First off, I believe you are referring to cartridge-base-to-ogive measurements (CBTO) rather than cartridge overall length (COAL). You are correct in that ideally we would like the caliper insert tool to seat at the exact point on the bullet ogive that first contacts the rifling. However, there are additional caveats to these measurements, the largest of which is often bullet length variance.
As shown in the cartoon below, any bullet nose length variance between the point at which the seating die stem contacts the bullet out near the nose and the point at which the caliper inserts seats on the ogive (i.e. "critical distance") will introduce variance into these measurements.
View attachment 1187236
There are various approaches to mitigate the effect of these discrepancies. Sorting bullets using a tool such as BoB Green's Comparator (
https://greensrifles.com/new-product-page), which allows sorting bullets based on the distance between seating die stem and caliper insert contact points is one approach. Because the majority of bullet OAL variance is usually found in the nose region, sorting bullets by OAL is kind of a poor man's approach to do what Bob's tool does. Taking the average of 5 to 10 bullet measurements is another way to let the statistics work in your favor. It may also be in your best interest to buy high quality bullets that exhibit the least dimensional variance.
With regard to load development, the distance between the point on the bullet ogive that first contacts the lands and the point where the comparator insert seats are typically much closer together than the seating die stem contact point much farther out toward the meplat. In fact, we typically act as though the land contact and caliper insert contact points are one and the same, even though they are not as you observed. One reason for this is that we can measure CBTO of a loaded round with a high degree of accuracy, typically .0005" to .001". Shooting loaded rounds whose CBTO is known very accurately will tell you what shoots well and what does not. Regardless of whether the caliper tool CBTO measurement is spot on with regard to bullet contact with the lands, or even a few thousandths off, you can still reproduce the CBTO in loaded rounds that gave the best actual test results with great accuracy, over and over again.
Further, we conduct incremental seating depth testing for the purpose of tuning the precision of a load. I generally use .003" increments, but others may use .005", .002" or whatever increment suits their desire. The point is that with sorted bullets and careful measurements, it is possible to generate loaded rounds with .005" to .001" CBTO variance. The limiting factor is really the precision of the calipers you're using and how consistently your loading setup seats bullets. If you are testing seating depth in increments of .003" then your loaded rounds need only have a CBTO variance of
less than that in order that the CBTO variance is no longer the limiting source of error. For example, if you are testing seating depth in increments of .003" and you have .0015" CBTO variance, none of your loaded rounds should have CBTO variance as large as a single seating depth increment, meaning the seating depth test will still show you what you need to see. It's all about limiting sources of error, those we can measure, and more importantly, those which actually measurably affect precision.
