When you dig into the data, the difference between SMU and TCU in his analysis is almost fully centered on instructional wages per student, with SMU's being much higher at $14,250 than TCU's $9,823. Both schools are lower-vulnerability due to healthy endowments and near-average percentages of international students. In terms of value, both have identical experience and credential scores. The remaining component of value is education, for which wages-per-student is a key element.
This comes back to the earlier discussion of whether it's a good thing that TCU is able to attract and retain quality faculty at lower-than-average wages. The assumption of the model is that TCU is overpriced relative to spending on faculty, which is sufficient to push it into the "Struggle" quadrant instead of "Thrive"--simply because faculty are willing to sacrifice some pay to be at TCU.
One scenario: The bubble will burst for Challenged schools as the international market goes elsewhere. Newly out-of-work faculty push wages lower across the board (with some faculty going overseas to serve the international market). TCU can't reduce wages as much as others but it will move closer to the market average as other schools cut pay. TCU's adjustments may fall to a combination of program/plant/tuition reduction.