http://blog.interviewing.io/you-cant-fix-diversity-in-tech-w...

The salient point is about halfway through when they put up the big chart of a couple thousand interviews and the stddev vs. mean of technical score, and say:

As you can see, roughly 25% of interviewees are consistent in their performance, but the rest are all over the place. And over a third of people with a high mean (>=3) technical performance bombed at least one interview.

Now, what this points to is a very clear failure of these standardized interviews with their allegedly "tee-ball" questions, because the results shown would be astoundingly unlikely in a world in which the interview process had any kind of consistency in identifying technical skill.

So, no. You're not doing what you think you're doing, and if you've got a high failure rate the problem is almost assuredly not with the candidates. It's with your interview process, or with you.