Only once you’ve shot enough groups to encompass the precision capability of the rifle/shooter combo.
Search through
@Ledzep posts.
You’ll see that once you fire a statistically confident amount of shots (even from a rifle without a shooter driving it), that you group size widens enough that *most* POI shifts from setting to setting are within the precision “cone” the rifle is capable off.
Therefore you wouldn’t be completely confident that the setting is repeatable until you’ve tested it thoroughly.
Obviously you then run into issues of throat erosion and such. So let’s say there was an absolute definitive test/s that proved that tuners absolutely work the way we think and are repeatable.
That still leaves us with the issue that the setting we *think* is ideal is *possibly* nothing more than standard statistical variance. There’s different ways to account for this and narrow it down to be more confident. But it’s definitely not being done by most shooters.
So, there’s a two prong issue. No one is saying something doesn’t work because the sample size is too small. They are saying the sample size is too small to be confident your end result was not “luck.”
And then there’s the second issue that we have just begun to scratch the surface on harmonics/vibrations.
A simple example:
If I took a person who had absolutely zero idea the odds of flipping a coin was 50/50 and had them flip it 5x. It lands on heads 5x.
They may think the odds are pretty good for landing on heads. It’s still 50/50. And the odds of the 5x heads is only 1/32.
Now, we make them flip it again. Most people won’t get another 5x heads. But, 1/1024 people will get heads again for 10x heads.
Now, use that example in the shooting/internet world. Most people aren’t going to test 2, 3, 4x, etc. They will just test the once. They get results that are the equivalent to the 1/32 above and then relay their “test” to the world as successful.
A few others will test a second time. And some will get the 1/1024. Definitely less than 1/32, but enough that some will go online and say it works.
Then you have the 6,7,8,9x in a row heads also telling everyone the 50/50 stat is not real based on their “testing.”
Now, it’s very easy for anyone to see it’s 50/50 because there are only two outcomes.
But, when you have an exponentially larger number of outcomes, that are also affected by the shooter, environment, etc etc……
Now the odds of most people experiencing standard statistical variance (the 1/32…..but now much less odds), us much, much, much higher. And so are the odds of people who did 2, 3, 4, 5 tests…….
So, now you can see how small sample sizes can easily take over and spread like wildfire. The Satterlee velocity flat spot method is a prime example. It perpetuated extremely fast.