Because of how it is sold by some of the expedition organizers. Climbing Everest gives you the distinction of climbing the highest mountain in the world and if you have money it can provide good safety to you. So, a lot of people who aren't particularly well trained also try it. Everest is definitely made easier by sherpas compared to some other mountains.
Kanchenjunga, Annapurna and K2 for example are much harder and have high fatality rate despite attracting experienced climbers mostly. If memory serves me right, they have a fatality rate of about 30%.
They used to have higher fatality rate in the beginning, then with new equipment, techniques and less-reckless approach fatality rate dropped significantly, as you can see here:
Avalanche forecasting and "snow science" is a fairly developed field.
I'm only familiar with avalanche advisories in the continental US, but if you explore this map http://www.fsavalanche.org/ (click a grey area, then the "more information" button in the info box, here's one for the Central Sierra https://www.sierraavalanchecenter.org/advisory ) it'll give you an idea of the information available to the general public.
During months with an active snowpack (ie, winter) these advisories are updated on a daily basis detailing the different types of avalanche problems by altitude, aspect, likelihood, and size. Courses that teach in-the-field avalanche assessment and snowpack evaluation are widely available. Familiarizing yourself with the changing conditions over time prior to a trip will give you a pretty good idea what to expect.
In more remote parts of the world, the online information is not as available but observations over time can still be made, and I'd expect any climbing party to have the skills and tools needed for making their own assessments.
All of that said, avalanches kill people every year, even in places with resources available. Avalanche forecasts are essentially risk probabilities and how people use them comes down to personal decision making and tolerance for risk.
I don't think that kind of avalanche forecast can be applied to Himalayas. Normally, one has to dig ~6ft hole and observe the layers of snow on a slope and deduce its stability. So you have a bunch of guys going out in the morning, digging holes at various places and making an estimated map of avalanche danger. Often they place explosives to start a small avalanche at exposed locations, reducing a chance of a big one. Now in Himalayas you don't have a chance to do this as the main problem are avalanches coming at high speed from far away where nobody climbs; there you likely rely on stats and estimates from known weather patterns and avalanche frequency. However, any time a serac can get loose, kicking off a massive avalanche, and world's best climbers are out of luck. In super remote areas with a lot of snowfall like Kanchenjunga, you are on your own and completely depend on luck.
One sign of a super dangerous avalanche terrain is hearing sounds a few ft underneath when one is walking. That's a sign that there are (at least) two distinct layers of snow with different consistency and the sound comes from the break between them; likely a huge avalanche layer is about to run down the valley. When you hear something like that, don't laugh that it's funny to hear deep sounds, but run for your life towards safer terrain.
I totally agree with you, it's great to have federal funding for public advisories stateside. I also agree that big icefall triggered avalanches like you describe are a different sort than what weekend warriors like me encounter in the US, where the vast majority of dangerous slides are triggered by humans.
My broader point, intended more for those interested in if avalanche forecasting is even possible than the Kanchenjunga set, is that there are methods for assessing conditions and evaluating risk using both observations in the field and tracking conditions over time - relying on stats, estimates, and deep noises as you note.
According to this[1], Kanchenjunga has about 22% fatality rate since 1990, can't find a newer stat. This[1] article mentions 220 climbs so far so it doesn't look like the attempts have increased a lot.
Yes, Kanchnenjunga is the only mountain where recently fatality rate increased and less than 250 climbers made it to the top. Seems like that is the ultimate challenge these days. Do you want to try? ;-)
Try Cho Oyu, if you avoid crevasses/avalanches and react to your health state properly, it's the easiest one - base camp is at 5,700m, almost reachable by a car, then climb is "only" 2,500m. And there are only 5 higher peaks ;-)
Kanchenjunga, Annapurna and K2 for example are much harder and have high fatality rate despite attracting experienced climbers mostly. If memory serves me right, they have a fatality rate of about 30%.