-country endurance drive. How much preventative maintenance was required to set the new EV and autonomous records? None. Zero. Nada.


Carl put it all in perspective. “No oil changes. Fan belts? Don’t need ‘em. The only things to worry about are windshield wipers and tires.”

Think about that for a second. Think about how much time and money we spend on our daily drivers on preventative maintenance and unforeseen repairs. On tracking down parts. On waiting for said parts. On cursing our decision to buy the first of year of any German car, or the third year of an Italian one. On wishing one had bought a Lexus. Or a Subaru. Then, once the bill is paid, lying to oneself about how great it is to own an out-of-warranty M, AMG or RS.

Lesson 3: Consumer Reports Is Wrong

Everyone was up in arms when CR lowered Tesla’s reliability rating to Below Average. $TSLA took a vicious hit. Reliability? Go back to Lesson 2.


Ninety-seven percent of Model S owners would buy one again. The source? Consumer Reports. Still scared of Teslas? Enjoy your Range Rover… at or near the bottom of CR’s reliability rankings for as long as I can remember. Or ask Tavarish about what it takes to maintain that AMG you got for the price of a new Camry.

Lesson 4: Tesla Ownership is Another World

Imagine a world with virtually no gas stations. Few mechanics. No AutoZones. That was the world of ICE car ownership in 1915. Until two weeks ago, this was my impression of Tesla ownership. A permanent state of fear. Fear of breaking down. Fear of lack of parts. Fear that one’s leap of faith in St. Musk would go unanswered, or answered with a $40,000 bill for a bricked battery.


Let’s deconstruct this. Tesla ownership is only partially analogous to ICE vehicle ownership in 1915, and then only in the good ways. The myriad array of parts inherent to ICE vehicles aren’t necessary. The physical wearables can be counted on one hand. Tesla support is superior in most ways to that of ICE cars today. How often does BMW send a technician to a 7 Series owner’s house? Or a loaner and a flatbed? I know Ferrari and Bentley owners who can’t get that kind of customer service. The ICE post-purchase dealership experience is almost universally wretched and overpriced. If it wasn’t, independent mechanics wouldn’t have flourished.

Where did one refuel an ICE vehicle in 1915? You went looking for a general or hardware store, or maybe a blacksmith. There was no GPS. Paper maps weren’t as common as you’d think. Bookstores? A Rand-McNally store? Ha! You found gas by word of mouth, or you waited a few years until “filling stations” proliferated, first in and around major cities, and then along the fractured pre-Interstate highway system. Refueling wasn’t easy, but it didn’t stop Cannonball Baker from setting his first cross-country record in a Stutz in 11 days, 7 hours and 15 minutes.

The Tesla Model S is about three years old, yet the ownership experience is infinitely more advanced than ICE was in the teens and twenties. There are 529 Superchargers in the United States, clustered around major cities and along major interstates. Despite anecdotal evidence of charger congestion, this too shall pass as locations proliferate.


Which brings us to...

Lesson 5: What We Mean When We Talk About Range Anxiety

What is Range Anxiety? I mean, what is Range Anxiety, really? According to Wikipedia, it’s “the fear that a vehicle has insufficient range to reach its destination and would thus strand the vehicle’s occupants.” But that’s not really what it is.


Range anxiety is really the fear of restriction of freedom. We’ve lived with range anxiety since we started riding horses, which needed to be fed. Our range anxiety then migrated to cars, in which our freedom of movement was restricted to the distance to the refueling stations so rare 100 years ago. But not for long, for as gas stations proliferated, so did the range of our curiosity and freedom. Once ICE ranges topped 400 miles, range anxiety largely evaporated…

Until the Model S came out. Oh, how people cried. A range of only 270 miles?

Guess what? The most popular car of 1976 was the Oldsmobile Cutlass. Range? 264 miles.


ICE fuel economy sure has improved. So will the range of EVs. What’s that? Tesla has just announced a battery upgrade that will extend the Model S’s range to 300? Is $3000 too much to pay one-time for 46 extra miles? That depends on how much driving you do sans charging, but I’d bet the difference between 270 and 300 has more to do with what Tesla’s marketing people need than actual customers want.

There is no Tesla range anxiety per se. There is only relative range anxiety, which is relative to where we might choose to go, even if we don’t. This is similar to horsepower anxiety, which is felt by teenagers insecure about showing up to the game in an EcoBoost Mustang when the captain of the football team has a GT. All that power you don’t need = all that range you think you want.


The overwhelming majority of Tesla owners live in homes with garages. They plug them in overnight. They don’t drive 270 miles in a day, let alone 100. Yes, as happy Model S owner Steve Wozniak pointed in his entertaining article, the Supercharger network doesn’t cover 100 percent of the nation’s tourist destinations. But that problem will be solved as battery power density improves, just as ICE fuel economy did.

If range anxiety worries you now, I mean, really worries you, then don’t buy a Tesla. Fine. But for every day that passes, Supercharger ubiquity approaches. Perhaps not quickly enough for some, but for the majority of owners sufficient ubiquity is already here. What’s “Sufficient Ubiquity?” It’s when a trio drives a Model S drive cross country, Supercharges two dozen times, and encounters five other Tesla owners the entire way.


Lesson 6: Gas Stations Are Boils On Society Anyway

Suddenly, in the haze of two dozen charging stops, Carl piloted us into a Supercharger at a gas station. My mind bent. Every other charging station had been behind a Courtyard by Marriott, replete with spotless bathrooms in the lobby and bemused desk clerks happy to point us to the free breakfast buffet.

Granted, there are Superchargers with no facilities at all, which forces your environmentally conscious Tesla owners to make another stop, generally at the usual crappy gas station, to use the usual facilities.


You know what I’m talking about. Push the door open with your sleeve. Touch nothing but the paper toilet seat cover as you pull it from the dispenser. Use the back of your hand to turn the water on and off. Hope no one spit in the soap dispenser. As Carl pointed out, the absence of facilities at many Superchargers is inspiring, because no one in their right would ever eat dinner at a gas station unless they were getting gas.

Plus, gas stations are at the bottom of the barrel when it comes to food and nutrition. Walk down the aisle of any gas station and look at the offerings. You don’t need to be Neil deGrasse Tyson to know that 99 percent of the food at Chevron is compressed balsa wood flavored with sugar and salt. Gas stations do human beings no favors.


We can do better, and Tesla, through cost-saving investments in the cheapest of Supercharger real estate, is leading the way. The absence of facilities removes an inherently unpleasant experience from the culture of driving. It forces us to seek something better, at least during normal business hours. In the meantime, there’s always the fruit cup and veggie sticks, if the plastic seal isn’t broken.

With any luck, EVs in general will lead to a withering of gas station culture altogether and an improvement in the general health of all Americans, but I wouldn’t count on it. Gas pumps will gradually be replaced by chargers, and the same swill will merely be foisted on the owners of greener cars.

Lesson 7: Doug DeMuro Is Wrong

The Tesla Model S isn’t cool anymore? No, Doug, It’s getting cooler every day, with every additional Supercharger, with every wireless update, and with every performance upgrade. Insane mode. Ludicrous mode. Pipe in a fake EV motor note like BMW does. EV motors sound soooo cool. Let me tell you what’s not cool. Refusing to admit that things are changing. That things will change. That Tesla hasn’t revolutionized EV’s as much as they’ve revolutionized the perception of EV’s.


Sure, the interior is weird. The screen is an eyesore. It needs interior storage pockets. The reading lights aren’t adjustable. It lacks coat hooks. Footnotes, all.

Lesson 8: Don’t Be Afraid to Embrace Change

I hope the Model 3 is big success. I hope they stay in business. If only I could hack the Autopilot to go faster than 90. If only I had a garage or a driveway where I could plug one in. And I had a family. And a 100-plus mile commute. And I didn’t own an SM. And a 928. And a E39 BMW MunchenAutobahnPolizeiVerfolgungJagerGeshutz M5. And an old Targa. And a Morgan 3 Wheeler. But I do.


The Tesla isn’t me. It’s isn’t who I am. But I admire anyone who takes the leap. Because it is something special. It’s as much the future today as my SM was in 1970, and relatively speaking, it is much, much more so.

I can’t wait to see what Porsche has to say. That Mission-E looks like the business.

Alex Roy is the author of the LiveDriveRepeat blog and a Kinja of the same name.

Roy is really proud of coining the phrase “Autonomotive Singularity.” Roy is President of Europe By Car, the founder of Team Polizei, a columnist for Jalopnik, a host on /DRIVE and author of The Driver - which depicts his 2006 NY-LA Transcontinental Driving Record, accomplished in 31 hours and 4 minutes. He also the Producer of The Great Chicken Wing Hunt & 32 Hours 7 Minutes, was Chairman of The Moth from 2002-2007, won The Ultimate Playboy on Sky One, has competed in LeMons & the Baja 1000, and holds a variety of driving records which must still remain secret.


Please follow him on Facebook, Twitter and Instagram.

Photos credit Alex Roy, The Drive