Sahuagin posted... why not? that's exactly the kind of thing compression does to simplify the image to compress it.
here's an example. things to notice include:
* the edges of the back wheel of the other bike are almost but not quite stationary, while the middle of the wheel is very clearly in motion. this is because the pattern on the edge repeats, and the compression is turning something that's "pretty close" to something that's exact. the middle of the tire is messy enough that the pattern doesn't repeat frame to frame, so the compression ignores it.
* even better, at times the rubber on the top of his front tire appears completely motionless. this isn't (just) because it's in synch with the camera, it's because the image on each frame look "pretty close" to the compression, so it converts it into something that's exact frame to frame.
https://www.youtube.com/watch?v=GBUJtUKk5KA&t=25m22s
I might agree compression is involved in causing this effect with the tire tread. When you described a bike wheel, I was imagining a stationary observer watching the bike roll past. The stroboscopic effect is also known as the wagon wheel effect as the spokes of a wagon wheel can do the same thing as propeller blades. I didn't imagine a scenario where the camera is looking at the tire treads which are held relatively stationary in a small part of the frame because the camera is traveling with the tire.
Sahuagin posted... it can't be for the bicycle wheels though since they're not moving fast enough.
The speed of the wheels doesn't need to be as fast as you think. Yes, an object could turn 360 degrees in one frame to create the effect. That video appears to be 25 fps, so that would require 1500 rpm for the wheel. I agree that's way too fast. But you pointed out the edge of the tread appeared motionless while the center of the tread didn't. For mountain bike tires, the edge of the tread is often lined with a repeating pattern of studs that repeats much more frequently than the studs along the center of the tread.
Let's say there are fifty studs in the tread aloud the edge of the tire just to make the math simple. Because there are fifty identical positions, it could rotate as little as 7.2 degrees per frame to match the frame rate of the camera. That's only 30 rpm. And the wheel could turn any multiple of 7.2 degrees to reach another identical position. So any multiple of 30 rpm would make the studs on the tread appear motionless. 150 rpm is around 10-12 mph depending on the diameter of the tire.
The studs along center don't repeat at the same rate so the effect doesn't manifest down the center of the tread. When they reach the top of the incline they're able to move more quickly and the rate of revolution is able to match one of the speeds capable of causing the center of the tire edge to repeat each frame.
So I think I'll still argue this effect is present in your linked video, though I'm not completely sure.
---
The only reason lightning never strikes in the same place twice, is that the same place isn't there the second time.