The video (you can skip to the last five or 10 minutes of it to get the gist of what I’m talking about) says that the discrepancy isn’t unexplained. It’s point is that the different models depend on a large number of variables, and one of those variables is the size and brightness of the first generation of stars in the early universe. Tweaking that part of the model can explain the discrepancy. (According to the video.)