Mechanical vs Practical accuracy

[rwilson1]

OK, I am a bit of a nerd. I have kept very meticulous records of my shooting for just over 3 decades. Over the past 6 months I have devoted an inordinate amount of time setting up and entering shooting data in SPSS (SPSS is a statistical analysis program for those of you not familiar with the term). I have found the process very enlightening. I found that under typical range conditions, with an AR pastern rifle I can consistently deliver 1.5-2.5 times the mechanical accuracy of the rifle I am using, with a mean of 2 times and a standard deviation of .03; under more practical conditions (stress, less conventional positions, ect.) I consistently deliver 2.4-3.7 times mechanical accuracy with a mean of 3 and a standard deviation of .21. All of that is a very nerdy way of saying that with an AR pastern rifle, with which I can shoot a 1 inch bench rest group at 100 yards, I have a 99.7% chance of hitting a 3.63 inch (call it a nominal 4 inch) target at 100 yards under adverse conditions. In other words, for me, at this point, with an AR pattern rifle, Practical accuracy = Mechanical accuracy x 3.63 (or call it 4 to minimize error variance). This means I can consistently deliver 2.75moa or better accuracy under typical range conditions, and 4moa or better accuracy under most any circumstances, with a 1moa rifle. I am curious how this compares to others. Has anyone else compared their mechanical and practical accuracy in any way? what did you find?

 

[illinoisburt]

My educated guess is the majority of shooters do a whole heap worse from unsupported field positions. The calculation of 4 moa is literally hitting a standard clay target 100 yards away 99 times out of 100 from a standing unsupported position. I know people who can do it (or get close), but its certainly not the majority of shooters at the range.

 

[rwilson1]

the data set for these results was collected over 381 months wit 17 AR rifles and 54,605 rounds fired, the most questionable part of the results is the "practical conditions" results because only 7,342 rounds were assigned this value and that assignment is a bit subjective. Another possible confounding variable, is that only 4 of the 17 rifles were 1.5moa or better bench rested, so from a skill point of view the data assumes that a 12 inch group with a 4moa rifle is equal to a 3 inch group with a 1moa rifle. I am not sure that assumption is valid. If I had unlimited time and resources, it would be interesting to try to find out.

 

[ontarget]

After my rifle is zeroed it doesn't see a bench again. What you and your weapon are capable of in a practical shooting scenario is the only thing that matters. That is unless the deer, coyote, elk...... Or bad guy are going to give you time to set up your bench. Varmints are a little different maybe.

 

[Nature Boy]

I know the capabilities of my "me/rifle/load" system more intuitively than empirically based on a lot of observed results down range. I'd love to do the analysis you've done to see what it says, but honestly I don't know how I'd use the data.

I also know I'm still the limiting factor in improving my accuracy.

 

[rwilson1]

Each rifle was benched to determine mechanical accuracy ( and ammo preference, zero, ect.) typically less than 100 rounds each. However, the vast majority of the rounds fired were under "typical range conditions" I.E. decent weather, good lighting, known distance, low stress, ect. While range, shooting position, stage set up ect. varied significantly. I assume this is true for most of us.

 

[entropy]

I must be confused. Wouldn't mechanical accuracy be what you get if you put the action in a vise, ala some of the Unlimited Class benchrest guns? I see no possible way one could derive better 'practical' accuracy (that is unsupported offhand) from a gun than that. There's a reason pistols are bench tested with a Ransom Rest- to eliminate the least accurate variable- the nut behind the trigger.

My educated guess is the majority of shooters do a whole heap worse from unsupported field positions. The calculation of 4 moa is literally hitting a standard clay target 100 yards away 99 times out of 100 from a standing unsupported position. I know people who can do it (or get close), but its certainly not the majority of shooters at the range.

My observation of years of myself and others while shooting (and as a 4-H leader this is quite a bit) is that usually prone is the most accurate position, Kneeing supported/seated (cross-legged with elbows on knees) is next, then target standing supported( the position Olympic shooters use; left elbow against hip, hand back by action) and 'field' standing unsupported the least accurate. For some, this is not 100%; often kids in a growth spurt will not shoot well from kneeling, while smaller kids (they start at 12) have a hard time holding the rifle up in either field position. .(22's and air rifles) If one anchors the stock in the 'pocket', leans forward, and pushes the rifle firmly into said pocket, this increases accuracy, particularly with follow-up shots. This has been my biggest challenge with the kids, is to get them to do so. Between natural inclination and seeing in movies people shooting AR's and M4's with the tip of the bottom of the stock on the shoulder, most of the kids don't assume the correct position even after being taught. The few that do have hunted with a .30-06 class rifle, and have learned the hard way.

After my rifle is zeroed it doesn't see a bench again. What you and your weapon are capable of in a practical shooting scenario is the only thing that matters. That is unless the deer, coyote, elk...... Or bad guy are going to give you time to set up your bench. Varmints are a little different maybe.

I return to the bench to check zero occasionally, but yes, I am much more interested in how a rifle shoots from a standing unsupported position for the same reasons. I do shoot some of my coyotes from prone, but I've shot most of them while deer hunting, from standing unsupported, or seated unsupported from a tree stand.

I also know I'm still the limiting factor in improving my accuracy.

Likewise. I've been tightening up the 'nut behind the trigger' for over 40 years.

 

[ontarget]

That's what I mean by practical shooting, any field position, supported by your own bone structure, or the ground in prone. It's very difficult to carry a shooting bench into the deer woods or combat.

 

[Jim Watson]

Note that the SR target for 200 yard standing High Power matches has a 7 inch, 3.5 MOA ten ring. So you can look up match scores and get a good handle on what people are doing with accurate rifles hand held.

from a skill point of view the data assumes that a 12 inch group with a 4moa rifle is equal to a 3 inch group with a 1moa rifle. I am not sure that assumption is valid.

I am not, either. I don't think shooter accuracy vs rifle accuracy is multiplicative. It is more nearly additive. If you can only hold a 12 inch group with a 4 MOA rifle, then you are more likely to shoot about a 9 inch group with a 1 MOA rifle. Contrariwise, if you can hand hold 3 inches with a 1 MOA rifle, you ought not to do worse than 6 inches with Uncle Bud's 4 MOA dirty thirty. Somebody once did the statistics on it and it was not QUITE that bad, but nearly so.

usually prone is the most accurate position

I have heard expert smallbore shooters to say that they can shoot a .22 target rifle more accurately prone with coat, glove, and sling, than from a benchrest.

 

[illinoisburt]

the data set for these results was collected over 381 months wit 17 AR rifles and 54,605 rounds fired, the most questionable part of the results is the "practical conditions" results because only 7,342 rounds were assigned this value and that assignment is a bit subjective. Another possible confounding variable, is that only 4 of the 17 rifles were 1.5moa or better bench rested, so from a skill point of view the data assumes that a 12 inch group with a 4moa rifle is equal to a 3 inch group with a 1moa rifle. I am not sure that assumption is valid. If I had unlimited time and resources, it would be interesting to try to find out.

I would be interested in the mean group size of the pratical shooting as a number rather than a percentage. I think it would be a very useful number to gauge ability of the shooter and help establish realistic expectations for field uses. I suspect it would vary by ergonomics. Comparing to the benched results could yeild insight into how well a particular firearm fits the shooter and their shooting form.

The old "pie plate" hunting standard was awful but in practical terms probably a reliable judgement of peoples abilities. (Distance at which you could keep 4 of 5 rounds on a 9 inch plate fired in rapid sequence unsupported as the maximum usable range for hunting. Most folks with common hunting firearms this is 75 to 150 yards tops and sometimes less with slug guns or revolvers.)

 

[rwilson1]

With only 7,342 rounds fired under "practical conditions" at paper targets in over 30 years with 17 guns, at all ranges, the overall mean group size from this data set isn't a great metric. Unfortunately, given the limits of the data set, the mean group size ant any specific range with any specific gun, under "practical conditions" seems less reliable. For example, with my current home defense AR that mean group size at 100 yards would be 7.65 inches, but that is only made up of 9 groups and a total of 26 rounds. the rifle shoots 1.8moa.(also, for the sticklers, 1 inch per 100 yards was used rather than true moa)

 

[Newtosavage]

I let the shot opportunity tell me whether or not I should shoot. I may not have a bench out in the field, but I do have mother earth, rocks, trees and other things to steady my aim. Every scenario is unique.

 

[Nature Boy]

The old "pie plate" hunting standard was awful but in practical terms probably a reliable judgement of peoples abilities.

But it only matters if you shoot said pie plate while resting your rifle on a fence post, or hood of a pick up truck, as long as it's '74 vintage or older, brand notwithstanding

 

[jmr40]

I think the 9" paper plate rule of thumb is a good one. In a hunting rifle much more accuracy isn't really beneficial. At whatever range you can keep all of your shots on a paper plate from field positions is your limit. And it doesn't doesn't have to be offhand. In fact firing from any unsupported position should only be used as a last resort. There are all sorts of ways to steady a shot in the field including shooting sticks or even bipods if you don't mind the weight.

This group was fired in the field with my rifle resting on a pack, range was 430 yards. I hit the target with a rangefinder and used the 400 yard dot on my scope. This qualifies as a field position. All 3 shots in a 2 1/2 group and none more than 3" from the aiming point. I've done almost as well with shooting sticks.

 

[Varminterror]

Practical accuracy = Mechanical accuracy x 3.63

What is your empirical scaling factor for a 2 IPHY mechanical rifle? My point being, in reading your post, it appears that you are multiplying two additive variables... Your mechanical group size is an angular dispersion, your wobble area is an angular dispersion. I suspect all of your regressions look pretty similar because you're shooting a lot of rifles with similar mechanical precision potential, and then you as the shooter have the same relative wobble area, which grows with your increasingly adverse conditions. But what happens when your mechanical potential changes significantly? Analytically, I'd say you fell into a statistical trap, developing a multiplicative scalar for additive variables. I'm happy to be wrong, but I don't see the explication in your original post.

I've done similar analysis as well, and I can say, with a half-moa rifle vs. a 1.5MOA rifle, my "adverse condition" shooting capacity is not tripled. Using your empirically determined 3.63x, a half moa rifle should yield ~1.8" groups under adverse conditions, whereas a 1.5MOA rifle would yield 6" groups. I can attest personally, I'd be laying high bets on myself to hit 6" targets with a 1.5MOA rifle at 100yrds, whereas I'd lay just as heavy of bet against myself to hit a 1.8" target off-hand, even with a 0.5MOA rifle. I regularly practice at 100yrds, iron sights, off hand, with two rifles. One is a 1.5MOA capable rifle (barrel shot out thousands of rounds ago, but she's holding steady at 1.2-1.5), one is a 1/2-3/4moa rifle from the bench. I use a 4" round steel target for either, and I consistently hit with both. I miss more often with the 1.5moa rifle, which may be the weight of the rifle, or may be the mechanical potential, but misses are pretty rare for either. My wife has a reliably 1/4" rifle - your 3.63 multiplier would claim, in your hands, you'd shoot this rifle to about 7/8" - can you shoot sub 1" at 100yrds?

And of course - I'd be interested to see how you've quantified "adverse condition severity" into your scalar? As I mentioned, I'm well practiced in off-hand shooting, but I can say it is much easier to shoot off hand on days where the wind is not gusting (I shot yesterday with 30 stable and 60mph gusts), and although I've only rarely done so, shooting from any seacraft or helicopter is absolutely terrible. Shooting from a boat on high swells and the wind which causes them is a very different level of "adverse condition" than shooting offhand in 0-5mph wind.

 

[Varminterror]

Service Rifle matches aren't as popular as they once were, and they were never hugely popular in general, but I can tell you, any of the guys who spent much time there knew their wobble area, and how it varied from prone to sitting to standing. Wobble area doesn't multiply linearly with by mechanical accuracy. One is about the rifle, one is about the shooter - they're additive.

 

[hdwhit]

rwilson1 asked:
Has anyone else compared their mechanical and practical accuracy in any way?

Not really. That's because, other than claims by the manufacturer or those in the gun press, I really have no way of objectively determining the "mechanical accuracy" of any of my rifles, so all of my accuracy measures would then need to be considered "practical accuracy".

what did you find?

The standard that I came up with for my own use as a teenager was whether or not I could consistently put at least 9 our of 10 rounds into the circular divot on the side of a plastic gallon milk jug at 100 yards firing offhand. Depending on the manufacturer, that divot is between about 2.75 and 3.5 inches in diameter.

I consider that sufficient accuracy for self-defense and hunting purposes. Since the onset of my neurological problems, I no longer shoot competitively.

 

[Varminterror]

In general, I've found, in ~18yrs of instructing new rifle shooters, the average Joe can't hit a milk jug at 100yrds off hand, regardless of rifle potential. Hand in hand, I have found, with practice, most marksmen should be able to achieve sub 4" at 100yrds with any rifle south of 1.5MOA in mechanical potential.

 

[rwilson1]

This exercise was as much about learning SPSS as it was about shooting for me. I have to learn SPSS for work and I have a lot of shooting data available. "typical range conditions" from "practical conditions" because my AR shooting at paper targets (the only targets that allow group measurement) was primarily just trips to the gun range, but some of it was competition and training that induced stress and very different results. I defined "practical conditions" rather subjectively. Comparing standing, kneeling and prone groups fired at exactly known ranges without serious time pressure to groups fired in the same positions modified according to available cover at estimated ranges, after sprinting to increase heart rate rendered very different results. The analysis is by no means perfect, but it is my first attempt at a statistical analysis based on a large mult-variable self generated data set. I am going to play with the data a lot more, who knows, maybe I will learn something. I haven't finished inputting handgun data yet. If this process has taught me anything, it is that sorting and quantifying data in a meaningful way is hard, and real world results are often very random. I have always been skeptical of statistics. I have always questioned the motivation and integrity of those doing the analysis. It is very frustrating that I don't even trust my own statistical analysis LOL,....but it is an interesting process.

 

[rwilson1]

I thought I kept impeccable records. I can tell you the exact number of rounds fired from every firearm I have ever owned. I can tell you the serial number of every firearm I have ever owned. I can tell you how many rounds I have fired from AR-15/M16 style rifles, how many of those have been fired at paper targets, from what distance, in what position, and what the groups measured. I can tell you the time of every timed drill and competitive stage I have ever shot. but even with all of this data, relevant statistical analysis is hard. I found that I have lots of unanswered questions. Despite my only motivation being to learn the "truth" I am having a hard time organizing the data in a way that presents an accurate picture of one persons performance over time. Knowing that, it is hard to believe the level of trust that we have in statistics generated with far less complete data, not just in understanding the past, but in predicting the future results or performance in a diverse population.

 

[sage5907]

I have a good friend who for many years schooled everyone about using crutches. Alcohol, tobacco, etc were crutches that should be avoided through your entire life. As a hunter I feel the same way about shooting from the bench and it was well said by ontarget above. When you get right down to being able to use a firearm the only tools that make the rifle work is what you can carry with you and what you can use quickly. If you want to shoot something and you look around for a tree, bush, rock or hump in the ground as a crutch you have a big problem. Most animals are only in view for a few seconds and to hesitate is to fail to get the shot. I personally have bent this scenario and I carry a single shooting stick and I can use it efficiently out to at least 300 yards. Target shooting to a hunter is only good to check the point of impact of the rifle. All hunting practice should be done under field conditions. Accuracy without a crutch is what I would call practical accuracy.

 

[Varminterror]

Like I said above, you fell into a really common trap. I'm sure there wasn't anything wrong with your regressions, but you made a choice to establish a linear relationship between independent variables. It's a simple trap to fall into, but it's not the right analysis. What you have assumed is Y = MX + B, practical precision = mechanical potential * Scaling factor + offset (which you assumed to be zero). What you actually have is two independent variables X + Y = Z, mechanical potential + wobble area = Group size, where wobble area dispersion is a function in itself, relating the stability of the shooter to his shooting conditions, so Y = f(E) .: X + f(E) = Z.... Never does wobble area depend upon the mechanical precision potential of the rifle, so your Y = MX formula just doesn't apply. That's not a statistical error, it's a system analysis error.

 

[Newtosavage]

OMG please tell me there isn't a "pie plate" test with rifles too. I have seen that applied to traditional bowhunting for over 30 years.

 

[Varminterror]

OMG please tell me there isn't a "pie plate" test with rifles too. I have seen that applied to traditional bowhunting for over 30 years.

Dude - we (rifle shooters) had the "pie plate" test way before bowhunters!!

 

[ontarget]

I let the shot opportunity tell me whether or not I should shoot. I may not have a bench out in the field, but I do have mother earth, rocks, trees and other things to steady my aim. Every scenario is unique.

I have a good friend who for many years schooled everyone about using crutches. Alcohol, tobacco, etc were crutches that should be avoided through your entire life. As a hunter I feel the same way about shooting from the bench and it was well said by ontarget above. When you get right down to being able to use a firearm the only tools that make the rifle work is what you can carry with you and what you can use quickly. If you want to shoot something and you look around for a tree, bush, rock or hump in the ground as a crutch you have a big problem. Most animals are only in view for a few seconds and to hesitate is to fail to get the shot. I personally have bent this scenario and I carry a single shooting stick and I can use it efficiently out to at least 300 yards. Target shooting to a hunter is only good to check the point of impact of the rifle. All hunting practice should be done under field conditions. Accuracy without a crutch is what I would call practical accuracy.

Bingo and Bingo. I wish every rifle shooter that brags about their sub moa rifle would pull that new whatever brand rifle off their lead sled and see what happens.
I'm not doubting the potential consistency of today's crop of rifles but for practical applications get it off the bench. Use your pack atop a stump or shooting sticks or anything else you may have available in the real world and practice that way.