Backfire 1 moa hoax video

[mrcheesemoose]

Anybody seen this video?

Came across it the other day, seems like the hunting rifles we’ve all been using aren’t actually as accurate as advertised.

 

[Makinchips208]

Anybody seen this video?

Came across it the other day, seems like the hunting rifles we’ve all been using aren’t actually as accurate as advertised.

Saw it and got a good laugh. Maybe he isn’t a good shot or he is using poor ammo?

Then TPH guys did a podcast about it. They are pretty funny guys.

 

[mrcheesemoose]

Saw it and got a good laugh. Maybe he isn’t a good shot or he is using poor ammo?

Then TPH guys did a podcast about it. They are pretty funny guys.

Did you watch the video? He wasn’t the one shooting

 

[Makinchips208]

Did you watch the video? He wasn’t the one shooting

Yeah I did. As his proof of his theory he says he looked back over all his group sizes and everything averages 1.82 moa.
All that means is he is a 2 moa shooter.
Then in the end he does shoot a group which he says is barely 1 moa.

 

[6.5SH]

He also repeatedly conflates accuracy and precision.
I do think he has a valid point that a cherry picked 3 shot group doesn't equal a consistent standard of precision.

The video showed a lot of really poor shooting fundamentals. A trip to a public range before hunting season will show a lot of the same, most people are not 1 MOA shooters.

 

[mrcheesemoose]

He also repeatedly conflates accuracy and precision.
I do think he has a valid point that a cherry picked 3 shot group doesn't equal a consistent standard of precision.

The video showed a lot of really poor shooting fundamentals. A trip to a public range before hunting season will show a lot of the same, most people are not 1 MOA shooters.

I don’t really buy that all people have bad shooting fundamentals. I mean he was out there for two days getting people to shoot groups with a factory bolt action, and not one could do it. That says a lot more about the quality of the rifles over anything else. Are you saying that not a single person in that group of people is a good shooter? Because that’s a lot less likely then saying the average cheap stock hunting rifle isn’t a true one moa gun.

 

[Emerson0311]

I don’t really buy that all people have bad shooting fundamentals. I mean he was out there for two days getting people to shoot groups with a factory bolt action, and not one could do it. That says a lot more about the quality of the rifles over anything else. Are you saying that not a single person in that group of people is a good shooter? Because that’s a lot less likely then saying the average cheap stock hunting rifle isn’t a true one moa gun.

The vast majority of shooters are not pro level, nor do they perform consistently. That being said, a 10 shot group done by a pro is rarely 1/2MOA.

 

[Andrew863]

I tend to agree with the video. Growing up as a hunter a box of ammo would last 2 years. We would go out before season and draw a 2 or 3 inch circle on a paper plate and when a bullet hit close to center, we were ready for season. I know most hunters in my areas do the same and they see target practice as wasting ammo. Most of the people I see at my private range getting ready for dear season have horrible fundamentals. I have been working with a buddy to improve his shooting that goes out west every year for elk hunts and has for the last 20 years. He can kill animals, but when it comes to shoot a 1in group he can't do it yet.

 

[doubloon]

Bring me your sub MOA rig and let me pick the ammo then we'll see if it holds 1 MOA.

 

[mrcheesemoose]

The vast majority of shooters are not pro level, nor do they perform consistently. That being said, a 10 shot group done by a pro is rarely 1/2MOA.

But we’re also talking about your average cheap factory hunting rifles. These things aren’t exactly built with supreme precision in mind. I don’t think a pro could shoot an average factory hunting rifle to sub moa consistently, because I don’t think the average hunting rifle is sub moa.

 

[Makinchips208]

The vast majority of hunters are terrible shots. Just listen to how many rounds are fired on opening and closing day compared to number of deer harvested.
Even Mr. backfire gut shot a deer.
He talks about cold bore shift like it’s valid, again, Shooter error.
The inaccuracy is the shooter, and possibly the ammo, not the rifle.

 

[1911Shootist]

Didn’t Litz publish a bunch of data showing even with full customs and handloads, once you start looking at 10+ round groups and stop cherry picking 5 round groups (not to mention 3), the vast majority of shooters, including pros and competitors, would only average ~ 0.9 moa?

 

[Haney]

The vast majority of hunters are terrible shots. Just listen to how many rounds are fired on opening and closing day compared to number of deer harvested.
Even Mr. backfire gut shot a deer.
He talks about cold bore shift like it’s valid, again, Shooter error.
The inaccuracy is the shooter, and possibly the ammo, not the rifle.

You are talking out your ass . You are in the vast majority of bullshit artists . Your opinion of local hunting sounds like you don't even hunt .

 

[supercorndogs]

He also repeatedly conflates accuracy and precision.
I do think he has a valid point that a cherry picked 3 shot group doesn't equal a consistent standard of precision.

The video showed a lot of really poor shooting fundamentals. A trip to a public range before hunting season will show a lot of the same, most people are not 1 MOA shooters.

The cherry picked 3 shot group thing isn't exactly news.

 

[supercorndogs]

LOL. His kid popping up behind the rifle like a prairie dog as soon as he shoots. Tighten the swivel on your bipod. He doesn't know how to shoot and if dad taught him and set him up, he must not be too far ahead. The vast majority of people can't shoot. If you haven't figured this out, you are either one of them, or you don't shoot. Many different kinds of ammo are not capable of MOA accuracy. Some ammo that is may not be in your rifle. There is a reason testing different kinds of factory ammo is suggested by everyone who knows their ass from a hole in the ground. I am glad I only wasted 30 seconds watching the intro.

 

[Mr.BR]

Sling shot 75rd group at 300m ,factory ammo no optics.

crossed of shots are sighters in 6x 10 rd groups that overlap to just shy of 1moa

electronic target and paper backer

 

[Newbie2020]

Shoot the Chris Way Kraft drill for a humbling reality check as to how good of shooter you are. Once we get off our belly we pretty much all suck.

The Ruger American uses the same action as the SubMOA RPR. Same for savage, Tikka, etc. Modern factory rifles are decent shooters. And they’re consistent. Don’t get me wrong, I prefer my custom to my savage and my Ruger. And blueprinting a factory action will improve it.

But I agree w the previous post: cheap ammo and bad fundamentals

 

[Evintos]

Didn’t Litz publish a bunch of data showing even with full customs and handloads, once you start looking at 10+ round groups and stop cherry picking 5 round groups (not to mention 3), the vast majority of shooters, including pros and competitors, would only average ~ 0.9 moa?

There's a scale factor for group sizes and he took the scale factor (based off normal distribution curve) and applied it to competition results and group sizes. Depending on the distances, the averages were either sub-MOA or above MOA.

If you were to scale a 3 shot group (actual fired) into a 10 or 30 shot group for example -

1 (MOA) ÷ 1.99 (scale factor extended to 30 shots) = 0.5025 MOA (3 shot)
1 (MOA) ÷ 1.58 (scale factor extended to 10 shots) = 0.6329 MOA (3 shot)

A 3 shot 0.5025 MOA group then equates to a 30 shots, 1 MOA average rifle.
A 3 shot 0.6329 MOA group then equates to a 10 shots, 1 MOA average rifle.

It's also noted (in the chapter) that if you take the average of many 3 shot groups and then apply the scale factor, you'll have a more accurate average (than taking a single 3 shot group).

How it was applied to competition results, example - 600 yard BR results (5 shot group average) was 0.672 MOA. When the scale factor was applied for 10 shot groups (1.24), a hypothetical 0.833 MOA group at 600yds was derived. With a 20 shot group hypothetical (using 1.45 scale factor), the 600yd BR shooters would've (theoretically) averaged 0.974 MOA if they had fired 20 shots. The winner of that particular 600yd BR competition had a 5 shot average of 0.352 MOA, a 10 shot (scale factor) of 0.437 MOA, and a 20 shot scale factor of 0.510 MOA.

Modern Advancements in Long Range Shooting Volume II by Bryan Litz - Chapter 1 Understanding Dispersion

 

[Makinchips208]

You are talking out your ass . You are in the vast majority of bullshit artists . Your opinion of local hunting sounds like you don't even hunt .

Sounds like you also missed a deer on closing day…
You hunt the same areas as I do?

 

[XP1K]

When I used to be a member of a gun club I always loved being out there in the couple weeks leading up to deer season. You always saw some wild stuff lol.

I watched the video. I wish he was closer so I could take his money. Personally, if I end up with a rifle that shoots like that, it doesn't stay around long. Or it ends up as a donor.

 

[R_A_W]

I still don't understand why people watch this guy

 

[Ronws]

I have seen the Backfire video before this thread. Then, again, in this thread.

Oddly enough, I had never heard of Texas Predator Hunting. And their rebuttal podcast was linked in here. I watched both videos. The initial take-down video where they take the piss out of him and then the follow-up video concerning his comments in their first video.

In my experience and that of others far more experienced than I am, the human is the least accurate part of the system. A truly fair test would be to have the gun in a vice or jig that is not able to move. Then see what the groups do.

There is always shooter fatigue, with or without pain. And I think TPH has a point about cold shooter, rather than cold bore. It was illuminating to find that he tried the cold bore zero a number of days in a row and the dispersion was the same as if he had done a group in one sitting. So, the best explanation for later shots to group better than the first shot is not because cold bore is off. In fact, if a hot barrel was supposed to disperse more, then why did Jim get really close shots after the cold bore and then what was obviously a flyer?

I have seen the Hornady podcasts on the size of groups. Episodes 50 and 52. Any rifle will open up in groups the more you shoot in a string.

So, Jayden, the lead ballistician at Hornady said you need to group the rifle for the job. So, a hunting rifle should be grouped no more than 3 shots. Not 5 or 10 because, as I believe and agree with others who believe this, if you cannot hit a deer in 3 shots, you are doing something wrong and need to stop.

As he states in the snippet, you can have a rifle that shoots .25 MOA in the first 3 shots but has a larger disperson at 10 rounds or more.

However, a target contest where you are going to shoot many rounds? Yes, do that. Or, as above, the extension math.

There are some budget rifle companies that make accuracy guarantees. Some do not. Thompson Center Arms offers 1 MOA in the first 3 rounds. And I have seen that happen.

And Mossberg offers no 1 MOA accuracy on their rifles. They simply say that with brand name ammo and good shooting fundamentals, you should get acceptable results. I have also seen good results from their rifles. Similar models of rifle, both using factory quality ammo. The only difference was the shooter.

Personally, I always assume I am the problem. I know when I have pulled a shot or failed to settle.

As for ammo, there can be differences in charge weights and differences in slug weight and mass distribution. But the more you can normalize that, the better.

 

[spamassassin]

When people that don't understand math talk about math, it's time to change the channel. If anyone takes a gun manufacturer's claims about accuracy seriously then they're just an example of why marketing is a trade. Someone has to be a fool for businesses to part fools from their money. There's no shame in it. That's how the world works. In the long long ago fools were very often killed before they could reproduce so foolishness was rightfully looked down on. Now that the economics of the world literally counts on 25% of humanity being actually retarded, there's nothing intrinsically bad about being a retard or even just a fool.

 

[physguy88]

I work in a scientific discipline where we deal with these types of experiments all the time. There are basically three sources of error:

Systematic error:

• Pencil thin barrels on hunting rifles heat up after a short string of shots and this opens up the dispersion of the group size. This is why a three shot group can be a better test than a 5 shot or a 10 shot group. The first three shots are likely to give you performance more like a hunting situation.
• Using the wrong ammo for the particular gun. Gun barrels have harmonics and the nodes in the vibration pattern can line up differently with different ammo loads, so you have the get the right ammo for the gun. A gun with a thin, short barrel is going to be more sensitive to ammo than a gun with a thick, long barrel because the thin short barrel is going to be more flexible and have less complete propellant burn.

Experimenter error:

• All of the bad shooting posture, etc, that people pointed out on the video

Statistical error:

• This gets to the heart of the problem that I think Backfire's video was meant to address. A 1 MOA guarantee, as typically worded by the manufacturers, is mathematically meaningless. In practice it has a bit more value but still has a lot of fudge factor.
• There are a lot of random factors in guns. Let's say I took out all the systematic error, and took out shooter error completely, the bullets coming out of the gun will still land in a random pattern around a central point.
• Because we are talking about random events, a proper definition of precision needs to give a confidence interval, e.g. 95%, 75%, 50%, etc.
• If I specify that my gun has a 95% confidence interval for 1 MOA precision using match grade ammo that works well with the gun, what does that mean?
  • That means that if I used perfect technique to shoot a 3-shot group, waited for the barrel to cool, shot another 3-shot group, waited, and repeated this over and over until I shot 100 bullets, 95 bullets will land in a 1 MOA circle.
  • For any individual 3-shot group, I have a 95% * 95% * 95% = 85.7% chance of getting sub 1 MOA
• So how good a 1 MOA guarantee from a manufacturer is really depends on their testing method.​
  • Let's say that the manufacturer's internal procedure is that the moment the rifle is clamped to a bench rest and shot by their staff using approved ammo, it has to have a 3-shot group under 1 MOA. That's probably close to a 95% confidence interval 1 MOA.​
  • Let's say another manufacturer allows testing five 3-shot groups, and the gun is considered a pass if any one of those five groups end up under 1 MOA. That's a much less stringent test. In fact, you could have a 60% confidence interval gun, shoot 5 groups of 3-shots each, and chances are one of those 3-shot groups will be under 1 MOA. What this means though is that for this second gun, 40 out of 100 shots will land outside of 1 MOA, purely based on the quality of the gun.​
  • Evidently, not all 1 MOA guarantees are created equal.​

The above isn't really news to most people here. Just summarizing. Manufacturers do advertising and are not always 100% transparent (what a concept). I think the Backfire video has a point, but it's not news. At the end of the day the limiting factor in most practical situations is still the shooter.

 

[mrcheesemoose]

I work in a scientific discipline where we deal with these types of experiments all the time. There are basically three sources of error:

Systematic error:

• Pencil thin barrels on hunting rifles heat up after a short string of shots and this opens up the dispersion of the group size. This is why a three shot group can be a better test than a 5 shot or a 10 shot group. The first three shots are likely to give you performance more like a hunting situation.
• Using the wrong ammo for the particular gun. Gun barrels have harmonics and the nodes in the vibration pattern can line up differently with different ammo loads, so you have the get the right ammo for the gun. A gun with a thin, short barrel is going to be more sensitive to ammo than a gun with a thick, long barrel because the thin barreled is going to be more flexible and have less complete propellant burn.

Experimenter error:

• All of the bad shooting posture, etc, that people pointed out on the video

Statistical error:

• This gets to the heart of the problem that I think Backfire's video was meant to address. A 1 MOA guarantee, as typically worded by the manufacturers, is mathematically meaningless. In practice it has a bit more value but still has a lot of fudge factor.
• There are a lot of random factors in guns. Let's say I took out all the systematic error, and took out shooter error completely, the bullets coming out of the gun will still land in a random pattern around a central point.
• Because we are talking about random events, a proper definition of precision needs to give a confidence interval, e.g. 95%, 75%, 50%, etc.
• If I specify that my gun has a 95% confidence interval for 1 MOA precision using match grade ammo that works well with the gun, what does that mean?
  • That means that if I used perfect technique to shoot a 3-shot group, waited for the barrel to cool, shot another 3-shot group, waited, and repeated this over and over until I shot 100 bullets, 95 bullets will land in a 1 MOA circle.
  • For any individual 3-shot group, I have a 95% * 95% * 95% = 85.7% chance of getting sub 1 MOA
• So how good a 1 MOA guarantee from a manufacturer is really depends on their testing method.​
  • Let's say that the manufacturer's internal procedure is that the moment the rifle is clamped to a bench rest and shot by their staff using approved ammo, it has to have a 3-shot group under 1 MOA. That's probably close to a 95% confidence interval 1 MOA.​
  • Let's say another manufacturer allows testing five 3-shot groups, and the gun is considered a pass if any one of those five groups end up under 1 MOA. That's a much less stringent test. In fact, you could have a 60% confidence interval gun, shoot 5 groups of 3-shots each, and chances are one of those 3-shot groups will be under 1 MOA. What this means though is that for this second gun, 40 out of 100 shots will land outside of 1 MOA, purely based on the quality of the gun.​
  • Evidently, not all 1 MOA guarantees are created equal.​

The above isn't really news to most people here. Just summarizing. Manufacturers do advertising and are not always 100% transparent (what a concept). I think the Backfire video has a point, but it's not news. At the end of the day the limiting factor in most practical situations is still the shooter.

That’s another good point. What is your definition of a “1 moa gun”? And also you could definitely eliminate the shooter error by shooting from a tied down rest, which backfire didn’t do. But that being said, I highly doubt that most hunting rifles are what I consider a “true 1 moa” gun. I mean we’re talking companies that are motivated to not be completely honest about a gun being true moa. We’re also talking about some of the cheapest guns in the market. A lot of military rifles, which have nicer barrels in many cases, aren’t considered to be 1 moa guns by the military. So why should a cheaper gun be considered better? You can say that the reason for people shooting bad groups all day every day is the person, but they are only one part of the system, and there are so many other pieces that could cause issues. Why ignore the second most important part of accuracy, and just write it off as something else?

 

[acudaowner]

I'd blame the shooters over blaming the rifle . take out the shooter which he never did in the video .

 

[DeathBeforeDismount]

Shoot the Chris Way Kraft drill for a humbling reality check as to how good of shooter you are. Once we get off our belly we pretty much all suck.

The Ruger American uses the same action as the SubMOA RPR. Same for savage, Tikka, etc. Modern factory rifles are decent shooters. And they’re consistent. Don’t get me wrong, I prefer my custom to my savage and my Ruger. And blueprinting a factory action will improve it.

But I agree w the previous post: cheap ammo and bad fundamentals

It's easier to shoot modified prone or sitting than prone IMO, or atleast for me. Laying prone is fatiguing and leads to poor focus and groups.

The actions have very little to do with accuracy. Blueprints an action is more about feel and reliability than accuracy.

Screwing a good barrel on is how you get great groups, along with good ammo.

Every tikka will shoot sub moa with good match ammo, same with sako and a few others. A tikka ctr using fgmm or hornady match will shoot sub moa.

Finding a bunch of fuddss who can't shoot is not representing of the equipment and how good it is off the shelf today.

 

[noremorse92]

When people that don't understand math talk about math, it's time to change the channel. If anyone takes a gun manufacturer's claims about accuracy seriously then they're just an example of why marketing is a trade. Someone has to be a fool for businesses to part fools from their money. There's no shame in it. That's how the world works. In the long long ago fools were very often killed before they could reproduce so foolishness was rightfully looked down on. Now that the economics of the world literally counts on 25% of humanity being actually retarded, there's nothing intrinsically bad about being a retard or even just a fool.

Underrated comment here

 

[DocRDS]

I work in a scientific discipline where we deal with these types of experiments all the time. There are basically three sources of error:

Systematic error:

• Pencil thin barrels on hunting rifles heat up after a short string of shots and this opens up the dispersion of the group size. This is why a three shot group can be a better test than a 5 shot or a 10 shot group. The first three shots are likely to give you performance more like a hunting situation.
• Using the wrong ammo for the particular gun. Gun barrels have harmonics and the nodes in the vibration pattern can line up differently with different ammo loads, so you have the get the right ammo for the gun. A gun with a thin, short barrel is going to be more sensitive to ammo than a gun with a thick, long barrel because the thin short barrel is going to be more flexible and have less complete propellant burn.

Experimenter error:

• All of the bad shooting posture, etc, that people pointed out on the video

Statistical error:

• This gets to the heart of the problem that I think Backfire's video was meant to address. A 1 MOA guarantee, as typically worded by the manufacturers, is mathematically meaningless. In practice it has a bit more value but still has a lot of fudge factor.
• There are a lot of random factors in guns. Let's say I took out all the systematic error, and took out shooter error completely, the bullets coming out of the gun will still land in a random pattern around a central point.
• Because we are talking about random events, a proper definition of precision needs to give a confidence interval, e.g. 95%, 75%, 50%, etc.
• If I specify that my gun has a 95% confidence interval for 1 MOA precision using match grade ammo that works well with the gun, what does that mean?
  • That means that if I used perfect technique to shoot a 3-shot group, waited for the barrel to cool, shot another 3-shot group, waited, and repeated this over and over until I shot 100 bullets, 95 bullets will land in a 1 MOA circle.
  • For any individual 3-shot group, I have a 95% * 95% * 95% = 85.7% chance of getting sub 1 MOA
• So how good a 1 MOA guarantee from a manufacturer is really depends on their testing method.​
  • Let's say that the manufacturer's internal procedure is that the moment the rifle is clamped to a bench rest and shot by their staff using approved ammo, it has to have a 3-shot group under 1 MOA. That's probably close to a 95% confidence interval 1 MOA.​
  • Let's say another manufacturer allows testing five 3-shot groups, and the gun is considered a pass if any one of those five groups end up under 1 MOA. That's a much less stringent test. In fact, you could have a 60% confidence interval gun, shoot 5 groups of 3-shots each, and chances are one of those 3-shot groups will be under 1 MOA. What this means though is that for this second gun, 40 out of 100 shots will land outside of 1 MOA, purely based on the quality of the gun.​
  • Evidently, not all 1 MOA guarantees are created equal.​

The above isn't really news to most people here. Just summarizing. Manufacturers do advertising and are not always 100% transparent (what a concept). I think the Backfire video has a point, but it's not news. At the end of the day the limiting factor in most practical situations is still the shooter.

AS a statistician/scientist, when one of my students writes anything about "errors" I mark it zero. If you know there is an error, fix it.

Now I know you mean variation/variability but just wanted to sharpshoot you and whip out my dick.

Sorry i just got done grading and was in a bad mood because even statistics students can be really stooooooooooooooopid.

And a reminder those litz factors are "average" factors. You want a 10 shot group? Shoot 10 shots. When I report my numbers, number of samples is key (as are confidence intervals, no one gives a shit about that here). Don't use those factors. Shoot the group. People don't understand the point he was trying to make and use them to "estimate" groups.

Don't estimate.

Measure.

 

[Ronws]

Next question, as a matter of interest. If you have a high speed rig, heavy bullet, fast MV, fast twist, something that causes more throat erosion, how much sampling can you live with in a limited barrel life?

The cheapie .308 W might be good for 5,000 rounds. Not so for maybe a tuned up 7 RM or even a .300 PRC.

I totally agree. If you want a 100 round sample, shoot 100 rounds. Or 10 or, whatever.

Then, to also address the statement of Jayden at Hornady. You could have a rifle that, given some consistent ammunition, shot less than 1 MOA on the first 3 shots, but widened out as the sample group size increases. What is best for what? If your round count is expected to be 20, 50, 100, then should that not be the goal. And then whatever you have is what you get because of limits on barrel life. Or at least usable barrel life. An older gun that has widened out to 1 MOA on the first three shots might still be good for hunting but not for competition.

I don't know. I may have ideas and opinions and like all buttholes, they stink.

 

[FredHammer]

Personal gripe - Backfire dude pegs my annoyance meter pretty quick. No fault of his, but his mannerisms and speech pattern remind me of mamby pamby 16-18yr old girl teenagers. Dude is a millionaire brainiac though. Fuck, I'm doing it wrong I guess!

Video test - application of the proper fundamentals of marksmanship would shrink those groups considerably. Picking a hunting rifle off the wall and arbitrarily shooting store bought ammo is not a recipe for small group success. 30yrs ago, before handloading, I always bought 5 boxes of various ammo and stocked up on what it shot best. Following this method of factory rifle and ammo would continue to shrink those groups. It's common knowledge some bullet/powder/twist rate combos work better than others.

Propper application of expectation - I had a pencil barrel Kimber 84 Hunter 708 that would always print the first two shots 1/8". Third shot .5" and past that 1.5" and more. It did that every range session where I shot more than two rounds, like clockwork. (@Ronws nailed it with hunting rifle expectation above, I've had a lot of coffee and want to expound on that.)
For a 2 shot hunting gun I was cooking with gas with those 1/8" groupings. Contrast this to MTU competition barrels and 2 shots don't mean shit, they better hold 20 shots MOA. (Best case scenarios) 6.5lbs mountain hunting gun vs 25lb belly gun. Hmmm, not much can translate between them. "Hammer, meet wrench."

 

[DocRDS]

warning: wall of text incoming

So I was musing on this. Earlier I was just yanking @physguy88 's chain, because what he said was essentially spot on, terminology aside. Having been through process control hell in several companies and even being the dreaded "black belt" of Six Sigma and teaching data science at a D1 school.

Back in the day I had a piece of equipment and there was a 'Spec' for its performance. This had VERY LITTLE to do with equipment, it was what was required for the product to function.

So my specialty was metals. Let's say my job was to deposit Nickel. My spec says that my layer of Ni has to be 100 +/- 10 units. In a perfect world this is my "six sigma" in other words the process I need only has a 1 in a million chance (roughly--long story short it really only 4.5 sigma) of producing a wafer where the thickness is 110 or greater or 90 or less. That means its 6 Standard Deviations in 10 units. Or in other words I have to run my equipment at 100 +/- 1.67 (1.67 being the STD DEV) or less units. to hit this specification.

BUT WAIT there is more! When I put my nickel down, its not 100 units everywhere. Some places it 99 and some places its 101. I'm working in semiconductors so I have a "within wafer" variation. So while the mean thickness may be 100, in some places its more, some places its less, so I have to take that into account as well! Typically I measured up to 49 points to get the mean and std dev within the run.

But behind Door #3 is a third problem:
Every time I run my machine its "destructive". It changes the output of my machine. If I run today and get 100 (mean), the next run I might get 101. Sometimes this is random, sometimes its systematic. In my case there was a large percentage systematic change, that we could account for. So now I have Wafer to Wafer Variation: some of which I can account for. I won't be giving any secrets away here but I had to adjust the height between my wafer and the "metal" being deposited every run. It was tiny, but something we could account for. Here's the fun part--this wafer to wafer (run to run) variation also effected the WITHIN run variation. (It got bigger if out of adjustment).

When I brought in a new piece of equipment to "qualify" for process matching it took us a minimum of 4 "runs" (wafers) if we were good and up to 6 if we were not.

Sidebar: Wafer to Wafer and Within Wafer variances add. (Variance is std dev squared) to that 1.67 is my budget for BOTH within and run to run

So who gives to shits about this and how does it affect accuracy.

Whether you fire 10 shot groups or 3 shot groups, there is a "shot to shot" variation. And then there is a "Session to session" variation. (or string to string). Along with this, we know from our OCD reloader friends that the "test" is destructive--every time you shoot, you put small wear and tear on the barrel that will effect both shot to shot and string to string. Its been a long time since I've done that version of statistics, but its pretty standard, but the long and short of it is:

You need 4 to 6 STRINGS of shots to "match" accuracy. (In other words, My rifle B matches the accuracy of Rifle A).
To ESTABLISH accuracy it would take around 40 Strings before we could call it "established" (40 points)

The long and short of it: 3 shot groups and accuracy guarantees are BULLSHIT. Even a 10 shot group is marginal. And if I want to "Guarantee" something I need a defect level. the "Six Sigma" is based on 3.4 per million. Now for a lower volume business like firearms, you may be ok with say 1 in a 1000 defects or 1 in 10,000 defects. And that means the AVERAGE accuracy is acutally going to be LESS than 1 MOA because the 1 in a million goes over 1MOA

But remember my whole "matching" comment earlier. You need at least 4 measurements (and usually more) to "match" a process and it takes upwards of 8 for a "disqualify/reject"

So yes, 3 shot guarantee is just "luck" If someone came to me with a statistical process and an N=3 (number of trials) I'd laugh their ass outta my office. If I tried that during process control days, I'd be fired. With N=10, you might just get a stern talking to.

Here's a test which one of these shoots better:

Hint: Its the same distribution and deviation (I just ran them back to back)

 

[DocRDS]

If you think I'm bullshiting

 

[mrcheesemoose]

If you think I'm bullshiting
View attachment 8225172

Would be helpful if you fixed the scale on the figures. If you don’t look at the hash marks on the bottom, or side those groups look pretty similar.

 

[Ronws]

Lies, damned lies, statistics. Did I get the saying right this time?

So, yeah, I get the gist of the above. You have variation in shot to shot. And variation in string to string. Kind of like Jimmy Page in concert. He never played the solo in "Stairway to Heaven" the same way twice. Which is funny because he started out as a studio musician and playing things over and over again before joining the Yardbirds.

A 3 shot group is not going to guarantee the next 3 shot or whatever shots. And each run is destructive. The barrel has a life. Although, something I read suggested that while getting rid of carbon but keeping copper fouling, the biggest problem is throat erosion. So, some have started out with a longer barrel and when it shoots out, they cut off 2 inches and re-chamber. But you are giving up speed when you do that. So, the advisability of that is suspect. Because you need the rifle to do a specific job. So, then, the plan is to simply replace barrels.

However, if the barrel life is aging and each string and each shot to shot cannot be depended on to be constant, are we back to luck?

I know am drifting from the original OP in the debate of whether or not off-the-shelf hunting rifles can be 1 MOA or less or the best to hope for is 2 MOA except when Jim Harmer is firing. Turns out that being a lawyer is good for accuracy, I guess.

 

[Makinchips208]

warning: wall of text incoming

So I was musing on this. Earlier I was just yanking @physguy88 's chain, because what he said was essentially spot on, terminology aside. Having been through process control hell in several companies and even being the dreaded "black belt" of Six Sigma and teaching data science at a D1 school.

Back in the day I had a piece of equipment and there was a 'Spec' for its performance. This had VERY LITTLE to do with equipment, it was what was required for the product to function.

So my specialty was metals. Let's say my job was to deposit Nickel. My spec says that my layer of Ni has to be 100 +/- 10 units. In a perfect world this is my "six sigma" in other words the process I need only has a 1 in a million chance (roughly--long story short it really only 4.5 sigma) of producing a wafer where the thickness is 110 or greater or 90 or less. That means its 6 Standard Deviations in 10 units. Or in other words I have to run my equipment at 100 +/- 1.67 (1.67 being the STD DEV) or less units. to hit this specification.

BUT WAIT there is more! When I put my nickel down, its not 100 units everywhere. Some places it 99 and some places its 101. I'm working in semiconductors so I have a "within wafer" variation. So while the mean thickness may be 100, in some places its more, some places its less, so I have to take that into account as well! Typically I measured up to 49 points to get the mean and std dev within the run.

But behind Door #3 is a third problem:
Every time I run my machine its "destructive". It changes the output of my machine. If I run today and get 100 (mean), the next run I might get 101. Sometimes this is random, sometimes its systematic. In my case there was a large percentage systematic change, that we could account for. So now I have Wafer to Wafer Variation: some of which I can account for. I won't be giving any secrets away here but I had to adjust the height between my wafer and the "metal" being deposited every run. It was tiny, but something we could account for. Here's the fun part--this wafer to wafer (run to run) variation also effected the WITHIN run variation. (It got bigger if out of adjustment).

When I brought in a new piece of equipment to "qualify" for process matching it took us a minimum of 4 "runs" (wafers) if we were good and up to 6 if we were not.

Sidebar: Wafer to Wafer and Within Wafer variances add. (Variance is std dev squared) to that 1.67 is my budget for BOTH within and run to run

So who gives to shits about this and how does it affect accuracy.

Whether you fire 10 shot groups or 3 shot groups, there is a "shot to shot" variation. And then there is a "Session to session" variation. (or string to string). Along with this, we know from our OCD reloader friends that the "test" is destructive--every time you shoot, you put small wear and tear on the barrel that will effect both shot to shot and string to string. Its been a long time since I've done that version of statistics, but its pretty standard, but the long and short of it is:

You need 4 to 6 STRINGS of shots to "match" accuracy. (In other words, My rifle B matches the accuracy of Rifle A).
To ESTABLISH accuracy it would take around 40 Strings before we could call it "established" (40 points)

The long and short of it: 3 shot groups and accuracy guarantees are BULLSHIT. Even a 10 shot group is marginal. And if I want to "Guarantee" something I need a defect level. the "Six Sigma" is based on 3.4 per million. Now for a lower volume business like firearms, you may be ok with say 1 in a 1000 defects or 1 in 10,000 defects. And that means the AVERAGE accuracy is acutally going to be LESS than 1 MOA because the 1 in a million goes over 1MOA

But remember my whole "matching" comment earlier. You need at least 4 measurements (and usually more) to "match" a process and it takes upwards of 8 for a "disqualify/reject"

So yes, 3 shot guarantee is just "luck" If someone came to me with a statistical process and an N=3 (number of trials) I'd laugh their ass outta my office. If I tried that during process control days, I'd be fired. With N=10, you might just get a stern talking to.

Here's a test which one of these shoots better:
View attachment 8225151
View attachment 8225157

Hint: Its the same distribution and deviation (I just ran them back to back)

Wafers? Now I’m craving Nilla Wafers!

 

[spife7980]

 

[DocRDS]

Would be helpful if you fixed the scale on the figures. If you don’t look at the hash marks on the bottom, or side those groups look pretty similar.

I know this one is new, but RNG is kinda a PITA.

Demo on how this is the same distro 1000 points each (with approved hashmarks!)

 

[FredHammer]

View attachment 8225337

I know this one is new, but RNG is kinda a PITA.

Demo on how this is the same distro 1000 points each (with approved hashmarks!)
View attachment 8225338

 

[Ronws]

Slightly getting back to the OP, dispersion increase aside, Jim at Backfire was saying that 1 MOA or accuracy guarantees are a hoax or misleading. Some budget friendly hunting rifles do offer a 1 MOA on the first 3 shots. Others do not. But still provide something close to that in practice.

So, Jim had a challenge of so many shots that measure at 1 MOA or less. And for this, I am going to be a whiny ass karen and insist that the shots fall inside of 1.047 inches, not 1 inch. TPH did offer the suggestion that they might do a video on that. I wish they would. Being at the SH has infected me. Show me paper or it does not exist.

I can prove my cheap ass rifle with the 1 MOA guarantee did meet that in the first two shots. And the analysis was not done by me. .54 inches from first shot to second shot on a rifle fresh out of the box using cheap ammo that turned out to not be the best ammo and in a factory synthetic stock (though it had aluminum pillar bedding, which helped a lot.) But that was one rifle on one day. And I got roasted like a pork shoulder because I had measured with my Stanley tape measure and someone else used Ballistic-X.

So, shit rifle, shit round (.308 W), and a shit optic (the one that everyone here loves to hate.) But I am no one, I am a nothing. Per my first post in this forum, ever, I will be the stupidest person here and, so far, I have not failed to fulfill that promise.

I think TPH should do a series where they shoot some "hunting" rifles for 5 or 10 shots and measure the disperson. Setting aside the statistical analysis. Just follow up on the challenge to Jim's claim.

Maybe Jim will eat his words. He did, a little bit, when he received a second rifle from Mossberg that did do better than the first one he had. He still does not like Mossberg and still loves Ruger uber alles.

 

[FredHammer]

Slightly getting back to the OP, dispersion increase aside, Jim at Backfire was saying that 1 MOA or accuracy guarantees are a hoax or misleading. Some budget friendly hunting rifles do offer a 1 MOA on the first 3 shots. Others do not. But still provide something close to that in practice.

So, Jim had a challenge of so many shots that measure at 1 MOA or less. And for this, I am going to be a whiny ass karen and insist that the shots fall inside of 1.047 inches, not 1 inch. TPH did offer the suggestion that they might do a video on that. I wish they would. Being at the SH has infected me. Show me paper or it does not exist.

I can prove my cheap ass rifle with the 1 MOA guarantee did meet that in the first two shots. And the analysis was not done by me. .54 inches from first shot to second shot on a rifle fresh out of the box using cheap ammo that turned out to not be the best ammo and in a factory synthetic stock (though it had aluminum pillar bedding, which helped a lot.) But that was one rifle on one day. And I got roasted like a pork shoulder because I had measured with my Stanley tape measure and someone else used Ballistic-X.

So, shit rifle, shit round (.308 W), and a shit optic (the one that everyone here loves to hate.) But I am no one, I am a nothing. Per my first post in this forum, ever, I will be the stupidest person here and, so far, I have not failed to fulfill that promise.

I think TPH should do a series where they shoot some "hunting" rifles for 5 or 10 shots and measure the disperson. Setting aside the statistical analysis. Just follow up on the challenge to Jim's claim.

Maybe Jim will eat his words. He did, a little bit, when he received a second rifle from Mossberg that did do better than the first one he had. He still does not like Mossberg and still loves Ruger uber alles.

Ain't no use for language like that!

 

[DocRDS]

They (Guarantees) are 'junk'

But that does;t mean a factory hunting rifle can't produce good groups. I'd be onboard with a "jig" of some sort that bolts down the rifle in a sled or something and fires a group to see what the "mechanical accuracy" is. I bet we'd be pleasantly surprised.

I've got my ol 300 WM Rem circa 1998 ish, penicl barrel, Nikkon scope. I need some range time. lesse if I can hold 1 inch @ 100 over multiple groups. Project for the fall...

 

[Ronws]

Ain't no use for language like that!

 

[mrcheesemoose]

Slightly getting back to the OP, dispersion increase aside, Jim at Backfire was saying that 1 MOA or accuracy guarantees are a hoax or misleading. Some budget friendly hunting rifles do offer a 1 MOA on the first 3 shots. Others do not. But still provide something close to that in practice.

So, Jim had a challenge of so many shots that measure at 1 MOA or less. And for this, I am going to be a whiny ass karen and insist that the shots fall inside of 1.047 inches, not 1 inch. TPH did offer the suggestion that they might do a video on that. I wish they would. Being at the SH has infected me. Show me paper or it does not exist.

I can prove my cheap ass rifle with the 1 MOA guarantee did meet that in the first two shots. And the analysis was not done by me. .54 inches from first shot to second shot on a rifle fresh out of the box using cheap ammo that turned out to not be the best ammo and in a factory synthetic stock (though it had aluminum pillar bedding, which helped a lot.) But that was one rifle on one day. And I got roasted like a pork shoulder because I had measured with my Stanley tape measure and someone else used Ballistic-X.

So, shit rifle, shit round (.308 W), and a shit optic (the one that everyone here loves to hate.) But I am no one, I am a nothing. Per my first post in this forum, ever, I will be the stupidest person here and, so far, I have not failed to fulfill that promise.

I think TPH should do a series where they shoot some "hunting" rifles for 5 or 10 shots and measure the disperson. Setting aside the statistical analysis. Just follow up on the challenge to Jim's claim.

Maybe Jim will eat his words. He did, a little bit, when he received a second rifle from Mossberg that did do better than the first one he had. He still does not like Mossberg and still loves Ruger uber alles.

I feel the same. Show me paper or it doesn’t exist. Honestly I want to see something much more comprehensive then backfires two day experiment from either side. If TPH really wanted to disprove Jim, they would get a couple hunting rifles. And fire a bunch of five round groups out of each one with the best ammo they could find off a lead sled. (By a bunch I mean like 10 or 20 or more) If they can prove that all, or at least the majority, of those five round groups were sub moa, I would be convinced that an average hunting rifle is sub moa most of the time. If Jim wanted to prove his point, he would probably run the same or a similar experience, then show that the majority, or all of the five round groups fell outside of sub moa. No paper, no proof. Anything else is just a screaming match.

 

[FredHammer]

I feel the same. Show me paper or it doesn’t exist. Honestly I want to see something much more comprehensive then backfires two day experiment from either side. If TPH really wanted to disprove Jim, they would get a couple hunting rifles. And fire a bunch of five round groups out of each one with the best ammo they could find off a lead sled. (By a bunch I mean like 10 or 20 or more) If they can prove that all, or at least the majority, of those five round groups were sub moa, I would be convinced that an average hunting rifle is sub moa most of the time. If Jim wanted to prove his point, he would probably run the same or a similar experience, then show that the majority, or all of the five round groups fell outside of sub moa. No paper, no proof. Anything else is just a screaming match.

Jim is just a Youtube popularity contest whore.

 

[DJL2]

This was hashed out on Reddit a while back. I guess I've got a couple observations:
1. 3 shot, 1 MOA guarantees are worthless... always have been, always will be.
2. Did you watch the shooters tossing 5+ MOA groups that weren't even on paper? Technique matters.

To the first point - I had a factory barrel that was garbage tier... and the first 3 shot groups it ever fired (zeroing) were "sub MOA." Most of the groups it fired, regardless of number of shots, were not sub-MOA or anywhere close to it. It was a 2 MOA gun, being generous, and it would reliably meet a 1 MOA guarantee. That's just the nature of random distribution for a 3 shot group.

To the second point - recoil management matters, and the lighter the rifle the more important it becomes. You can see the article on the front page "How heavy is too heavy - rifle weight" for a more nuanced discussion on that point, along with a bunch of other articles written by Frank, Marc Taylor, Litz and others on the topic.

I can't really watch most of the videos on his channel... his delivery/style just doesn't jive for me. That's true of a lot of firearms related content, honestly. If you wanna do goofy shit for likes, you gotta go 9-hole or Garand Thumb. ;-)

 

[Triple D]

He’s a YouTuber…he set out to make a controversial video from the start. It’s about clicks and views, not accurate content.

The data observed from fudds shooting is worthless. Even “shooters” often can’t shoot sub moa. At my local gun club, I’d guess that less than 20% of the shooters can consistently shoot sub moa. If they can hit a 3” target at 100yds, they think are doing good, and shooting well enough for deer season.

Ammo choice matters. A custom 223 that can shoot 3/8moa with match ammo is going to suck with xm193.

 

[Ronws]

To the second point - recoil management matters, and the lighter the rifle the more important it becomes. You can see the article on the front page "How heavy is too heavy - rifle weight" for a more nuanced discussion on that point, along with a bunch of other articles written by Frank, Marc Taylor, Litz and others on the topic.

Exactly. Imagine the ouch of the T3x Superlite in .300 WM. Heavier recoil impulse on a very light weight rifle? And how do you get good with it when you will hurt unless you do something about recoil?

 

[FredHammer]

Exactly. Imagine the ouch of the T3x Superlite in .300 WM. Heavier recoil impulse on a very light weight rifle? And how do you get good with it when you will hurt unless you do something about recoil?

Not even a brake? That’s a brutality recipe right ther, yassir!

 

[Ronws]

Not even a brake? That’s a brutality recipe right ther, yassir!

Exactly. No suppressor or brake. With that dinky little thing they call a recoil pad.