Are 5 shot groups worthless statistically?

[Treefrog]

That part is not binary. It is a calculation of degree, a percentage. You might compute the exact center of a collection of shots but when you predict where the next shot will land you hedge your bets. "It will land at X,Y within +/- "delta" with a X% probability."

Actually. Hit or miss IS binary. Two possibilties there. Hit - 1, Miss - 2. I actually did the arithmatic. (Credit to Mrs. Steeleman, my first grade teacher.)

And this leads me to ask a question that I often ask myself. What exactly is the purpose of this exercise?

J~

 

[elh0102]

I tried to watch most of the video, but I found it pointless. No, not right, it isn't pointless, it just doesn't address the question originally asked, "why doesn't the bullet go through the same hole every time?" Math and statistics may suggest future performance based on results obtained with a host of variables. But if you are a competitive BR shooter, that information is not helpful. You know the variables involved in constructing your ammo and in shooting conditions at the bench. You have to load and test ammo to make your choices. If load A is 10% more likely to shoot .25 MOA as compared to load B, I don't care and don't need to know. I'm going to shoot load A. The exercises described may be of academic interest to a scientist, and that's great, but it's not a short cut to developing the best load. The answer to the question asked is finally answered in the video. Many variables are involved that affect where the bullet goes. Some examples: bullet (many types and weights), powder type and amount, primer brand and type, brass, neck thickness, primer pocket dimension, bullet seating depth, neck tension (based on sizing bushing used), etc. And these variables are in play with the environment, temperature can change the whole recipe. The purpose of the exercise shown in the video, IMO, is academic in nature. It may be very useful if one is wishing to compile shooting statistics based on the performance of a plethora of variables, but I see little proactive value in developing shooting performance, or in choosing the number of shots you wish to shoot in your groups. But then my wife says I'm always wrong, and I cannot say the same about her.

 

[ctshooter]

I prefer to go with 10 shot groups myself. Better portrayal of the rifles consistency. But that still doesn't equate to statistical analysis. I didn't see it discussed yet but from a purely statistical standpoint in a production or manufacturing environment, you do not have usable statistical information until you have 20 consecutive datapoints. Many of you probably already knew this but I offer this for those that don't.
I see ctshooter must know as he mentioned 20!!
Thx
Dan

Dan - you give me much more credit that I deserve as I just picked 20 out of a hat.

 

[Hateful McNasty]

I guess to maintain consistency.

I dont do scores but i like to keep it all together. Like i say gun , pellet, hold consistency. Can i hold my mouth just right. Trigger pull and follow through where i lack and tend to get hold lazy.. true the pressure is on his f going for ten agter a good 6 .

 

[beerthief]

I think that when you move from 5 shots groups to 10 it becomes more of a test of the shooter ( it is for me at least )

when I go past 5 I start to get more and more nervous about keeping the group the same size.

just a personal though/experience,.

you need to relax , shoot every shot the same . Only think of the shot you are taking now . try it . if you think of the last shot stop start over .

 

[drpietrzak]

These are interesting discussions.

1. Statistics don't lie, they are used by people to lie. It is just a tool or device people who lie may use. Sound like a familiar point?

2. There does not appear to be a consistent hypothesis being tested here so any means to "know" what statistics might apply are left unclear. Differences? Predictions? If so, consistency metrics are descriptive stats not differeces or predictive stats. The utility of consistency stats (descriptive stats) varies by context. They are not the end of the journey but a precursor for the journey. Differences or predictions are not tested by consistency stats. Now consistency establishes the potential for predictive stats or differences stats to be applied. As such they have a role in understanding the error band in predictive or differences applications.

3. I have not, to my recollection, ever seen a predictive or differences statistical testing approach applied to group sizes. I sure it has been done at JPL or at a grad program somewhere, but rarely so I would suspect. Eley, Lapua and others set pseudo-arbitrary groups size bounds when ammo grading but do not publish statistical tests of differences. They do run them that I am aware of at all.

4. There is a lot of incorrect use of the term statistical significance here. Especially when there have been no statistical test output being provided. It appears the phrase is being used with a common language definition not research definition.

5. A sample size is not the same as statistical testing and is not a statistical test. It is really just an estimate about the representative number of trials needed, if there is a perfect sampling method being used. Sampling is not really a statistical test. It, like consistency, might be considered a descriptive stat but it usually not used that way in research dialog. Sampling provides a foundation to tell us how big a difference is needed to detect a real difference. Or, how strong a predictive value will have as to be to be be useful. Type 1 and 2 error. It interacts with consistency.

In sum: In reality we really only care about practical utility in this instance. This red herring of statistical tests is fog that has little practical value here. The question is does this work for my application? Essentially the application is 1 shot. The next one.

 

[Emu]

Human factor is always important when shooting.

It is easier to keep concentration on less shots before getting tired from both, body posture and eye focus.

 

[dens228]

I wish I had that problem.

Instead I have heaps of slugs that don’t shoot for crap. Statistically speaking.

I had numerous slugs that didn't shoot well consistently. Funny thing is I have no real need to shoot slugs. I do it just because I can.

 

[Hateful McNasty]

Without a scope or spotting scope i guess the most shots i do per bulls is less hoofing it down range to chech my hits .

I have a thread on shooting 100y i figured i walked to and back from that rack = 2.5 miles ( without looking back at that thread) .. no wonder why my dogs were barking at the end of the day..

 

[OldSpook]

Actually. Hit or miss IS binary. Two possibilties there. Hit - 1, Miss - 2. I actually did the arithmatic. (Credit to Mrs. Steeleman, my first grade teacher.)

And this leads me to ask a question that I often ask myself. What exactly is the purpose of this exercise?

J~

Clever.

Actually you must have failed to notice that I underlined a portion of your post. I underlined the part where it says, "predict where the next shot will land". Which to me begs the question, "with what level of confidence?". Probabilities being what they are, are non-binary, hence the observation.

Sorry not trying to ruffle your feathers.

Regarding your question about exercises; what do you think the purpose is?

 

[PumaCarl]

Interesting thread. I only shoot groups to figure out what my barrel likes best, then zero with the favorite. When that is all done I put the gun away and take it out and shoot exactly one shot every day. I only care about the first shot, and once I can trust it, I check that every week. If a gun can't do a first shot hit, it don't hunt. This is why I got tired of my FWB124D, I had to shoot 10-20 warm up shots into the dirt before starting a hunt to get it to settle down before I would shoot a pidgeon, that's bs compared to pcp. Usually I only get one shot on a woodchuck, so only that one shot matters to me, the four other shots are statistically worthless. I'm shooting woodchucks, not groups. We're all shooting for different things I guess.

 

[Airgunmann]

I typically do 1 mag or more groups, as in 2 mags, 3 mags, 4 or even 5 sometimes. Sometimes when the 5 , 6 or 7 shots are phenomenally placed I cheat and will stop the group short.

Stop short at the 1 shot group lol

 

[OldSpook]

@PumaCarl Oh yes, "the Cold bore shot", when I stopped shooting varmit rifles and started shooting airguns again, I made the mistake of mentioning on GTA that I was seeing a "Cold Bore like" behavior on my spring rifles. That one went over like a lead balloon. It didn't matter how certain I was that's what I was seeing. So I just stopped talking about it and kept track of it anyway. I still see that kind of behavior on all my rifles, some much more than others. I have a similar procedure to yours. I always shoot that first shot on a different Target during each shooting session.

 

[JimD]

To me the value of the video is to remind us that a single 5 shot group is not terribly reliable evidence of the accuracy of whatever we were testing. It is data but it isn't enough to draw a very reliable conclusion. More groups increase confidence but how much depends on how I am shooting that day. But I am not going to shoot hundreds to thousands of shots to be absolutely certain.

I knew my P35-22 was accurate the first time it put three shots into exactly the same hole at 25 yards. Not very many shots, statistically may not be significant but my conclusion has been confirmed with thousands of subsequent shots. It later did it again with a different head size pellet. But these pellets don't shoot the same on 30 yard challenge targets. One is noticably better than the other. The "bad" pellet is still better than my other rifles, however. But the rifle is nicely accurate.

 

[PumaCarl]

@PumaCarl Oh yes, "the Cold bore shot", when I stopped shooting varmit rifles and started shooting airguns again, I made the mistake of mentioning on GTA that I was seeing a "Cold
Bore like" behavior on my spring rifles. That one went over like a lead balloon. It didn't matter how certain I was that's what I was seeing. So I just stopped talking about it and kept track of it anyway. I still see that kind of behavior on all my rifles, some much more than others. I have a similar procedure to yours. I always shoot that first shot on a different Target during each shooting session.

Oldspook, the Brits talked about it and documented it extensively when the web was young. You are right, a lot of people (here in the U.S.) didn't want to admit it, but I did my own testing with two of them for three decades and nobody is telling me different. GTA had some overpowering personalities that stifled more than helped, a lot of the time, back then. Once I saw guys on the web changing mags and pellets without the need for seasoning the barrel, and of course the cold bore accuracy of pcp's, the penny really lost its shine on springers for me.

 

[drpietrzak]

@PumaCarl Oh yes, "the Cold bore shot", when I stopped shooting varmit rifles and started shooting airguns again, I made the mistake of mentioning on GTA that I was seeing a "Cold Bore like" behavior on my spring rifles. That one went over like a lead balloon. It didn't matter how certain I was that's what I was seeing. So I just stopped talking about it and kept track of it anyway. I still see that kind of behavior on all my rifles, some much more than others. I have a similar procedure to yours. I always shoot that first shot on a different Target during each shooting session.

Mine has one. Always has had 1.

 

[KatoKevin454]

I usually do a 1 shot group, since my Target is small and furry and usually there is not much of it left to shoot any more than once, besides it makes a nasty mess.....

 

[Nomadic Pirate 66]

you need to relax , shoot every shot the same . Only think of the shot you are taking now . try it . if you think of the last shot stop start over .

I'm not a target shooter, I don't really like it and never really do since it does not interest me.

 

[triggertreat]

I would agree with multiple 5-shot groups. This way the shooter gets to reset as well and can make adjustments for improvements to himself or the gun.

 

[Smitty911]

I would agree with multiple 5-shot groups. This way the shooter gets to reset as well and can make adjustments for improvements to himself or the gun.

What software is being used for measuring the data?

Smitty