Anybody with a rudimentary understanding of statistics, which apparently does not include World Athletics decision makers, would have realized that the false-start system jumped the shark. It was eliminating runners below a 0.10 mark when multiple runners were posting marks of 0.10x and 0.11x. That's not how error detection systems work. They can't distinguish between "just this side of correct" and "just the other side of correct." They must be constructed with barriers.
Thus, if the official DQ level is 0.10, then those who constructed the system must think that nobody could legally start in less than, say, 0.12 seconds. Build in a 0.02 buffer, to acknowledge that 0.12 is merely a signpost and that there can be no hard-coded signal for reactions that occurred before the gun and those that occured after the gun, and that's how you get 0.10.
So when runner after runner is posting start times of less than 0.12, you screwed the chicken. They didn't all anticipate the gun by a tiny fraction of a second. There's no way at all that happened. What that means is your system is broken. Because if it worked correctly then false starts would look like this.
0.155
0.153
0.144
0.139
0.137
0.131
0.126
0.122
0.94 - DQ
But that is not what is happening out there. The false starts are much closer to the average. As an earlier post noted, the standard deviations with the false starts when compared with the legal starts aren't even statistically signficant, that is how close the margin is.
If the World Athletics engineers built a bridge, it would collapse the first time that the wind blew hard.