Home » When Should You Try To Swerve To Avoid A Crash?: Comment Of The Day

When Should You Try To Swerve To Avoid A Crash?: Comment Of The Day

R1t Crash Top

Mercedes is galavanting through California with her lovely wife, Sheryl, meaning I have the pleasure of making fun o—I mean, bestowing upon some of you the honor of “Comment of the Day.” Today’s COTD winners (yes, plural today) have questions about the physics of crash testing, but they also suggest some advice that you absolutely, positively should ignore. Read this; actually, don’t.

Welcome back to Comment Of The Day! Every day, we read every single comment posted on our site and pick the one that made us laugh, get informed, or feel warm inside. You don’t have to go into our comments sections and write thousand-word stories about why you love a car so much, but a lot of you do, and that means a lot to us. So we’re highlighting some of the most excellent bits of thought that you’ve formed into words and digitized onto our website.

Anyway, I shouldn’t be writing this post, because my garage just flooded due to a burst pipe, and all my tools/car parts are actively freezing to the garage floor. In truth, I’m not that worried, because I’m certain that, so long as I can crack the ice into small enough pieces, those pieces should lift right up off the garage floor given that the concrete is completely saturated with oil. That ice won’t be sticking on there very well. Still, I’d rather not have to use a hammer and pick to pack for my move to LA, so I really ought to pick those tools up, like, right now.

But Comment of the Day ain’t gonna write itself (well, it is, but the article about the comment of the day won’t), so let’s give some love to — or, in this case, let’s poke some fun at —  readers hugh crawford and BolognaBurrito, who, upon seeing how strenuous the IIHS’s Small Overlap Rigid Barrier (SORB) test — which involves the very outer 25 percent of the vehicle’s front end hitting a rigid barrier, thus replicating a crash between two vehicles driving towards one another but only just hitting one another — asked themselves: Would it be better to just hit a car straight on?

So does this mean that you should not try to avoid a crash, since doing so could put you in a small overlap crash situation, and instead just lean into it and go full-frontal? This idea is, of course, absolutely ridiculous — so much so that I figured I’d share these comments with you:

Screen Shot 2022 12 20 At 7.46.21 Pm

hugh crawford acknowledges the ridiculous nature of this question with “That also seems like the definition of a perverse incentive,” and that’s true. Avoiding a crash is obviously the best answer. But what if trying to avoid a crash results in stresses to your car that put occupants in more danger than if the driver hadn’t tried avoiding the crash? I mean “stresses the car” not just by focusing crash energy on a smaller area (stress equals force over area, after all, and in this case we’re talking about 25 percent of the frontal area), but — it’s also worth considering — by possibly rotating the vehicle and inducing a side collision or rollover instead of a frontal collision. (This is why automakers tend to build understeer into cars — so that they crash facing their tree or car or whatever, since the nose of the car is best designed for an impact). And then there’s risk to your surroundings — are there pedestrians or other vehicles that are potentially in your crash-avoidance path?

At the same time, leaning into a crash instead of trying to avoid it — for example, hitting an object perpendicularly — can transmit more of your kinetic energy into material destruction than if you hit it at an angle, as some of that entry energy could remain and/or convert to rotational energy (you can see how in some small overlap crash tests, the vehicle continues moving after the test).

I think the answer is: You’re probably not going to have time to make a choice, anyway. IIHS’s small overlap test does imply that hitting another car or a rigid barrier with just 25 percent of your car is more strenuous than hitting those objects with 50 percent of your car’s front end or, it follows, with its entire front end. But none of these is better than zero percent of your front end, and that’s likely what you’re instinctively going to reach for; hopefully your car’s stability control can keep you under control during such a swerve.

Anyway, I’m looking forward to more discussion on this in the comments.

Share on facebook
Share on whatsapp
Share on twitter
Share on linkedin
Share on reddit

26 Responses

  1. Ohh, comment of the day! Where are my my valuable prizes?
    On hitting things, when I’m on my bike in city traffic and someone opens a car door in front of me I aim for the hinge rather than try and miss the door which may open wider, not to mention getting deflected into traffic. And you might hit the driver who is softer than the edge of the door. Safest of course is to ride in the middle of the street.

  2. As a highly qualified armchair physicist, I would theorize that a small overlap would be better as it leads to a deflection rather than a dead stop (pun intended) from a direct hit.

    Really though, all accidents are different and I’m sure there’s no hard rule on it. Velocity would also play a large role, obviously.

    1. With strictly rigid bodies, yeah maybe. However today’s cars are most definitely not designed to be rigid bodies.

      With a small overlap I believe the tendency is to spin off of each other, with full overlap you both expend all of your energy coming to a stop. On the surface that sounds like the worst case scenario but you also have to think of time of impact and how that relates to force, ie: longer time = less force {F=(m*delta-v)/t }
      It is going to take much longer to come to a full stop than it will to bounce off the other car (inelastic collision.) Again though, not the full story; then you need to take into account the change in velocity after impact (elastic collision.)

      You would probably need a spreadsheet comparing vehicle weights and speeds and at that point if you need to consult a spreadsheet before you crash into someone something has gone very very wrong.

  3. What a great question, I would really like to hear some thoughts on this matter. My drive to work is over a two lane coastal mountain road, full of twisties, along with people trying not to be late for work, and of course a lot of commercial trucks making their morning deliveries. Having made this drive around 8K times, I have swerved to the point of scraping the mountainside more times than I can count, it’s my default mental state to be ready. If I really can’t avoid it, I would love to know what’s best to do. As long as I don’t get pushed over the side into the revine.

  4. Well, this i something I unfortubately have first hand experience in. The instinctive reaction in a case like this is to try to steer away from the crash. 9 years ago an oncoming car came into my lane and I swerved to make a 75% overlap crash into a 25% crash much like in the test shown. Both cars were probably traveling at roughly 50mph. My 2002 CR-V vs a quite bit smaller Mitsubishi Space Star (not available in the US afaik )hatchback. And while it was definitely a hard hit, the cars did deflect snd some energy ws turned into rotation as both cars turned in behind each other after swiping each other. Would still recommend just hitting the brakes as I think we just got very lucky that I didn’t swerve enough to go off the road, or for the other car to hit us in the doors first. Still mostly destroyed the whole side of both cars and crushed my front wheel into the floor under the a-pillar/clutch pedal. But no serious injuries.

  5. “and all my tools/car parts are actively freezing to the garage floor”

    “Still, I’d rather not have to use a hammer and pick”

    You have salvaged the hammer and pick from the floor, just in case? Or are they stuck to the floor already?

  6. Tossabl is 100% right. And, I would add, as long as you have ABS, slam down on the brake pedal, and steer where you want to go. The ABS will give you maximum braking while you are still able to steer.
    That’s the best you can get. Learned this through some on track exercises at Spring Mountain Motorsports in Nv.

    1. Exactly, even for car-aware folks like the commenters on this site, it’s tough to remember that you can both brake and swerve at the same time. But if you’re in a multilane highway and you’re swerving into another lane with cars possibly overtaking, or if you’re going into the oncoming lane without visibility, swerving without knowing what’s in that lane could prove to be much, much worse than just hitting the rear of the car in front of you, with all of your car’s front end (well unless you’re in an Isetta, in which case, you have already presumably made your peace with God) and airbags to help out.

  7. Just a quick tip/suggestion. Could you guys resize the browser window to a more vertical orientation when talking the screenshot of the comments. It will make it much easier for us mobile users and shouldn’t hurt the desktop users very much.

    1. They should just not screenshot text. The other writers who have done COTD lately have figured that out, but apparently David was busy ice skating in his garage when that memo went out. 😉

  8. I came here expecting some napkin math to suggest one way was “right” or at least under some vague hand waving and simplification that at least one way was right given certain conditions/assumptions/situations. I got nothing.

    1. So this isn’t really a physics problem, as physics is just the basics of the problem. Because this has man-made objects that react differently depending upon how the problem is set up, this is an engineering problem and the answer to every single engineering problem ever given is “it depends.”
      It depends on how you set the problem up, it depends on your assumptions, it depends on how granular you want to get with the answer.
      Hell, it even depends on what you meant by “fixed objects.” I hear that in relation to car accidents and I don’t think parked cars, I think oak trees. Hitting a parked car is vastly preferable to hitting pretty much any type of tree.

      Ultimately this is going to require a spreadsheet, vehicle weights, speeds, and a whole host of figures.

      I went into a bit of it on a reply to CoolDave

  9. it’s my opinion, unless you have incredible spatial awareness, you shouldn’t swerve, just plant your foot down, brake hard straight, look around while doing that, if possible, swerve at the last second, when speed has been reduced considerably.

  10. Can we just take a moment to say how impressively the Rivian performed in this test? The front end was completely demolished but the cabin hardly budged at all. I keep rolling the clip back and forth over the moment when the barrier violently separates the tire from the rim, the rim from the face of the wheel, and fires the brake rotor like a clay pigeon. Oh, except for a chunk of the rotor which apparently stayed in the caliper. I can’t stop watching it… like, um, a car wreck, I guess.

  11. Side discussion: If it is an animal, hit it. Not because I am mean but a lot of people end up kissing a tree trying to avoid a deer. Bad outcome, sometimes fatal.

    1. Depends on the animal… if you hit a horse (or moose) at speed, there’s a very high chance its entire body is coming through the windscreen, making you “at one with nature” in the most undesirable way possible

  12. I had a car accident involving my rare 1971 Alfa Romeo 1750A Berlina (yes, with automatic gearbox). It was raining little bit hard, and I was driving through the construction site along the future Dallas North Tollway in Far North Dallas. Suddenly, the idiot in his Chevrolet pick-up truck, crossing from the left side, stopped in the middle of the intersection for no reason. The construction barriers on both sides prevented me from doing the evasive manoeuvre. My car’s brakes weren’t stellar for the wet surface during the panic braking. So, I veered slightly onto the oncoming lane and aimed for the weakest part of his truck: rear passenger-side fender, which seemed to help lessen the collision damage on my car.

    After the witnesses calmed me down, I asked the idiot why he suddenly stopped in the middle of the intersection. He had no idea why he did. The police came to investigate and determined that he was on medications and shouldn’t be driving at all. My car only needed the new front clip and some repairs, yet the parts were hard to come by and quite expensive.

    His insurance carrier was a professional arsehole, determining that my Alfa Romeo wasn’t worth anything and offered me the pittance of $800. My father dissuaded me from lawyering up because he was kind of “happy” to see my car damaged. He was fed up with its high maintenance and frequent issues. Instead of using $800 to fix up the car, he took the insurance money and bought me something very reliable, a 1982 Buick Skylard (spelling is intentional), from his company fleet. That put a lot of strain on our father-son relationship for a long time…

      1. I, unfortunately, have never seen Death Race 2000. But, if video games have taught me anything it’s that you get points for style, too. So, Larry C is still correct, it’s about the points.

Leave a Reply