Range Report Group size means little

[47guy]

and to really through a wrench in the gears here i personally think that a chronograph is the biggest mind fuck piece of equipment made as well as a very useful piece of equipment...i watched my smith shoot a 2" 5 shot group at 601yds on saturday with a 50fps ES then shoot a 1.5" 3 shot group at 600yds through our shot marker with the same load.

now with a 50FPS ES you'd expect to see some issues and when you dont do you chalk it up to an odd round or do you start looking for a problem?

 

[BCX]

ES will start to show its demons at distance. 600 yards is still within 22lr and handgun range

 

[BCX]

600 yards I'm at 2.4 mil drop (52 inches) w 2450 fps. Im still burning powder in my tail wind...

 

[morganlamprecht]

and to really through a wrench in the gears here i personally think that a chronograph is the biggest mind fuck piece of equipment made as well as a very useful piece of equipment...i watched my smith shoot a 2" 5 shot group at 601yds on saturday with a 50fps ES then shoot a 1.5" 3 shot group at 600yds through our shot marker with the same load.

now with a 50FPS ES you'd expect to see some issues and when you dont do you chalk it up to an odd round or do you start looking for a problem?

yup...thats the reason i lean towards bergers over hornady...ive seen bergers consistently print smaller vertical at 800 yds when 100 yd groups and chrono #s are the virtually identical between the 2

 

[47guy]

ES will start to show its demons at distance. 600 yards is still within 22lr and handgun range

lol!!

 

[47guy]

yup...thats the reason i lean towards bergers over hornady...ive seen bergers consistently print smaller vertical at 800 yds when 100 yd groups and chrono #s are the virtually identical between the 2

that’s all we shoot is Berger’s

 

[morganlamprecht]

that’s all we shoot is Berger’s

ill shoot either, both plenty capable for most things, but in controlled situations the bergers win for sure...you just pay a little more for it

last years NRL matches i shot 2 matches with hornady and 2 with bergers...6mm and 6.5mm once for each

 

[47guy]

ill shoot either, both plenty capable for most things, but in controlled situations the bergers win for sure...you just pay a little more for it

last years NRL matches i shot 2 matches with hornady and 2 with bergers...6mm and 6.5mm once for each

hornady never shot as well as Berger’s for me.

 

[PAYDIRT]

I don't own a chronograph, sure I'd like to have one but.... I do have 100-700 right outside my door, literally my front yard, and out to 1500 walking distance from the house. Doesn't take long to find out if a load falls apart at distance. I can't imagine "working up a load" then driving miles to check it at distance and it goes from touching @100 to complete trash at 850 and beyond.

 

[morganlamprecht]

ive never had a load shoot tight at 100 and not make it to 1000-1500 just fine...with capable speeds and proper bullets

i have had situations shooting certain types/weights of bullets or lower consistency hunting bullets that wouldnt make the trip...but no load work up would have helped me...the bullets werent consistent enough or they just started getting wild in the 1100-1300 fps range at distance

cheap 120 gr nosler hunting bullets cant really keep up with high BC bergers from my 6.5 @ 800 yds, nor would i expect them to
certain types of 155 and 168s from a 308 wont hang with 178-185...one of my 308 barrels will shoot ragged holes @ 100 w/ 155s and wear plates out inside 700 yds, but i cant buy consistent hits on a 10" diamond @ 1k...same barrel wont shoot 185s as good at 100, but it wears the plate @ 1k out

other 224/6/7/6.5/308 match bullets ive pushed well into subsonic and they made the trip with no issues

 

[Greg Langelius *]

My views about groups and performance standards are not razor clear, not based on metrics, but on the binary certainty of success or failure. I honestly believe that many of us get hung up in the numbers, and that this can be regrettable.

How I got there is more visceral than analytic.

A successful shot has no groups size since it's but a single instance. There is no spread, there is no average.

My criterion is whether I can engage a target at random and defeat that target. One can reduce the variables to the minimum, but there is an inarguable reason (whatever that may actually be) why perfection is unattainable. I accept that I will never achieve perfection, and I probably smile too much as well. Some answers are genuinely optional.

In my estimation, a conventional match is simply a string of individual target encounters, many starting very similar, but the environment is always changing. Score is just a byproduct. My match is but a series of one shot matches; won or lost one shot at a time.

It has less to do with target size or group size than it does about the individual encounter.

Greg

 

[Ledzep]

Here's some SD-by-the-shot action. Notice that at 5 shots I'd believe I had 6fps SD, but the long term actual SD is closer to 11fps. At the 5-shot mark this can go either way. It can be horrible (15+) and level out to normal, it can be right on the long term average, or it can be below the long term average. Moral of the story is that they ALL level out at 20-30 shots into the test.

And here's MPOI vs. shot count for horizontal and vertical position. The numbers on the left side are in inches (for scale) but are just from a reference location on the target. Again, this typically settles out about 20-30 rounds in, and is moving around quite a bit for the first 5-15 shots. Things to think about the next time you re-zero based off of a 3 or 5 shot group, then encounter a tenth or two of error between real life and your calculator at range.

As the barrel fouls velocity increases and vertical POI does seem to shift. I did see a loose trend between MV and vertical POI across tests, which is nothing mind blowing.

 

[2aBaC̶a̶]

I don't own a chronograph, sure I'd like to have one but.... I do have 100-700 right outside my door, literally my front yard, and out to 1500 walking distance from the house. Doesn't take long to find out if a load falls apart at distance. I can't imagine "working up a load" then driving miles to check it at distance and it goes from touching @100 to complete trash at 850 and beyond.

That would be a dream. Driving 2 hours to "test" a load at distance blows.

 

[Ledzep]

Plotted final group. 50 shots, 5x 10 shot groups. Each 10-shot is it's own series, all corrected to the final 50-shot Mean POI.

 

[Tchitcherine]

That's gonna give you a really solid zero!

 

[lash]

Here's some SD-by-the-shot action. Notice that at 5 shots I'd believe I had 6fps SD, but the long term actual SD is closer to 11fps. At the 5-shot mark this can go either way. It can be horrible (15+) and level out to normal, it can be right on the long term average, or it can be below the long term average. Moral of the story is that they ALL level out at 20-30 shots into the test.
View attachment 7304779

And here's MPOI vs. shot count for horizontal and vertical position. The numbers on the left side are in inches (for scale) but are just from a reference location on the target. Again, this typically settles out about 20-30 rounds in, and is moving around quite a bit for the first 5-15 shots. Things to think about the next time you re-zero based off of a 3 or 5 shot group, then encounter a tenth or two of error between real life and your calculator at range.
View attachment 7304785

As the barrel fouls velocity increases and vertical POI does seem to shift. I did see a loose trend between MV and vertical POI across tests, which is nothing mind blowing.

Okay so after looking at these charts, reading your comments and relooking and re-reading, I have a couple questions:

Are you starting with a clean barrel for each string of 50? Your comment starting with, “As the barrel fouls...” implies that to be the case.

Doesn’t it seem as if 10 shots gets you within 90-95% of your eventual answer after 50 shots? I may be reading those charts incorrectly,but that’s what they seem to be telling me.

I am not picking nor poking. I like good data and good charts. It just seems like that is what I’m seeing.

 

[memilanuk]

@Ledzep as I said, we all 'get' your view on sample sizes.

So how do *you* deal with that during your load development? Shoot 5 x 10 of each increment you're looking at? For final confirmation, sure. For 'exploratory' workup that seems a tad impractical.

 

[2aBaC̶a̶]

I think you can do better.

Load up another 50 each with +.2 and -.2 gn and try again.

 

[Ledzep]

Okay so after looking at these charts, reading your comments and relooking and re-reading, I have a couple questions:

Are you starting with a clean barrel for each string of 50? Your comment starting with, “As the barrel fouls...” implies that to be the case.

Doesn’t it seem as if 10 shots gets you within 90-95% of your eventual answer after 50 shots? I may be reading those charts incorrectly,but that’s what they seem to be telling me.

I am not picking nor poking. I like good data and good charts. It just seems like that is what I’m seeing.

No, not starting clean. I cleaned them every 200-300 rounds and shot 10 foulers after cleaning before starting a test. Velocity bumps up over the shot count between cleanings regardless. .223 was the worst, IIRC it was something like 40fps over 30 shots on average.

10 shots is better than 5. The above is one sample of many. Some are better, some are worse. By that I mean some level off quicker, some take 25 to settle out. How many shots that takes is beyond my control, so 20 shots is where you start getting to the point that you've got a really good chance of being within 1 click (.25 MOA or .1 mil) of actual long-term MPOI, getting within 1-2fps of actual long-term SD, and getting a really good picture of total dispersion.

@Ledzep as I said, we all 'get' your view on sample sizes.

So how do *you* deal with that during your load development? Shoot 5 x 10 of each increment you're looking at? For final confirmation, sure. For 'exploratory' workup that seems a tad impractical.

I shoot 20 shots. I don't fine tune loads, I pick a charge based on the velocity I want and run with it. My testing has shown that adjusting powder charge and seating depth requires gross adjustments to make minor changes. Testing every .2gr and every .010 or .005" for seating depth comes out in the wash if you shoot enough shots.

If 41gr of RL16 with a 140 (20 shots) in a 6.5 Creedmoor doesn't work, I'll try H4350 (20 shots). If H4350 doesn't work, I'll try a 147 (20 shots), etc...

The rough formula is:
.025" off the lands
1gr under max thru max book load (if this is a new barrel work your way up from a few grains under)
Favorite brand of primer (Primer makes little to no difference in my testing for ES/SD, and dispersion)

If the barrel is going to shoot, it's going to shoot. If it's a dog, it's a dog and I'll get another.

Big picture: Shoot and document enough rounds and you'll see that by shooting small sample sizes you're acting upon noise. In general, Barrels and bullets make accuracy, powder and cartridge design make ES/SD small.

 

[kthomas]

Its funny how some people get tied up in 100 yard groups just to get that load out at distance and it falls apart. FB forums are great for a good laugh lol

FB is full of a bunch of brain dead, can't think for themselves, blindly follow what their chosen "leader" says and spit out the rhetoric. Anyone that challenges the general rhetoric gets piled on by all the brain dead zombies.

The precision rifle groups on FB were the final nail in the coffin for me and FB. The rhetoric and stupidity is full steam ahead there. I for one can't stand echo chambers.

 

[Dthomas3523]

If the group is solid at 100, and the ES is within the acceptable area, there’s little to no reason a load will fall apart at 1k that didn’t fall apart at 100. If the load work up was done properly and the bullet is a stable/known commodity.

 

[lash]

No, not starting clean. I cleaned them every 200-300 rounds and shot 10 foulers after cleaning before starting a test. Velocity bumps up over the shot count between cleanings regardless. .223 was the worst, IIRC it was something like 40fps over 30 shots on average.

10 shots is better than 5. The above is one sample of many. Some are better, some are worse. By that I mean some level off quicker, some take 25 to settle out. How many shots that takes is beyond my control, so 20 shots is where you start getting to the point that you've got a really good chance of being within 1 click (.25 MOA or .1 mil) of actual long-term MPOI, getting within 1-2fps of actual long-term SD, and getting a really good picture of total dispersion.



I shoot 20 shots. I don't fine tune loads, I pick a charge based on the velocity I want and run with it. My testing has shown that adjusting powder charge and seating depth requires gross adjustments to make minor changes. Testing every .2gr and every .010 or .005" for seating depth comes out in the wash if you shoot enough shots.

If 41gr of RL16 with a 140 (20 shots) in a 6.5 Creedmoor doesn't work, I'll try H4350 (20 shots). If H4350 doesn't work, I'll try a 147 (20 shots), etc...

The rough formula is:
.025" off the lands
1gr under max thru max book load (if this is a new barrel work your way up from a few grains under)
Favorite brand of primer (Primer makes little to no difference in my testing for ES/SD, and dispersion)

If the barrel is going to shoot, it's going to shoot. If it's a dog, it's a dog and I'll get another.

Big picture: Shoot and document enough rounds and you'll see that by shooting small sample sizes you're acting upon noise. In general, Barrels and bullets make accuracy, powder and cartridge design make ES/SD small.

Thanks. I appreciate the clarifications.

 

[2aBaC̶a̶]

Here's some SD-by-the-shot action. Notice that at 5 shots I'd believe I had 6fps SD, but the long term actual SD is closer to 11fps. At the 5-shot mark this can go either way. It can be horrible (15+) and level out to normal, it can be right on the long term average, or it can be below the long term average. Moral of the story is that they ALL level out at 20-30 shots into the test.
View attachment 7304779

And here's MPOI vs. shot count for horizontal and vertical position. The numbers on the left side are in inches (for scale) but are just from a reference location on the target. Again, this typically settles out about 20-30 rounds in, and is moving around quite a bit for the first 5-15 shots. Things to think about the next time you re-zero based off of a 3 or 5 shot group, then encounter a tenth or two of error between real life and your calculator at range.
View attachment 7304785

As the barrel fouls velocity increases and vertical POI does seem to shift. I did see a loose trend between MV and vertical POI across tests, which is nothing mind blowing.

Am I looking at this correctly? The difference between your 5th shot and 40th shot.

ES = 4fps
Horizontal MPOI = .01"
Vertical MPOI = .05"

 

[memilanuk]

If the barrel is going to shoot, it's going to shoot. If it's a dog, it's a dog and I'll get another

No way to be 100% sure (even with hind sight and all that) but I'm guessing that's a large part of some of the frustration I've battled off and on over the years. Specifically, my stubborn streak / reluctance, once I sink the money (and time) into getting a new barrel put on, to say "f$ck it" and get another one. It's probably cost me *more* in terms of time and money, in the long run, but as they say, hind sight is 20/20 and all that.

I've known people who made similar claims about particular lot #s of bullets (even Bergers) - and seen a little bit of that myself. One lot, you just about couldn't make them shoot *bad*. Others... shot pretty well... with a bit of massaging/prep, and a lot of time at 'load development'. Again, once you've sunk the time and money, it's hard to quit beating that particular dead horse.

Primer makes little to no difference in my testing for ES/SD, and dispersion.

In general, bigger picture terms, I'd agree. I've seen a couple instances over the years, one of them quite painful, where that wasn't the case. That one was more of a matter of a bad lot #, rather than brand, though. Couldn't get any more of the original lot #, so changing brand was the quickest 'fix' - went from ES/SD of 41/18, to 10/4. And yes, the sample sizes were 'small', but they directly reflected the vertical dispersion I was seeing on target @ 1k.

 

[Ledzep]

Am I looking at this correctly? The difference between your 5th shot and 40th shot.

ES = 4fps
Horizontal MPOI = .01"
Vertical MPOI = .05"

They are running numbers. The SD graph shows the SD at 2 shots, 3, 4, 5, 6, 7, .... 45, 46, etc... Basically if I took the SD of the total number of shots to that point.

The other graphs show a running average of point of impact, horizontal and vertical.

At 5 shots, had you stopped there, your chrono would tell you that you had a 6fps SD, and your 'zero' would be .1" left and .025" or so high from the long term actual zero.
At 10 shots, had you stopped there, your chrono would tell you that you had 10fps SD, good horizontal, and .125" vertical shift from long term actual.

This is one set of data. I was briefly going through it earlier and there was one that moved over 0.3" from 5 shots in. Some settle down in 5 shots. Some are stable across the board, some settle down at 30 shots. The point is that you DONT KNOW IF WHAT YOU HAVE IS STABLE IF ITS NOT A STATISTICALLY SIGNIFICANT SAMPLE SIZE.


They almost all do this one one form or another:

What I'm getting at is that you need to shoot 20 shots minimum (like rock fucking bottom minimum) to have a good chance of being past the noisey data and into the stable data. It still wanders around a little bit, but the changes are much more gradual and much smaller at that point. Like .05" shift or less instead of .1-.3".

Again,

Here's another way to look at it. If I take a large sample size (100+ rounds) and I shoot them in a shit load of 5 shot groups, then repeat the exact same test with a bunch of 10 shot groups, then a bunch of 15-shot groups, then a few 20 shot groups, then a couple 30 shot groups, and compare it to the average of a couple 50-shot groups, this is what it would look like:

So you tell me when you shot a single, or 2, or 3, or 4, 5-shot groups... which one(s) of the red dots you were given by the gods of randomness.

ETA: Now, if you shoot 6-10 5-shot groups, and record a common reference (same POA, for example, and overlay all of the individual groups into 1 composite), you'll find that the total average of those compiled 5-shot averages will (nearly) equal the same thing as the 50 shot average. That is fine, you have real data. But a few 5-shotters here and there, and "hmm must've pulled that one" data discards, with no common reference, means almost nothing from a real statistical analysis point of view.

Which is fine. You can do that, and you can be happy without a statistical analysis point of view. Whatever makes a dude happy. .5 MOA rifles sound great, and if you believe in it, great! .5 MOA rifles exist, and are "proven" to exist by all of the independent sub-.5 MOA 3 and 5 shot groups people post on the internet. You don't have to understand that in terms of total dispersion no such thing exists and you're gambling on every shot past ~600-700yd (target size dependent) to have fun.

 

[kthomas]

I still think if you really want to nerd out on this, you have to do testing with the shooters influence completely removed, putting the rifle in some sort of testing apparatus.

Shooting 30+ rounds will be testing the shooter more then the gun. Between shooter fatigue and discrepancies/consistency issues between groups, you are not getting a fair representation of what your gun and ammo is doing. Very minor changes in how the shooter interfaces the rifle with skew the POI. Different loading pressure on bipod, different hand pressure, different cheek wield pressure, differences in managing the rear bag, etc. Very small changes can have a big impact on the POI. More then most think. Then you have stuff like sun shift to think about.

Neat academic discussion, but without proper testing, I think we are testing the shooter more then the rifle and ammo. I believe a lot of what we are talking about is shooter induced.

You geeky motherfuckers should put your rifle in a testing apparatus and shoot until the cows come home. Have proper controls and a testing procedure in place, and get back to us and let us know how the results compare to your hypothesis.

Not knocking anything here, and I'm sure there's some truth to it. But I think there's a lot of shooter induced errors accounted for in the results of the current conversation that people are dodging around.

 

[47guy]

I still think if you really want to nerd out on this, you have to do testing with the shooters influence completely removed, putting the rifle in some sort of testing apparatus.

Shooting 30+ rounds will be testing the shooter more then the gun. Between shooter fatigue and discrepancies/consistency issues between groups, you are not getting a fair representation of what your gun and ammo is doing. Very minor changes in how the shooter interfaces the rifle with skew the POI. Different loading pressure on bipod, different hand pressure, different cheek wield pressure, differences in managing the rear bag, etc. Very small changes can have a big impact on the POI. More then most think. Then you have stuff like sun shift to think about.

Neat academic discussion, but without proper testing, I think we are testing the shooter more then the rifle and ammo. I believe a lot of what we are talking about is shooter induced.

You geeky motherfuckers should put your rifle in a testing apparatus and shoot until the cows come home. Have proper controls and a testing procedure in place, and get back to us and let us know how the results compare to your hypothesis.

Not knocking anything here, and I'm sure there's some truth to it. But I think there's a lot of shooter induced errors accounted for in the results of the current conversation that people are dodging around.

EXACTLY!!!!

again the most over thought thing on the planet!

 

[memilanuk]

I recall years ago when I was shooting Service Rifle with an AR-15, there was a guy on the old nationalmatch.us forum who was an armorer for some Guard teams. He mentioned testing their guns in test fixtures for exactly the reasons listed above.

He also said that in his experience a really good match AR would agg about 0.75-1 moa for a 30 shot string. Everyone else (including me) protested, with examples of our quarter to half moa ARs.

Starting to think he might have been on to something ?

 

[morganlamprecht]

i still have people tell me that bench vs prone vs barricade will give different POIs

While they sit there and watch me shoot from all 3 positions and punch the same holes @ 100 yds...

“well, for me....”

every rifle I’ve shot in a NRL/PRS match for the last 2-3 yrs was zero’d and doped from a bench...cause I don’t feel like laying down when I don’t have to lol

rifles don’t get a free pass from me...I’ve got enough rifles and barrels to know better

 

[Ledzep]

That has also been tested in a hydraulic return accuracy fixture by someone I know. 0.7 MOA for 30 shots was what they were getting (different bullets, caliber, case prep, etc.. than I use. Not sure on the details there). A friend that shoots BR volunteered 30 rounds to the cause and produced ~.9 MOA group (8.4fps SD). Myself with 12-14lb rifles have been in the 1.0-1.3 range with Kreiger/Bartlein and 1.2-1.5 range with a Wilson blank, 1.2-1.6 with a factory CZ barrel all for 50 shots. The 1.0 MOA Barrel consistently shoots tighter than the other 2 6.5cm barrels. The BR rifle had full OCD case prep done with neck-turn-to-fit cases.

I'm not saying I'm perfect for 50 shots every single time, but the vast majority of the shots that break feel good. I bag everything out and shoot in 10-shot strings with 10-20 minutes between to let the barrel cool and let my eyes rest. Glass is an AMG and it's not hard on my eyes. Nonetheless, there's probably about a .100" diameter zone that I could be aiming in and not really be able to tell the difference. Also bear in mind I'm shooting 50 shots, the people I've talked into trying this out have shot 30.

My interest in starting all of this is first to find out what matters and how much as far as status quo load development dogma, AND to also get a better feel for system capabilities to apply down the road to hit probability. In the latter application the significance of having good data shows through. In other words, convincing myself that I have 15fps ES, 3fps SD, and a .3 MOA rifle does me a disservice by making me overconfident in hit probability. Even using accuracy rest fixture data doesn't tell me anything about reality. Understanding that in any given situation I'm actually drawing from a 1.1 MOA pool of random selections, NOT a 0.5 MOA pool ultimately makes easier my ability to assess shots, and better address how much work and of what type to put into load dev.

Anyway, end of the day if the shooter adds .3-.5 MOA, the shooter adds .3-.5 MOA. I know there are better shooters out there, but I suspect I'm right in line with the better half of them. Nonetheless, I'm happy to have someone prove that I suck, and happy to see sub-MOA 30-50 shot strings from a 'tactical' (non BR/Fclass) rifle if you can do them. If there's a way to get better dispersion, better MV spreads, etc... then I'm all ears because that was the intent of this to begin with. But so far I've tested most (not all) of the status quo methodology ad nauseam and come up with fewer tools than I thought I had at the beginning.

 

[morganlamprecht]

Also just to note...myself and Led began testing what we were testing completely independent of one another, i only know him thru the hide, not personally

the only reason we crossed data (after we both had come to similar conclusions) was because he randomly posted some data or a graph in a thread a couple months back and it happened to catch my eye

like led, I tested commonly known methods and methods of close friend shooters...over and over, multiple days...I’ve challenged a handful of them to prove otherwise also...none have been able to with the results they’ve sent me

some days they think they have it locked down, as they shoot more to prove it...ain’t worked out

the #1 thing that directly affects my scores and impacts is how close my dope is to to being perfect followed by how well I nail the winds that day...how the rifle grouped prior, is a wash

 

[Dthomas3523]

That has also been tested in a hydraulic return accuracy fixture by someone I know. 0.7 MOA for 30 shots was what they were getting (different bullets, caliber, case prep, etc.. than I use. Not sure on the details there). A friend that shoots BR volunteered 30 rounds to the cause and produced ~.9 MOA group (8.4fps SD). Myself with 12-14lb rifles have been in the 1.0-1.3 range with Kreiger/Bartlein and 1.2-1.5 range with a Wilson blank, 1.2-1.6 with a factory CZ barrel all for 50 shots. The 1.0 MOA Barrel consistently shoots tighter than the other 2 6.5cm barrels. The BR rifle had full OCD case prep done with neck-turn-to-fit cases.

I'm not saying I'm perfect for 50 shots every single time, but the vast majority of the shots that break feel good. I bag everything out and shoot in 10-shot strings with 10-20 minutes between to let the barrel cool and let my eyes rest. Glass is an AMG and it's not hard on my eyes. Nonetheless, there's probably about a .100" diameter zone that I could be aiming in and not really be able to tell the difference. Also bear in mind I'm shooting 50 shots, the people I've talked into trying this out have shot 30.

My interest in starting all of this is first to find out what matters and how much as far as status quo load development dogma, AND to also get a better feel for system capabilities to apply down the road to hit probability. In the latter application the significance of having good data shows through. In other words, convincing myself that I have 15fps ES, 3fps SD, and a .3 MOA rifle does me a disservice by making me overconfident in hit probability. Even using accuracy rest fixture data doesn't tell me anything about reality. Understanding that in any given situation I'm actually drawing from a 1.1 MOA pool of random selections, NOT a 0.5 MOA pool ultimately makes easier my ability to assess shots, and better address how much work and of what type to put into load dev.

Anyway, end of the day if the shooter adds .3-.5 MOA, the shooter adds .3-.5 MOA. I know there are better shooters out there, but I suspect I'm right in line with the better half of them. Nonetheless, I'm happy to have someone prove that I suck, and happy to see sub-MOA 30-50 shot strings from a 'tactical' (non BR/Fclass) rifle if you can do them. If there's a way to get better dispersion, better MV spreads, etc... then I'm all ears because that was the intent of this to begin with. But so far I've tested most (not all) of the status quo methodology ad nauseam and come up with fewer tools than I thought I had at the beginning.

I think this is why a lot of higher echelon F class shooters are finding powder node with chrono and then putting the effort into seating depth tests (And tuner if they run one). They know their loading technique is correct and therefore should produce consistent numbers. Then they get things dialed in as tight as possible with seating depth and adjust as they go (basically as their sample size gets larger, they adjust based on their data).

Meanwhile, us PRS and hunter types are still shooting groups and trying to decipher something from it.

 

[Ledzep]

The biggest new tool I have is discovering that most of the time the standard deviation of the radii of the shots form (more or less) a normal distribution centered on the mean radius. This has led to pretty remarkably accurate predictions of total dispersion, which has in turn led to a much better level of understanding and observation of things like hit % on KYL rack plates.

Also my skills in Excel are much better!

ETA: A barrel tuner is on the list of things I want to play with. Inherently I tend to believe that messing with mode shapes can dampen barrel whip that I assume is what (in part, at least) causes the dispersion.

 

[kthomas]

This discussion is purely academic until a test with proper controls and procedures, eliminating the shooter from the equation is done.

As much as we all like to think we are amazing shooters, there's way to much influence from the shooter to merely say that "I felt I broke the shot well so it must be the ammo and/or barrel".

Right now it's all purely anecdotal evidence and conjecture.

@Ledzep, @morganlamprecht I'm not saying that you guys are wrong. I'm sure there are truths in what you are saying. But I don't trust any testing method that doesn't eliminate the shooter as a variable, as the shooter has too much influence on the results for the data to be 100% pure and meaningful. Even the best of shooters will introduce errors into the shooting through inconsistencies, to varying degrees. No one is 100% consistent behind the rifle, and the more rounds shot, the more potential for shooter error you introduce.

Without isolating the shooter, it's hard to say how much of it is being caused by the shooter. And until then, it's difficult to ascertain whether this is a real problem we should be accounting for, or something merely academic to discuss on bored lonely covid-19 induced nights once we get bored of our FB news feeds.

 

[Dthomas3523]

The biggest new tool I have is discovering that most of the time the standard deviation of the radii of the shots form (more or less) a normal distribution centered on the mean radius. This has led to pretty remarkably accurate predictions of total dispersion, which has in turn led to a much better level of understanding and observation of things like hit % on KYL rack plates.

Also my skills in Excel are much better!

ETA: A barrel tuner is on the list of things I want to play with. Inherently I tend to believe that messing with mode shapes can dampen barrel whip that I assume is what (in part, at least) causes the dispersion.

I am running the first tuner for me. Just in the break in load I’m running, the tuner took individual groups from the .6 area down to .3 average.

That’s for individual groups. Taking into account the dispersion you are describing, probably .7. And that’s not with more than 80 rounds on barrel. So probably only about 20 shots in this actual group from the tuner.

 

[Dthomas3523]

This discussion is purely academic until a test with proper controls and procedures, eliminating the shooter from the equation is done.

As much as we all like to think we are amazing shooters, there's way to much influence from the shooter to merely say that "I felt I broke the shot well so it must be the ammo and/or barrel".

Right now it's all purely anecdotal evidence and conjecture.

@Ledzep, @morganlamprecht I'm not saying that you guys are wrong. I'm sure there are truths in what you are saying. But I don't trust any testing method that doesn't eliminate the shooter as a variable, as the shooter has too much influence on the results for the data to be 100% pure and meaningful. Even the best of shooters will introduce errors into the shooting through inconsistencies, to varying degrees. No one is 100% consistent behind the rifle, and the more rounds shot, the more potential for shooter error you introduce.

Without isolating the shooter, it's hard to say how much of it is being caused by the shooter. And until then, it's difficult to ascertain whether this is a real problem we should be accounting for, or something merely academic to discuss on bored lonely covid-19 induced nights once we get bored of our FB news feeds.

True to a point. I think there is very much valid data if it’s a good shooter and there are enough shots.

Regardless of shooter error, there will be a very consistent group that will show up over time. And there will be outliers that we’re likely the shooter. This will all show up fairly obviously over time.

Also, the shooter is part of the actual system as we don’t shoot PRS matches from a hydraulic.

So, the flip side of the argument would be “so, your group dispersion was .6” over 200 rounds. So what about with a shooter involved.”

And as I’ve hinted at a bit, I think the other disciplines have already discovered what this thread suggests a long time ago.

 

[MadDuner]

I cannot wait to go burn through 50 rounds to see what kind of dispersion pattern I end up with. My first trip out with my new rifle to sight it in resulted in multiple sub-MOA groups of 5 with milsurp ball ammo.... We all know that’s not right. I’m perfectly happy to claim the lucky happy accident of me missing and the rounds all keyholing anyway. I’m Just glad I had witnesses!

But I will go shoot 50 and see where I end up.

 

[PAYDIRT]

 

[kmontang]

Maybe this is a silly thought, but isn't a group with a higher number of shots more likely to be affected environmental variables that have nothing to do with the gun's mechanical accuracy but will lower perceived accuracy? 50 shots have 10x as much chance for changes in wind, barrel temp change, shooter error, etc...

While I don't doubt that more rounds=increased group size due to probability, I also think that in the real world, more rounds=increased group size due to uncontrollable factors. And in the real world, I'm not sure how you'd separate those at a certain point.

 

[Dthomas3523]

Also, damnit.....don’t talk people out of stopping at the 1/2 moa on the kyl racks!!

 

[morganlamprecht]

Pretty much everything in that BR video points more to my point IMO...

they're shooting machines, over flags, and constantly changing things at the bench while shooting paper to maintain their tiny groups...if an f class shooter with a tuner gets to the match and his rifle isn’t shooting how he expected, he’ll start tweaking the tuner...

why would I expect to maintain similar from a shoulder fired rifle on a bipod and squeeze bag or from various positions with 500 rounds of ammo loaded days/weeks/months ahead of time?

I’m not carrying my arbor press out into the hills hunting with me or stage to stage in a match

 

[memilanuk]

While I don't doubt that more rounds=increased group size due to probability, I also think that in the real world, more rounds=increased group size due to uncontrollable factors. And in the real world, I'm not sure how you'd separate those at a certain point.

There's actually a whole field called "design of experiments" aka DoE, parts of which deals with exactly this. It *could* have some interesting applications in developing (and periodically validating) 'robust' loads... but considering that most reloaders run screaming from any math more complex than ES... and most of the rest of us are more interested in shooting than testing... it's not happening any time soon.

The simplified version is "randomize what you can, and block what you can't".

 

[Ledzep]

The best bergers you ever bought aren't good enough for those guys. I remember reading up on what they were shooting once and IIRC it was basically a monopoly of some guy that hand makes bullets to literal perfection.

 

[memilanuk]

they're shooting machines, over flags, and constantly changing things at the bench while shooting paper to maintain their tiny groups...

The interesting bit is that if I recall correctly, most of the smallest individual groups are *not* fired from rail guns like that, but from bags. The *aggs* might be a different story.

if an f class shooter with a tuner gets to the match and his rifle isn’t shooting how he expected, he’ll start tweaking the tuner...

Things have to be pretty bad for most F-class shooters I know to start fussing with the tuner during a big match. At that point you're just trying to put out the dumpster fire.

 

[morganlamprecht]

The interesting bit is that if I recall correctly, most of the smallest individual groups are *not* fired from rail guns like that, but from bags. The *aggs* might be a different story.



Things have to be pretty bad for most F-class shooters I know to start fussing with the tuner during a big match. At that point you're just trying to put out the dumpster fire.

I can’t speak for all of them but the guy who told me, makes one of the more popular tuners...I took his word for it

he said he’ll adjust the tuner and monitor how it affects groups/shape while doing load tests...if he starts seeing something occur during sighters or whatever he’s shooting, he makes adjustments, as necessary, to correct it

 

[memilanuk]

I'm guessing we may know the same guy... I've been using Erik's tuners for years now.

If it's a string with unlimited sighters, that's a viable option. If it's 2 and go for record... then it's a little more like roulette.

I've seen it go well - watched a person recover mid string from a disappointing start (actually that's when I started using them), and I've watched people mind-fuck themselves and instead of paying attention to the range conditions actually causing vertical, they made things worse.

How good your notes are plays into it as well.

 

[BCX]

If the group is solid at 100, and the ES is within the acceptable area, there’s little to no reason a load will fall apart at 1k that didn’t fall apart at 100. If the load work up was done properly and the bullet is a stable/known commodity.

I will give 3 good reasons and this is directed to a group not the load.... The shooter, mirage and wind.

 

[Dthomas3523]

I will give 3 good reasons and this is directed to a group not the load.... The shooter, mirage and wind.

Well, that woudn’t be “falling apart.” That’s just shooting conditions. I’m saying there’s not much reason your load won’t perform properly given the shooter performing properly. I.E. you don’t need to go back to load development.

 

[spife7980]

Yeah things may agg out over a massive sample but I will still trust that those test loads that shoot shitty will continue to shoot shittier than the ones that shot good; so why not go after the ones that shot good instead of just settling at whatever becasue "its all a wash in the end and life is a meaningless pit of despair?"

Id like to see a test where someone does a 50round sample on what the OCW says should be in a node and what is clearly not and compare how the 2 samples compare. Im sure the good group will grow with sample size but I bet its still much better than the other. I shoot groups every weekend but Im not down for 100 marathon rounds in one go like that.


My method? Shoot an ocw at 500 and fine tune it. Sure its small sample sizes during the course of development before I confirm at the end but I can see that a shift up and down in powder isnt moving me off the point of aim which is really what I care about.


Chronos help me identify if something is wrong, it doesnt help me at all in finding out whats right. All my samples Ive ever had an sd of 1-2 on have shot noticeably worse than my good groups with an sd of 8.

 

[2aBaC̶a̶]

The graphs make things look extreme. All the variance from 5-40 shots take place in under .5" or less.

Were really picking the fly shit out of the pepper on this.

I would be curious to see if Velocities followed a similar trend.