5 or 3 shot group when determining ES/SD???

I appreciate all the replies guys lots of good info. It sounds like the answer to my question is no there is no standard # of shots used by everyone to determine an accurate ES/SD. So when Joe or Bob say they have an ES of 10 nobody knows how they came to that conclusion. Joe might shoot 30 shots in 3 shot groups and average out the 10 groups while Bob might shoot 3 and call it good. Makes me wonder how useful it is to hear the results of Joe or Bob's tests without a detailed description of how they came to such results?

Great point... which is why I say "the target is the final arbiter." I'd rather see Joe or Bob's long range target photos, preferably with witnesses. :)
 
another thing we have to consider is that barrels don't stay the same across hundreds of rounds of shooting--especially for barrel burning cartridges.
That's news to me. Having worn out 3 or 4 match grade barrels in magnum cartridges getting no more than 1200 or so rounds producing sub 3/4 MOA accuracy at 1000 yards, they all performed the same from first shot to the end of barrel life. "Death throes" of each barrel lasted only a few dozen shots. First indication of nearing the end of barrel life was sight settings for a given load, range and altitude had to come up about 1 to 1.5 MOA due to throat erosion.

Shoot 30 three shot groups. And, calculate the average for each 3 shot group. Then, take the ES/SD of the 30 averages and you'll begin to have a pretty good idea of future performance.
One's better off shooting one 90-shot group then do the math for ES/SD as well as mean shot radius. 'Course, if one doesn't belive tried and true statistics, then do something else.

Arsenals testing small arms ammo for all sorts of data shoot several dozen (a few hundred in some instances) shots per test group to get the best, most reliable data possible. Ever seen a 300-shot test group of 30 caliber ammo done at 600 yards?
 
Bart,
Are you saying for every load one wants to test they should shoot 90 shot groups or is this a hypothetical? You would need your barrel maker's # on speed dial!
 
That's news to me. Having worn out 3 or 4 match grade barrels in magnum cartridges getting no more than 1200 or so rounds producing sub 3/4 MOA accuracy at 1000 yards, they all performed the same from first shot to the end of barrel life. "Death throes" of each barrel lasted only a few dozen shots. First indication of nearing the end of barrel life was sight settings for a given load, range and altitude had to come up about 1 to 1.5 MOA due to throat erosion.

One's better off shooting one 90-shot group then do the math for ES/SD as well as mean shot radius. 'Course, if one doesn't belive tried and true statistics, then do something else.

Arsenals testing small arms ammo for all sorts of data shoot several dozen (a few hundred in some instances) shots per test group to get the best, most reliable data possible. Ever seen a 300-shot test group of 30 caliber ammo done at 600 yards?

3 shot groups, 5 shot groups, 90 shot groups don't matter.

There are ways to structure your testing to isolate different variables.

And, there are ways to statistically analyze whatever data you collect.

Regardless, the more good data you have, the higher confidence you can have in predicting future performance.

As to the OP, I get the feeling that most people eyeball the numbers and targets and just throw out numbers. I pay attention. But, it's just a reference point. I mostly focus on documenting my own performance so that I can tell whether I'm moving forwards or backwards because that's really all that I can control anyways.

-- richard
 
Are you saying for every load one wants to test they should shoot 90 shot groups or is this a hypothetical?
No. 90 shots in one group's better than thirty 3-shot ones.

For any one load, 15 to 20 shots is a good cost-effective number. It'll tell you what will happen all the time with about 75% to 80% confidence; statistically speaking, that is.
 
Bart... what magnum cartridge were you using that was ready for a new barrel at 1200 rounds? Or did I misunderstand your post?
 
Bart... what magnum cartridge were you using that was ready for a new barrel at 1200 rounds?
.264 Win. Mag. lasted 640 rounds. Three .30-.338 Win. Mag. barrels lasted about 1200 rounds.

The 26 caliber one was a sub .6 MOA rifle at 600 yards, sub .8 MOA one at 1000. That's what the best match bullets available back in the late 1960's would produce.

30 caliber ones were/are sub .4 MOA at 600 and sub .6 MOA at 1000.

I know of no 26 or 30 caliber rifle producing 25+ shot group accuracy at that level burning 65 + grains of powder doing so for more than 1300 rounds. Of course, there may well be some that'll shoot smaller groups with fewer shots once in a while (but there's other groups that are much larger) and have more barrel life under those conditions, but for long range high power competition, you gotta have good accuracy that lasts for at least 25 shots fired a minute or so apart.
 
.264 Win. Mag. lasted 640 rounds. Three .30-.338 Win. Mag. barrels lasted about 1200 rounds.

The 26 caliber one was a sub .6 MOA rifle at 600 yards, sub .8 MOA one at 1000. That's what the best match bullets available back in the late 1960's would produce.

30 caliber ones were/are sub .4 MOA at 600 and sub .6 MOA at 1000.
It's important to realize that if one's rifle using these cartridges starts out with less accuracy, its barrel life will tend to be longer. That's because barrels starting out wil less accuracy take longer to notice when they open up 25% to 50%, A factory rifle for one of these cartridges starting out at 1 or more MOA at these ranges may well go 2 to 3 times as many rounds of barrel life.

Note that to me, accuracy is the largest many-shot test group fired. That's what can be counted on all the time. All those micro-measured single few-shot groups are the smallest percentage of all groups fired.
 
I guess I might be the anal one here but everytime I go to the range for load development, sight in, trigger practice ect. I set up the chrono and record the information. I have several rifles that have over 300 rounds over the chrono which has allowed me to trend as little as a 10 degree ambient temp difference for loads used in long range shooting. ES and SD beg to be averaged over more than 3 or 5 rounds if you are actually going to use the information in your long range shooting. Statistically, the more information given to the input,the better the result. If you are trying to use ES and SD with 3 to 5 rounds to find a good load, IMO you are better off looking for accuracy nodes and not wasting your time with a chrono. If your intent is to gain information to use in the field to help with shot placement at long range then ES and SD information obtained at the range will only help.
 
Sleepyeddie pretty much summed up what I was tring to say in my post...I don't think it is anal at all to accumulate as much info as you can...The long string SD calculations are especially important for the competition shooter but I believe the day after day in any kind of weather statistics are the most important for the long range hunter...Especially that first cold bore shot...Unlike the competition shooter who has to depend on long strings and is able to compensate for differences as he goes, the long range hunter has to make that first one count....As was said the ambient temperature, elevation and other factors can be figured in any ballistics program but things like the differences in powder type and temp changes are not included in your off-the shelf programs....All the data collected in all weather types can be analyzed and reasonable hypothesis may be generated from this info...ie; "Oops, left that one laying on the hot bench too long".....This accumulated over time data
should tell you your maximum "do-able" range on any given day and maybe even tell you when you should not shoot.

There is no substitute for actual field testing but the equipment we have today will help us move closer to our actual goals...

It has been said in this post numerous times that the most accurate SD compilations are based on the largest number of instances and no one can argue that fact.....For me it is way more important to have a lot of "short string" information accumulated over time with variable conditions because that is how I am going to use it....The numbers themselves are actually all "relative" at that point and might not necessarily be exactly what others may think they should be....

IMHO,
Randy
 
I guess I might be the anal one here but everytime I go to the range for load development, sight in, trigger practice ect. I set up the chrono and record the information. I have several rifles that have over 300 rounds over the chrono which has allowed me to trend as little as a 10 degree ambient temp difference for loads used in long range shooting. ES and SD beg to be averaged over more than 3 or 5 rounds if you are actually going to use the information in your long range shooting. Statistically, the more information given to the input,the better the result. If you are trying to use ES and SD with 3 to 5 rounds to find a good load, IMO you are better off looking for accuracy nodes and not wasting your time with a chrono. If your intent is to gain information to use in the field to help with shot placement at long range then ES and SD information obtained at the range will only help.

Plot your 300 samples and it should approximate a normal distribution a/k/a bell curve.

Calculate the mean and standard deviation of all 300 and you should have roughly 95% confidence that any future shots fired will fall within 2 standard deviations of the mean.

You could then pull 30 random cold bore samples and calculate the mean and standard deviation followed by a two tailed T test to see if there's a statistically significant difference in MV for your cold bore shots vs all shots fired.

Well, that's a very rough example. But, most of us unintentionally tend to focus on our smallest group or tightest ES regardless of the facts.

Figures don't lie. But, liars figure.

The most important thing is to measure your own performance and keep in mind that when the numbers continue to shrink, you're doing something right and when they grow, you're going in the wrong direction.

-- richard
 
Now we've gotten to the meat and potatoes. My title for this thread was mis leading. I understand the more info the more accurate the hypothesis. I was just wondering if there was a consistent number of shots that everyone uses to determine ES/SD. Now I know that there is not but I also learned how to conduct proper testing while developing loads for long range hunting. I like the way you summed it up sleepyeddie but all of the info in this thread has been quite informative. Thankyou for sharing your knowledge guys!
 
I would say 5 shots would give a good SD and ES to draw a reasonable conclusion on the accuracy of a load.
While this is a popular thing to do, it ain't always right. Sometimes it produces the worst accuracy at the greater ranges.

For example, when a bunch of us developed loads for Sierra's then new .308 caliber 155-gr. Palma bullet back in 1991, the load that produced the smallest spreads in metered charge weight, peak pressure and muzzle velocity also had the worst accuracy at long range. The load with medium values for these things had the best accuracy.
 
Warning! This thread is more than 13 years ago old.
It's likely that no further discussion is required, in which case we recommend starting a new thread. If however you feel your response is required you can still do so.

Recent Posts

Top