This site is not affiliated with the Discovery Channel or MythBusters. Please visit the Official MythBusters site for official content.

Episode 28: Toy Cars vs. Real Cars, Buttered Toast, and Contagious Yawning

  • Toy cars can go faster than a real car under the influence of gravity only:confirmed ( toy car faster at 100 ft, real car faster at 400 ft)
  • Yawning is contagious: confirmed
  • Buttered toast always lands buttered side down: mythbusted

Toy cars faster than real cars

Myth: Toy cars can go faster than a real car under the influence of gravity only.

Related links: * RaceGrooves on filling an order for a 1/4 mile track and some great photos from the actual busting

Test 1: toy car max speed

The build team ran some tests to check the feasibility of this myth - they first wanted to see what the top speed of a toy car is.

In the first test setup, they created a mount that held a toy car on top of a modified belt sander. The toy car did fine with the sander at its top speed, but only then did they determine that the sander couldn't go any faster than 43 mph.

To get up to higher speeds they decided to use a car tire in place of the belt sander. After propping up one of their real cars against the side of the building, they were able to retry the experiment. This time one of the toy car wheels exploded at 35 mph. Jamie inspected their mount and didn't like it -- he decided that the toy car wheel blew up because the mount was putting too much down pressure. They created a new mount that was basically a U-shape around the back and sides of the toy car, with nothing on top.

With the new mount, the toy car got up to 85mph before it flew out of the mount (with a lot of bouncing up to that point). They inspected the toy car and found that it was still in good condition, which was good enough results to go forward.

Wind resistance of toy cars

They also wanted to check the wind resistance of the toy cars to see what effect that would have on the max speed.

The "wind tunnel" was a leaf blower hooked up to an upwards sloping clear plastic tube. The build team measured the speed at which the cars stayed suspended in the tube, which turned out to be between 73 mph and 76 mp depending on the model.

Custom toy cars

The build team, Adam, and Jamie built their own toy cars to come up with faster models. Build team modified existing toy cars and give them better bodies, whereas Jamie and Adam both built car from scratch. Adam's model was a lead brick -- according to Jamie, "That looks like it's built for killing children." Jamie's was more like a sleek aluminum sports car.

They had a toy car ramp setup in the shop that they raced off with. Adam's brick was the fastest of the bunch.

The real test

They need to find a sloped, straight quarter mile. Suprisingly, they could not find any in SF, so they instead decided to do their testing up in Tahoe.

Test setup: * quarter mile toy car track (they claim it's the longest ever built) -- this would be changed later * Dodge Viper

They immediately had problems when the sun started warming up the road. The track became 5-6' longer in the sun, which put large kinks in the track -- by actual testing time it was far from the straightest, smoothest track that it could have been.

They also had another big problem -- when they tried to run toy cars down the track they kept on getting stuck on flying off far before the end of the track (200-400 ft down). They decided

Under multiple tests, couldn't really get a car to go more than 400 ft without flying off or stopping, so they shortened the track to 400 ft.

With the shortened track, they started timing the toy cars down the 400 ft track: * stock toy cars: 23 seconds * Adam's lead brick: DNF (too top heavy) * Jamie's racer: 18 seconds (fastest)

Toy car vs. Real Car

The final test setup was simple. The toy car and Viper were both 'released' at the same time, both going only under the power of gravity.

  • Stock toy car vs. Viper: stock toy car got to a fast start, but the Viper caught up by 200 ft and won easily.
  • Jamie's vs. Viper: Jamie still ahead at 100 ft, but Viper won overall in 13 seconds.

Jamie's car hit 28 mph, which was below the top speed of the cars and also showed that there was more potential to be had (better track, greater slope).

Final result: toy car faster at 100 ft, real car faster at 400 ft (it would have been even less of a contest at a quarter mile).

Photos from RaceGroves:

Photos from RaceGrooves Photos from RaceGrooves Photos from RaceGrooves Photos from RaceGrooves

More photos at

Is Yawning Contagious?

I have to say: this particular MythBusting was pretty lame. I yawned several times during it, and I don't think it was from the contagious aspect.

Experiment 1

The build team put out a casting call on the Internet for extras to come to MythBuster's HQ, without actually being told what for.

Control group: they crammed a bunch of people into a room for 10 minutes with no talking allowed, while filling out forms. No one yawned.

Test group: same setup, except each person was given some tax law to read. Two yawns from same person, but no spreading.

Adam and Jamie told them to do it all over again with more people.

Test 2

They built a custom observation truck that they took to a flea market. Inside were 3 yawn cubicles and a separate observation room where Tory was stationed.

Kari yawned at 2 out of every 3 people that entered the observation truck, and they then recorded whether or not the person yawned while in the room. Over 5 hours they tested 50 subjects.

Final results: * Control/No Kari yawning group: 25% of the subjects yawned * Stimulus/Kari yawning group: 29% of the subjects yawned (10 people)

According to the flaky MythBusters statistical analysis that was present throughout this episode, they judged this one as confirmed.

If I understood their setup correctly, though, this means that 4 people yawned in the control group, and 10 people yawned in the stimulus group. If just one more person had yawned in the control group, the results would have been 31% to 29% -- hardly a result in favor of this myth.

fuzzy mythbusters statistics used in support of dataset with insufficient data

[According to LiveScience: "Empirically, this is tough to deny; perhaps you'll yawn while reading this. The real question is whether there's actually something physiological at work here, and the answer is likely yes: even chimpanzees mimic each other's yawns."]

Buttered Toast: which side does it fall on?

There were two things being tested here: 1. If buttered toast falls off the table, does it prefer to land butter side down 2. If a buttered toast falls through the air, which side does it prefer to fall on?

First (Adam's) rig: Adam's rig most closely replicated a piece of toast falling off of a table top. Testing with a control sample of unbuttered test, the dominant behavior was for the toast to flip once and land top side down. They didn't need to do any more testing with actual buttered toast, as the rig clearly had a bias.

Second (Jamie's) rig: Jamie's rig tested whether or not, all things being equal, which side toast prefers to fall on. It shoots toast straight down.

With control sample testing, toast kept landing down. Once again they were statistically challenged, as they stopped after 10 samples. They determined that 3 ups and 7 downs was enough to show a clear down bias, and once again, if just one of those had been different, they it would have been 4 ups and 6 downs, which doesn't seem biased at all.

Third rig: based on Jamie's original design, but with way more over-engineering to be more automated, regular, and MythBuster-y. A conveyor belt toaster dropped the toast off onto a second conveyor belt that carried toast over to Adam, who marked the toast and loaded it into a dropper that was then released with a switch.

  • 11 up and 13 down with control sample
  • 12 up and 12 down with buttered sample

They determined this to be "less biased", so they then brought it to the roof of MythBusters HQ. From the top of the roof:

  • 26 up and 22 down with control sample
  • 29 up and 19 down with buttered sample

Jamie's theory was that for a lot of the buttered toast that landed butter side up, the buttered side was pressed in, forming a cup that affected the way the toast dropped. Regardless:



I think that a toy car is not faster then a real car. Because a toy car is influence of gravity but still it is not faster then car. That is my comment not to be rude or anythin

I'm doing the toast project for my science fair this year! I hope it works!

The toast experiment was flawed. If the toast falls of a normal table (2m hight). it will always land on the butter side. Mythbusters launched their toast in the wrong oritation.

I saw the buttered toast expirement, and i thought it was a cool test, so I decided to try it out and do it for a science project. Me and my dad built a shoot and droped the toast down it then recorded how it landed. I tested wheat toast compared to white

@Ali C: Very cool! How did the experiment go? The MythBusters also have a book out that has some more science experiments that you can try out at home.

We're doing the toast one for science fair too! ^_^ But when we smacked it off the table, the toast landed butter-side-up almost 3/4 of the time... o_o

"If the toast falls of a normal table (2m hight)..."

SAY WHAT?!? That is a 6.56 foot tall table. What pray tell is "normal" about that?!?

I did the buttered toast myth and took a ruler and a banana and made a catapult that would flip the toast at least three times before landing. The buttered side landed down 62% of the time. Not quite 50:50 as expected.