• This Forum is for adults 18 years of age or over. By continuing to use this Forum you are confirming that you are 18 or older. No content shall be viewed by any person under 18 in California.

Converting 3-shot group size to 5- and 10-shot group size

I believe that. I regularly shoot 5 shot groups. Why do they get larger is the question.
Simply because you're increasing the sample size. The theory is that shots are 'normally' distributed. As you draw more sample/take more shots, you're more likely to see samples/shots that are out further in the distribution.
At the same time, those samples/shots that are further out in the distribution could show up within the first five shots - which would simply produce a larger 5 shot group.
 
If I wanted to convert a 3 shot group into a 5 shot group I would need to fire 2 more rounds. That’s only way it makes any sense to me. I don’t think extrapolating or using some computer software would provide me with convincing evidence.
The best way to "convert" a three shot group to a five shot group is to shoot 2 more rounds. Or you can use the statistical factor previously cited to estimate what the long run average five shot group will be. Still you only have an estimate.

Even if you shoot two more rounds you have only one data point for estimating what the next five shot group will be. The statistical confidence level you choose will deterring how many more five shot groups you have to shoot to achieve your desired confidence level. I understand statistical analysis well enough to use it but not well enough to teach it.

I have observed that when this subject, statistical analysis, Is introduced the long knives come out and those that do not understand it and can not teach it are the biggest critics.

Statistical analysis is a powerful tool for analyzing groups, but you have to know when and how to use it.
 
The best way to "convert" a three shot group to a five shot group is to shoot 2 more rounds. Or you can use the statistical factor previously cited to estimate what the long run average five shot group will be. Still you only have an estimate.

Even if you shoot two more rounds you have only one data point for estimating what the next five shot group will be. The statistical confidence level you choose will deterring how many more five shot groups you have to shoot to achieve your desired confidence level. I understand statistical analysis well enough to use it but not well enough to teach it.

I have observed that when this subject, statistical analysis, Is introduced the long knives come out and those that do not understand it and can not teach it are the biggest critics.

Statistical analysis is a powerful tool for analyzing groups, but you have to know when and how to use it.
In my prior life before retirement I was pretty deeply involved with six-sigma projects and actually do understand statistical analysis. But I would rather just fire two more rounds. I know----------Minitab is your friend, just spend some time wallowing in the data...................Or I could just fire two more rounds :) No knives:)
 
If I wanted to convert a 3 shot group into a 5 shot group I would need to fire 2 more rounds. That’s only way it makes any sense to me. I don’t think extrapolating or using some computer software would provide me with convincing evidence.
True, but the OP was asking about converting so he could get an estimate of group size from what others post as their 3 shot group. It is strictly for estimating purposes.
 
Better off shooting a 20, 30, 40 round group on a target at 200m and then count how many rounds land inisde 1MOA, 2MOA, 3MOA of the point of aim. What percentage fall inside of each circle in 1MOA increments will be more telling of the accuracy of the rifle and ammo combination.

If the rifle, ammo and shooter are too good this will not be very telling if you are basicly stacking rounds so close that you get a massive circle of nothing left.

In the case of a BR or competition rifle and ammo you would want to build a jig to stack targets in perfect allignment and shoot 3, 5, 10 shot groups. You would repeat this say 10 times and compare the groups from datam line. You would aggregate the total and individual group.

As an someone that has worked in engineering in automotive manufacturing and aviation that is how I would set up testing accuracy of a rifle if I had to create data for a client. That said there are many ways to skin this cat depending on what data you are actual looking for.

I would not try to do a meta analysis of data that was not created by myself unless I knew all the details of how that data was generated and measured and what model was used to process the data. Sometimes you do not have a choice but often for empirical data you are often better off generating fresh data so long as you can have enough sample size to be significant for purpose. Often we are looking more for trending data than anything absolute unless we have some unknown we are trying to discover and solve for so it can be accountedf for and solved for!

Estimating group size is a fools folly in my opinion. In my youth I lifted weights and played sports. Their was a chart for coverting weight lifted by reps to failure into predictive load capcity at other rep counts. So if you could do 200lbs. for 10 Reps to failure it was suppossed to predict what you could lift for say a 3RM 3 repetition max or 1RM and the other way as well. I never found a single person that this wall chart was ever accurate for. I asked the coach if he had ever met an athlete where that chart predicited their conversion accuracy and after heming and hawing for what seemed like forever he said "No not really." so outside of it breaking up the bad paint job int he gym like wallpaper or a painting on the wall it was not worth while and the very premise of it is absurd.

I have been reloading since 1995 and to be honest only with a few cartridges does reality shake out with predicitive expectations. If you doubt me watch video's about people that reload and then test loads on youtube. Even using loading software looking for nodes and then doing a shooting ladder if you watch enough video's of this it quickly becomes evident that it is not as empirical, logical and intuitive as some would have you believe it is.

There is a very real reason tunners are more popular than ever amoung compittors than ever. Likewise depending on which shooting sport you are talking about some high level shooters are actualy using factory loaded match ammo which no one would even consider 20 years ago in any shooting sport that did not require it like Palma did for some time.
 
Better off shooting a 20, 30, 40 round group on a target at 200m and then count how many rounds land inisde 1MOA, 2MOA, 3MOA of the point of aim. What percentage fall inside of each circle in 1MOA increments will be more telling of the accuracy of the rifle and ammo combination.

If the rifle, ammo and shooter are too good this will not be very telling if you are basicly stacking rounds so close that you get a massive circle of nothing left.

In the case of a BR or competition rifle and ammo you would want to build a jig to stack targets in perfect allignment and shoot 3, 5, 10 shot groups. You would repeat this say 10 times and compare the groups from datam line. You would aggregate the total and individual group.

As an someone that has worked in engineering in automotive manufacturing and aviation that is how I would set up testing accuracy of a rifle if I had to create data for a client. That said there are many ways to skin this cat depending on what data you are actual looking for.

I would not try to do a meta analysis of data that was not created by myself unless I knew all the details of how that data was generated and measured and what model was used to process the data. Sometimes you do not have a choice but often for empirical data you are often better off generating fresh data so long as you can have enough sample size to be significant for purpose. Often we are looking more for trending data than anything absolute unless we have some unknown we are trying to discover and solve for so it can be accountedf for and solved for!

Estimating group size is a fools folly in my opinion. In my youth I lifted weights and played sports. Their was a chart for coverting weight lifted by reps to failure into predictive load capcity at other rep counts. So if you could do 200lbs. for 10 Reps to failure it was suppossed to predict what you could lift for say a 3RM 3 repetition max or 1RM and the other way as well. I never found a single person that this wall chart was ever accurate for. I asked the coach if he had ever met an athlete where that chart predicited their conversion accuracy and after heming and hawing for what seemed like forever he said "No not really." so outside of it breaking up the bad paint job int he gym like wallpaper or a painting on the wall it was not worth while and the very premise of it is absurd.

I have been reloading since 1995 and to be honest only with a few cartridges does reality shake out with predicitive expectations. If you doubt me watch video's about people that reload and then test loads on youtube. Even using loading software looking for nodes and then doing a shooting ladder if you watch enough video's of this it quickly becomes evident that it is not as empirical, logical and intuitive as some would have you believe it is.

There is a very real reason tunners are more popular than ever amoung compittors than ever. Likewise depending on which shooting sport you are talking about some high level shooters are actualy using factory loaded match ammo which no one would even consider 20 years ago in any shooting sport that did not require it like Palma did for some time.
I am curious how you do load development and test variables.
 

Upgrades & Donations

This Forum's expenses are primarily paid by member contributions. You can upgrade your Forum membership in seconds. Gold and Silver members get unlimited FREE classifieds for one year. Gold members can upload custom avatars.


Click Upgrade Membership Button ABOVE to get Gold or Silver Status.

You can also donate any amount, large or small, with the button below. Include your Forum Name in the PayPal Notes field.


To DONATE by CHECK, or make a recurring donation, CLICK HERE to learn how.

Forum statistics

Threads
165,594
Messages
2,198,869
Members
78,989
Latest member
Yellowhammer
Back
Top