However, if you have a 5% applicant rate being women and you start hiring 20% women to try and force address the issue artificially ... I stop agreeing with the narrative/agenda. Two wrongs don't make a right.
I'm virtually sure that I agree with you in philosophy, but I think this example is flawed.
Imagine a field (say, software engineering) that both pays very well and is inherently interesting to some people who have an aptitude for it. That 19:1 male to female ratio might exist because all of the people who find it interesting and have an aptitude apply and (in addition) a bunch of men apply because it pays well and, since most jobs suck anyway, you might as well try to work at one that pays well.
Further imagine the applicant pool consists of 4 women with aptitude, skill, and interest for software engineering, 16 men with aptitude, skill, and interest for software engineering, and 60 men who picked software because they liked video games and Google or their high school guidance counselor gave them a list of the highest paying professions.
In such a situation, you might very well find that the average female candidate is better than the average male candidate (by virtue of a bunch of the addition of that last group of men to the pool). It doesn't even have to be that 4x as many men are trying to join the field because of pay. The candidates who have lesser skills, interview poorly, and aren't selected don't just dematerialize; they keep applying for jobs, meaning they keep showing up and not getting selected until one day they wiggle a little higher on the interview performance or click a little better with the interviewer and get hired. When they do land a job, they might be fired more frequently (and thus be over-represented in the applicant pool from this effect).
Just because a group is hired at a higher rate doesn't mean there's inappropriate discrimination at play. In the example above, an ideal interviewing process would hire all 4 women and the 16 men with aptitude, skill, and interest for software engineering. That would result in 20% of hires being women from a pool that was only 5% women applicants, but would be an entirely "correct" outcome.
Anecdotally, in 30 years of working in software, I almost never run across a female software engineer who is terrible. I've run across quite a few men who are terrible at it. It's quite possible that the women who would be terrible at it are filtered out earlier in the pipeline (or alternatively, that the men who are terrible at it were attracted to it for reasons other than inherent interest and competence).