So, they are wrong. First bad advice ever on the internet. Lmao
What does it mean when there is no velocity/pressure node?
- Thread starter MoneyWaster
- Start date
If I understand your post correctly, your MV data is the average of 7 identical rounds at each charge weight. If that's correct, then your data is a LOT more significant than that gathered by people trying to show the same thing with only one round per charge weight.
I would say you might be looking for something which is more subtle than a genuine "flat spot". And some times that information is hard to envision, especially if you're looking at a matrix of numbers.
Here is a graph of your data showing MV vs Charge Weight along with a linear trend line in red:
That "flat spot" you're searching for may actually be only a slight dip and that's why they say a picture is worth a thousand rounds, or something like that. At a charge weight just over 24 grains, where you got your second best group, is exactly where the MV vs Charge Weight line takes a dip. Perhaps that's all there is there, just a little deviation from linear. But it does give a hint that the theory you're investigating might have something to it.
Of course, your best group is right where the line goes the "wrong" way. Does that disprove the theory?
Argue among yourselves.
In God we trust. All others bring data.
One data point is not statistically valid.
To prove a theory is valid it must be repeatable.
Based on what constitutes valid information I wou.d say repeating the test several times is warranted. After all this is a process that is supposed to be a short cut.
So now who is snarky? The guy who added nothing to the discussion?
One data point is not statistically valid.
To prove a theory is valid it must be repeatable.
Based on what constitutes valid information I wou.d say repeating the test several times is warranted. After all this is a process that is supposed to be a short cut.
So now who is snarky? The guy who added nothing to the discussion?
Mulligan
Silver $$ Contributor
Good morning there Sunshine
The OP said he shot 7 rounds of each load, if i read that wrong my bad.
That is plenty of data to move forward while narrowing down the chase.
I personally never use a chronograph until the load development is mostly complete. Most of the time i couldn’t care less what the velocity or corresponding SD and ES is while doing load development. The target speaks for it’s self, at 600 and 1000 yards there will be flat spots in point of impact over a small range of powder charge weights. Many time several flat spots.
Yes, you have to go back out and prove the load, however is does save a bit of barrel life. In my opinion this is a great way to go about load development.
In order for a sample to be statistically sound and pass a peer review by standards accepted in the scienctific community the sample size has to be so large that we as shooters are not willing to go there.
Dealing with such limited populations (barrel life), a valid sample would still have to be near a 1000 data points. None of us are willing to do that to our barrels just testing. We look for the best info at hand and go for it, making adjustments as we go along.
Testing at the distance you chose to compete at is solid advice. A load that works at 1000 may not be your best load for 600 or up close.
Good morning and hope the new year is full of joy for you and yours.
Kind Regards
Clay
Edited to include pic
Last edited:
Mulligan
Silver $$ Contributor
Right at 24 grains looks to be the most forgiving spot in the data presented
CW
I'm glad you spoke up, not sure how many gun forums or other avenues all here gather info from, but this and weaker methods of load dev are becoming prevalent. I've yet to see a flat spot, other than just before your load becomes pressure, plateaus out before spiking.
There's a fraternity of guys going off numbers only, sd numbers of 3 or less and calling it a day. < Even clowns refusing to tighten things with seat depth changes. As long as the sd numbers tight, it will hold vertical at distance, and horizontal drift is automatically outside interferences, wind which is now aerodynamic jump or elliptical swerve(spelling on purpose), drunk bullets, lol
Even a friend of mine, who I've worked with on 3 previous rifles, I showed him how things work at 100 yards compared to 500 and beyond. Anyway a third in my circles, helping again, we showed some groups shot at 500 yards, the only question muttered was, "how do the numbers look". Well, here's the target at 500, here it is again at 1K, who cares. I've lost him, lol
I have seen flat spots. In my 338 Lpaua Imp for an example. I tested 103, 104 amd 105 grains of RE33 witha 300 grain Berger at 1000.,the 104 and 105 load overlapped one another and shot a 4 1/4 group. I colored the bullets amd hre was no vertical. Even though the chrono siad tjere was 50 feet,per second difference. A lot of times this happens right at peak load. Matt
Matt, I was only discussing flat spots in numbers, I know all about charge weight ranges holding flats on target, it's what it's all about, not numbers.
mikegaiz
Stay frosty, my friend.
6.5/47 Lapua Berger 140 and Lapua 139 in .5 grains increments.
Wrong bubba
Everyone knows a statistically valid sample size is 30 for a normal distribution.
Everyone knows a statistically valid sample size is 30 for a normal distribution.
Mulligan
Silver $$ Contributor
https://www.surveymonkey.com/mp/sample-size-calculator/
Thirty is a rule of thumb for some stuff. Checkout the link above.
Have a great evening
CW
30 is fine this purpose.
Your suggestion of a huge sample size is only a smoke screen you pasted from the net.
Your suggestion of a huge sample size is only a smoke screen you pasted from the net.
Last edited:
memilanuk
Gold $$ Contributor
The sample size >= 30 thing comes from t-tests, where you're trying to tell if sample 'A' from the same population as sample 'B'. Not really what we're after here. Sample size and 'power' varies with the amount of (in)accuracy you're willing to accept in the answer, and with the experimental budget you're willing to expend to tighten up that answer. There are specific calculations for what the correct sample size is needed to get the statistical 'power' you want - and when you get into things where the difference between sample 'A' and sample 'B' is very small, then the technically correct sample size to be able to say that they are indeed different is surprisingly *large*. Way more than any sane person is likely to want to burn thru their barrel.
Or you can let it go. 99% of what we do in reloading falls more under the category of 'exploratory data analysis' aka EDA, than trying to statistically *prove* there is a significant difference between each increment in load development. If you want to load up a big batch of rounds to prove (to yourself, cuz no one else cares) that your new final load is better than the one you started with, or that of the two most promising candidates - load 'A' really is better than load 'B'... go for it. Saying that someone's load development is bunk because they don't shoot 30+ of each increment is just about as silly as someone who shoots *one* of each increment and calls it good. One is wasteful excess and a misapplication/misunderstanding of theory, and the latter is placing *way* too much faith in the bullet fairy.
Or you can let it go. 99% of what we do in reloading falls more under the category of 'exploratory data analysis' aka EDA, than trying to statistically *prove* there is a significant difference between each increment in load development. If you want to load up a big batch of rounds to prove (to yourself, cuz no one else cares) that your new final load is better than the one you started with, or that of the two most promising candidates - load 'A' really is better than load 'B'... go for it. Saying that someone's load development is bunk because they don't shoot 30+ of each increment is just about as silly as someone who shoots *one* of each increment and calls it good. One is wasteful excess and a misapplication/misunderstanding of theory, and the latter is placing *way* too much faith in the bullet fairy.
To think every barrel is going to have an accuracy node at x, y, or z velocity seems awful simple minded. You'd think it was depending on the variables of barrel diameter, length, caliber, and velocity. If every barrel is accurate at a known velocity why even test?
Some forums have embraced this simpleness without question.
Some forums have embraced this simpleness without question.
mikegaiz
Stay frosty, my friend.
Not all barrels like all bullet / powder / primer combination. We look for repeatable barrel harmonics. Some barrels will not shoot well know matter what you put through them. Barrel weight, Length, boar size, twist, # of lands, taper, matter, not to count powder, bullet, primer, case, etc. That is why we test.
Last edited:
If I understand your post correctly, your MV data is the average of 7 identical rounds at each charge weight. If that's correct, then your data is a LOT more significant than that gathered by people trying to show the same thing with only one round per charge weight.
I would say you might be looking for something which is more subtle than a genuine "flat spot". And some times that information is hard to envision, especially if you're looking at a matrix of numbers.
Here is a graph of your data showing MV vs Charge Weight along with a linear trend line in red:
View attachment 1030691
That "flat spot" you're searching for may actually be only a slight dip and that's why they say a picture is worth a thousand rounds, or something like that. At a charge weight just over 24 grains, where you got your second best group, is exactly where the MV vs Charge Weight line takes a dip. Perhaps that's all there is there, just a little deviation from linear. But it does give a hint that the theory you're investigating might have something to it.
Of course, your best group is right where the line goes the "wrong" way. Does that disprove the theory?
Argue among yourselves.
I think that I will load up 25 rounds at 24 grains and shoot 5 sets of 5 shot groups to see what the average group size is. Thanks for the graph!
Monte, Invoking a Hegelian dialectic to tar us “bullet fairies”, equating our highly successful and economical methods with those silly lost souls at the other end of the spectrum, is equivalent to saying you would like to eliminate Half of all government waste
memilanuk
Gold $$ Contributor
Said methods are only economical when they actually work... which they often do, or no one would espouse their use. But when they *don't* work, when you end up back tracking and starting over from scratch... the same 'cost' in time and effort could have been better spent up front with a somewhat more rigorous testing method.
Some people have great fortune with abbreviated test methods. Kudos for them. Me, not so much. I've been 'lead astray' too many times to put much faith in doing things that way. But if it works for you, rock on
Upgrades & Donations
This Forum's expenses are primarily paid by member contributions. You can upgrade your Forum membership in seconds. Gold and Silver members get unlimited FREE classifieds for one year. Gold members can upload custom avatars.
Click Upgrade Membership Button ABOVE to get Gold or Silver Status.
You can also donate any amount, large or small, with the button below. Include your Forum Name in the PayPal Notes field.
To DONATE by CHECK, or make a recurring donation, CLICK HERE to learn how.
Click Upgrade Membership Button ABOVE to get Gold or Silver Status.
You can also donate any amount, large or small, with the button below. Include your Forum Name in the PayPal Notes field.
To DONATE by CHECK, or make a recurring donation, CLICK HERE to learn how.