Skin cream is the control group factor of the experiment. Nobody cares or has any preconceived cognitive bias towards using or not using skin cream and therefore attempts to examine the numbers in a rational manner.
The other difference is, its entirely plausible that a skin cream would not work, or cause more problems. It doesn't matter if the numbers are fake or hypothetical, either case presented, makes a plausible case.
When those same numbers are presented as gun control results, they don't work, because in the case of gun control, having gun control, does not increase crime. The numbers only make sense when presented one way.
There isn't really any demonstration of political bias here, because the numbers only make sense in one direction, ever, when it comes to gun control.
The only real interesting part of the video is how bad people are with proportional math. FWIW, the thumbnail onscures some of the numbers, but seeing the full data set, it was immidiately obvious what they were going for there. My thought process was along the lines of, "The bottom set looks worse but about 25 is 1/4th of about 100, and 75 times 4 is 300, which the big number is not even close to, the bottom set is better.