How can this be useful to know?
Basically, this site measures everything in terms of Normal Ticks, with Shift Ticks as a secondary, more pinpoint measurement. Whenever possible, we avoid Shift Ticks, since they are a little harder to control and count accurately. However, if an auto calls for "CA ↑ 1 ST" then instead of going up 1 Shift Tick, you can simply aim up to the Top Angle and then come back down to the Corner Angle. This will have the same effect, because it will be 1 Shift Tick higher than the actual Corner Angle. If an auto calls for "CA ↓ 3 NT ↓ 1 ST" you can do the same thing, except THIS time you'll aim all the way down to the Low Angle, come back up to the Corner Angle, and then just go down 3 Normal Ticks. The result will be the same as if you had gone down 3 Normal Ticks and then down another Shift Tick.
Why does this happen?
Notice that if you go up to the Corner Angle and then aim up, 1 Normal Tick at a time, you'll get very close to the Top Angle... but then for the last tick, without holding shift, you'll go up 1 Shift Tick automatically. Same for the Low Angle. This is because the range of all angles cannot be divided into a whole number of Normal Ticks; there is a remainder of 2 Shift Ticks.
Think of it this way: you're doing math, and you're not allowed to use decimal answers. What's 50 divided by 5? Answer: 10. Okay, now what's 51 divided by 5? Well, it's not going to divide evenly, so you could say "10 with a remainder of 1," since 10 x 5 = 50, then add 1 and you get 51. In other words, imagine that in TW3, Normal Ticks are like increments of 5, Shift Ticks are like increments of 1, and the Top Angle is 51, not 50. So you aim up to the Top Angle using only Normal Ticks (5's), and you reach 50... but that last tick is going to be a Shift Tick (increment of 1) even if you don't hold Shift.