certainly...
Of course the reach will increase, but if you are comparing two frames and do the same on both, then both have the same increase in reach. What you are overlooking is the fact that every rider has only ONE saddle height and ONE saddle setback amount that will be the same on both frames being compared. All you've been doing with the saddle height information is figuring out where the top of the seatpost will sit. That makes no difference, as long as the saddle ends up at the same location on both frames, relative to the center of the BB. That might require a post with a different setback but it does not change the fit of the frame.
What you need to figure out is horizontal distance from the center of the BB to the exact same point at the head tube for both frames - that's the difference in reach. That will also tell you the stem length difference required to make both frames fit the same. The difference in seatpost setback should not be intermingled with this calculation. The seatpost setback difference is simply (cosA-cosB) times the saddle rail height (not the actual saddle height), where A and B are the two STAs.
I've had frames with very different STAs and used a zero setback post on the frame with a 72.5 degree STA to position the saddle exactly the same as my other frame with a 74.5 degree STA, and a 25mm setback post. Measuring from the saddle tip to the center of the bars, the frames also measured the same, with the same stem length. The TT lengths were much different, but what was the same is the REACH. The same reach requires the same stem length. The saddle on these two bikes could be moved up, down, forward or backward, but as long as the same movement was done on each one, they both have the same fit.
Read the instructions below and try making the drawing I have described. It proves with no doubt that saddle height and setback are not needed to compare the fit of any two frames. It merely assumes that the saddle will be in exactly the same position, relative to the BB on both frames.
Here’s a graphical method that might convince some skeptics that the concept of frame reach will predict the difference in the fit of two frames without any knowledge of the rider’s saddle height. I drew up a representation of two frames on graph paper. Using graph paper makes it much easier to make the drawing and measure the reach. In this example, I am defining reach as the horizontal distance from a vertical line through the bottom bracket to the head tube/ top tube intersection point. Cervelo uses a slightly different point at the top-center of the head tube, but the same principles apply. The defined point at the head tube can be set to any desired vertical stack height, but that stack height must be the same on both frames under comparison.
1) First draw a vertical line of a convenient length. I made mine 3 inches long, with each end at a horizontal grid line on the graph paper. Mark a point at the lower end of the line to consider the center of the bottom bracket.
2) Draw a horizontal line intersecting the top of the vertical line. Mark this line TT for top tube.
3) Next, draw a line that represents one seat tube centerline, starting at the center of the bottom bracket. Rather than worry about a specific angle, I drew mine so the upper end of the line was two squares behind the vertical line and horizontal line intersecting point.
4) Next draw another line representing another frame’s seat tube centerline. Make this line have more setback. Instead of two squares, make the setback four squares for an extreme difference.
5) Extend both of the “seat tube” lines further up, to represent a saddle height. The amount makes no difference – just make the lines the same length and some reasonable amount above the TT line.
6) Next, draw a horizontal line representing the saddle and end the line behind the vertical line through the center of the bottom bracket. Consider the end of this line to be the saddle tip, with some amount of setback, behind the center of the BB. If the saddle tip setback and height are the same on both frames, the sitbones will also be in the exact same position.
7) Draw two points or short vertical lines along the TT line to represent the ends of the top tubes on each frame. Make each mark the same distance from the starting point – the intersection point of the TT line and the seat tube line. That will represent two top tubes of the same length. Notice how one of these points or vertical lines is much closer to the vertical line through the center of the BB. The amount should be the same as the difference in the seat tube setbacks. On my drawing, this is two squares on the graph paper. Two squares is the difference in the reach of the two frames that I have drawn, but both have the same TT length.
8) Draw a vertical line extending up from the end of each of the top tubes. They should be the same two squares apart, just like the seat tube setbacks. Make the lines long enough to extend higher than the saddle height line.
9) Measure from the point representing the saddle tip to each of the vertical lines just drawn. The difference is still the same two squares and the same as the difference in the reach. You could move that sadle tip point anywhere you want, up, down, forward or back, but the difference between the reach of these two frame will remain the same - two squares.
To make these two "frames" fit the same, the stem on the frame with the steeper seat tube angle must be shorter, or the stem on the frame with the slack STA made longer, by the difference in the reach - whatever amount the two squares on the graph paper represents. Then the horizontal distance between the saddle tip and the handlebar centerline will be the same on both frames, with the saddle in the same position relative to the bottom bracket. No knowledge of the actual saddle height or saddle setback was required for this analysis. The saddle height or setback can be anything you want, as long as it's the SAME for both frames being compared.
I’ve also stated that the stack height must be the same to make a valid comparison in reach. To convince yourself of that, draw an angled line from the end point of either top tube to represent an extended head tube. The further up the line is extended, the closer it gets to the vertical line through the center of the BB and the shorter the reach. If you compare reach values that are calculated at different stack heights, then you will not get a proper prediction of the stem length difference required to make each frame fit the same. A correction that is accurate enough just requires an increase in the reach difference of 3mm for each 10mm of stack height. If you look at two frames with that have a 10mm difference in the published reach (from the manufacturer), but one has a 20mm taller stack height, then the real difference in reach is about 16mm. A good example is the 51 and 54cm Cervelo R3 frames. Since these two frame have the same STA and HTA, the difference in the TT lengths should match the difference in the reach. When two frames have the same STA, all you really need to do is take the difference in the TT lengths to get the difference in the stem length required to make the frames fit the same, but it good to know that a proper reach comparison produces the same answer (within 1mm). I suspect a little roundoff error somewhere.