That's an interesting point and yes, you wonder what real-event evidence it's based on - it would be nice to know what is the best method

In the three examples you mentioned, did you actually have time to exit the building before the quake had finished?

I've had three earthquake (two were quite unnerving) experiences where things have started smashing (technically four big ones but one of them I was on the beach). Of those three, two were in high-rise buildings - and you're right, that instinct to get out is strong but in the first of those that just led me out onto the balcony - then came an aftershock and I realized that the balcony may not be the safest place and between shocks I went down seven flights of stairs and got out of the building - where everyone was milling around - by that time it was pretty much over

The second time was on the fifth floor of a department store - in the glassware and ceramics section! (which amplified the drama but actually it was the least serious quake of the three Lots of stuff smashing and people screaming - but there was no way to really make a run for it. The good news was that as it struck, I was admiring a very nice, but expensive glass bowl - which miraculously survived the quake - apart from a a small chip - for which I got a very good discount from a startled sales lady once the pandemonium had subsided

The third was here in a one-floor steel and tin hut - I did move outside - but actually - in that situation I realized that a tin house (though not ideal in a typhoon) is pretty much earthquake-proof so I spent the time after the main bump just securing things that hadn't already smashed

One thing that might help is to start cluing people into fore-shock patterns. There were two 6M's before the 8.1 in Japan in the preceding two days. Could it be that it's when this relatively common pattern starts, that it would be a good time to exit the building, rather than waiting around for the big one?