It's important to understand that there are some
very strong selection biases here in the U.S.
While fatalities per passenger-mile may be 24x cars (as below), you have to realize that the population riding motorcycles in the U.S. is not a random sampling. These people aren't trying to get around, they're acting out an image. For that behavior I don't blame them, but it is what it is.
Anecdotally, a little less than half of U.S. motorcyclists are fat, bald, drunk dudes riding Harleys or Harley look-alikes with the legal minimum protective gear (which in some states is none at all), while another little less than half are young, shirtless dudes on sportbikes. To their credit, they are wearing full-face helmets (because you can't go 200mph without one). Also, something like half of motorcycle fatalities involve alcohol[0]. Yikes.
<10% of motorcyclists are riding reasonable, well-maintained bikes with a modicum of skill and all of the proper protective gear.
So while motorcycling qua motorcycling may be more dangerous than driving a car per vehicle-mile, the difference is probably a lot smaller when controlling for differences in the behavior of people doing it. To put it in perspective, in the U.S., if motorcycling is ~30x deaths per vehicle mile compared to cars, bicycling is ~20x (though you do get exercise, which extends your life). And miles per motorcycle (or bicycle) are << miles per car per year in the U.S. People don't realize, but moving to the suburbs and driving everywhere vs. living in the city is taking the same sort of risk one might by motorcycling. </rant>
[0]: https://en.wikipedia.org/wiki/Motorcycle_safety