The "Scott Satterlee" method isn't a good statistical analysis to find a node. I've come to this conclusion after a few years of reloading for precision rifle, many of them using the "Scott Satterlee" method. I use to buy into that method (hey, how doesn't like a faster and easier way to do things?). But I've come to the conclusion that it isn't the best method for finding the "node" for your loads. I don't think a single one of my dialed in low SD loads would have ever been found in a "node" produced by the "Scott Satterlee" method.
I use a ladder to test velocities and pressure signs. It's basically irrelevant for finding nodes. Once I have an idea of what my velocities are for each charge, I'll load up groups of 5 rounds for each charge I want to test. I'll shoot those groups, and from that I'll have a better idea of which charges will produce a more consistent velocity. 5 rounds of a charge will provide a much better statistical analysis than 1-2 rounds.
For example, for my 6BRA, using the "Scott Satterlee" method, my so called "node" loads end up with SD's of 7-8+ (5 shots). Using the method I describe above, my SD's are 2-3 (5 shots). Same story with my .300NM. And my many barrels of 6.5 Creedmoor.
Before blaming the AMP, I would use a better method than the "Scott Satterlee" method for finding a good load. I use an AMP for all my reloading, and I love that machine. All my loads have SD's of 5 or less (this is over 10+ rounds).
If you don't believe me, try it. My eyes started to open up when I started to shoot two ladders side by side, and get different results. The nodes would be different between every ladder, and I started to realize the "node" information from this method was garbage. Any of my statistic teachers would have called me an idiot if they found out that previously I used a ladder with only 1-2 rounds to find a velocity node.