Sometimes, for some, it's easier to visualize if it's in a visual format so.........
If you had a perfect bullet aligned perfectly in a perfectly aligned case you
would have something that looks like this. Numbers are for demo only but are actual measured dimensions in the drawings. In this case you can see that there would be "0" runout anywhere you measured either on the case or on the seated bullet.
If you had a perfect bullet poorly aligned in a perfectly aligned case you
would have something that looks like this. In this case the bullet misalignment pivoted around the base of the bullet. Misalignment would be different depending on what point along the center axis of the bullet the misalignment took place.
Because the line the measurements are taken from is parallel with the center line of the perfect case/bullet the runout measures the same as if it was taken from the center line. The misalignment is greatly exaggerated to emphasize the misalignment.
You can also see that it makes a difference as to where on the bullet you take your measurements. The TIR will be different for different places. Most people indicate where they take their measurement when stating TIR. I believe you'll find that most people will measure the bullet runout at the juction with the case neck so as to have a consistent place for measuring. Your measurements can be on the bullet or the case but as stated previously, the goal is "0" for all TIR measurements.
If you want to get really carried away you can measure the TIR of a case neck on a loaded round, mark the high point and record it. Then measure the TIR of the bullet, mark the high point and record it. Then what you do with those measurements is up to you but get too carried away and you'll just end up with a headache.