I've tossed this topic around in my head for quite a bit and I wanted to share test data from new barrel. Mainly, because a lot of shooters that are new to handloading are hearing about Satterlee's method from YouTube and elsewhere thinking it will give them a magic load in 10-20 rounds. It might, but statistics are not on their side.
Before continuing: I have the utmost respect for all shooters in our sport and this is not a dig/slight or anything similar to Scott Satterlee, 6.5 Guys or anyone else that uses a 10 round method for development. Use whatever type of load development process that works for you. This is not a "best way" post per se, but a comparison of the two methods to help guys understand the downsides of single sample velocity testing.
I've seen quite a few posts, videos and articles about the Satterlee method and it's merits but in most cases, the merits are over hyped. This is purely based in statistics: larger samples are better. PERIOD. In it's best form, the Satterlee method is simply shooting 10 rounds to find velocity at charge intervals and max pressure. Then using the "flat spot" in velocity to predict an accurate, low SD node. But honestly that's not really a method; that's part of loading safely and again, statistically speaking, it's pretty unlikely to have the "flat spot" actually represent a low SD node.
There are instances where the initial velocity testing 10-shot style methods are valuable and even advisable: an unfamiliar catridge, vastly different bullet from other loads, powders without reference manual data. But this is good for finding max pressure; not for finding best "nodes".
Here is a picture of a multishot/OCW style test with a new Bartlein 26" 6.5CM barrel chambered by Area 419, 140 ELDMs, Hornady brass 4th fire, BR2 primers, (40 rounds through barrel when test began) :
Here is the data if you look at like running 5 Satterlee tests with the same charges. Notice that each String (like running a new Satterlee test) shows different results. I've hightlighted the areas that would called nodes using this method, but its different or ambiguous for each new sequence. But if you simply use the SD for each charge weight, the decision is Simple and easy: between 41.6-41.8gr.
I logged the data averages into Excel to get this set of graphs:
To be clear, the data in the charts is not a Satterlee plot, it's the avg of 5 shots per charge, which means the velocity listed per charge has MUCH higher confidence than if I simply used one or two shots each.
The upper graph is looking at SD by Charge weight and the lower is velocity by charge weight (AKA Satterlee Method).
If I used Satterlee Method: I would have fired 10 rounds to find my node for future work up, I would have tested 41.1-41.5 at .1gr intervals with 5 rounds each (another 25 rounds following the results from above). Then I would need to test seating depth at best SD/group from above (another 5 rounds each +-.010 increments; about 25-30 rounds). Total rounds needed to find load: 35 without changing seating depth and 60-65 with seating depth test.
BUT I would have been testing the wrong node and ignored a series of charges that would have reduced my SD by +50%! I would never know it existed if I didn't fire more rounds to begin with!
Now using the OCW like method (this is not an OCW exactly; that would evaluate relative POI by charge weight): I fired 5 round groups at .2gr charge increments (30 rounds total) and have the following results from 30 rounds: an average velocity for charges that is reliable, avg group size by charge, POI shift by charge, plus most importantly: SD for each charge. I can say with confidence that any load with 41.6,41.7, or 41.8 should have an SD around 3-5fps, velocity between 2749 - 2765 and produce groups of about .5-.6 or better. If I feel I need to decrease group sizes, I could then test seating depth (another 25-30 rounds) and be done.
Essentially, I've gotten to the same round count with both methods but I have better data shot in the same conditions, and best of all, I don't have any doubt that I've picked the best node possible.
Some might wonder why not choose 41gr as it's the smallest group and SD is still single digits. Simple: all of my "5th shot fliers/misses" are left which indicates is probably me. I was having difficulty with NPA shifting during string and my common miss is left. But the other 4 shots tell me 41.6 is capable of .3" and 41.8 is capable of .4" with practice.
So in summary, both methods will require a minimum of about 30 shots to develop a load, but any method with more shots per charge will at least let you make the right choice with better data more confidently and more quickly with fewer range trips. If you don't know the cartridge or bullet well, you might not shoot some of the loads due to pressure, but you will have good data for rest.
Edited: updated with all data and comparison as if running 5 separate Saterlee tests to check reliability of data/results.
Before continuing: I have the utmost respect for all shooters in our sport and this is not a dig/slight or anything similar to Scott Satterlee, 6.5 Guys or anyone else that uses a 10 round method for development. Use whatever type of load development process that works for you. This is not a "best way" post per se, but a comparison of the two methods to help guys understand the downsides of single sample velocity testing.
I've seen quite a few posts, videos and articles about the Satterlee method and it's merits but in most cases, the merits are over hyped. This is purely based in statistics: larger samples are better. PERIOD. In it's best form, the Satterlee method is simply shooting 10 rounds to find velocity at charge intervals and max pressure. Then using the "flat spot" in velocity to predict an accurate, low SD node. But honestly that's not really a method; that's part of loading safely and again, statistically speaking, it's pretty unlikely to have the "flat spot" actually represent a low SD node.
There are instances where the initial velocity testing 10-shot style methods are valuable and even advisable: an unfamiliar catridge, vastly different bullet from other loads, powders without reference manual data. But this is good for finding max pressure; not for finding best "nodes".
Here is a picture of a multishot/OCW style test with a new Bartlein 26" 6.5CM barrel chambered by Area 419, 140 ELDMs, Hornady brass 4th fire, BR2 primers, (40 rounds through barrel when test began) :
Here is the data if you look at like running 5 Satterlee tests with the same charges. Notice that each String (like running a new Satterlee test) shows different results. I've hightlighted the areas that would called nodes using this method, but its different or ambiguous for each new sequence. But if you simply use the SD for each charge weight, the decision is Simple and easy: between 41.6-41.8gr.
I logged the data averages into Excel to get this set of graphs:
To be clear, the data in the charts is not a Satterlee plot, it's the avg of 5 shots per charge, which means the velocity listed per charge has MUCH higher confidence than if I simply used one or two shots each.
The upper graph is looking at SD by Charge weight and the lower is velocity by charge weight (AKA Satterlee Method).
If I used Satterlee Method: I would have fired 10 rounds to find my node for future work up, I would have tested 41.1-41.5 at .1gr intervals with 5 rounds each (another 25 rounds following the results from above). Then I would need to test seating depth at best SD/group from above (another 5 rounds each +-.010 increments; about 25-30 rounds). Total rounds needed to find load: 35 without changing seating depth and 60-65 with seating depth test.
BUT I would have been testing the wrong node and ignored a series of charges that would have reduced my SD by +50%! I would never know it existed if I didn't fire more rounds to begin with!
Now using the OCW like method (this is not an OCW exactly; that would evaluate relative POI by charge weight): I fired 5 round groups at .2gr charge increments (30 rounds total) and have the following results from 30 rounds: an average velocity for charges that is reliable, avg group size by charge, POI shift by charge, plus most importantly: SD for each charge. I can say with confidence that any load with 41.6,41.7, or 41.8 should have an SD around 3-5fps, velocity between 2749 - 2765 and produce groups of about .5-.6 or better. If I feel I need to decrease group sizes, I could then test seating depth (another 25-30 rounds) and be done.
Essentially, I've gotten to the same round count with both methods but I have better data shot in the same conditions, and best of all, I don't have any doubt that I've picked the best node possible.
Some might wonder why not choose 41gr as it's the smallest group and SD is still single digits. Simple: all of my "5th shot fliers/misses" are left which indicates is probably me. I was having difficulty with NPA shifting during string and my common miss is left. But the other 4 shots tell me 41.6 is capable of .3" and 41.8 is capable of .4" with practice.
So in summary, both methods will require a minimum of about 30 shots to develop a load, but any method with more shots per charge will at least let you make the right choice with better data more confidently and more quickly with fewer range trips. If you don't know the cartridge or bullet well, you might not shoot some of the loads due to pressure, but you will have good data for rest.
Edited: updated with all data and comparison as if running 5 separate Saterlee tests to check reliability of data/results.
Last edited: