Standard Deviation?
And while we are in the mood for asking questions, can any of the Experts out there explain what "Standard Deviation" means?
I know I am sounding like a dunce here, but this is another term that I have never understood from day one of my shooting career.
Please someone out there explain it too me so that I can understand it, and explain it to some of my shooting partners.
Thanks as always for your help,
BigBarn[:o)]
"Respect the land, and it will respect you"
I know I am sounding like a dunce here, but this is another term that I have never understood from day one of my shooting career.
Please someone out there explain it too me so that I can understand it, and explain it to some of my shooting partners.
Thanks as always for your help,
BigBarn[:o)]
"Respect the land, and it will respect you"
0
-
BigBarn: Try this link for an explanation:
[:D]
http://www.robertniles.com/stats/stdev.shtml
Don't assume malice for what stupidity can explain.0 -
It is the normal variation above or below a given average.It can also be referred to as tolerence. A replacement part may work if it is plus or minus .OO2 since it falls within the standard deviation.Two rounds may show different velocities, even though they have exactly the same measured components.
Love them Pre-64's!!!!-Bob0 -
stand at a counter in a quicky mart and pay for the exact cost of a pepsi. total comes to 75 cents. you only have 74 cents, so the cashier reaches over and throws a spare penny laying in the counter ashtray into the payment. you get the soda and the standard deviation of the sale was 1 cent. even though you were shy a penny, the sale took place. go back the next day and this time you are 5 cents shy for a pepsi. cashier says, im sorry, youre 5 cents short. have a nice day. you were out of tolerence or beyond the standard practice of politeness, or, the standard deviation. follow me? 0 -
BigBarn,
I'll throw little more fuel on the fire...
"The standard deviation is a number that describes uniformity. The smaller the number, the more uniform velocity."
That's as simple as I have seen it written with regard to firearms and velocity. Here is the entire article:
http://www.naaminis.com/standev.html
Best.0 -
It's a measurement of dispersion of data from a statistically normal
curve. The smaller the number the better you are as regards pressure, velocity, group size etc.
Like a stream you have to cross: to get through it you have to understand statistical concepts and get wet.0 -
About anything manufactured has a specification for some attribute.
Let's say our Muzzle Velocity is supposed to be 1700 fps. Every cartridge produced isn't going to yield exactly 1700 fps - the manufacturer isn't going to want to pay for what that would cost and neither are you. So a compromise is made. The spec is actually 1700 fps plus or minus 30 fps. If the measured muzzle velocities are all in spec, then all that fall within plus or minus 10 fps would all be within 1 standard deviation, those that fall within +/- 20 fps would be within 2 standard deviations and those that fall within +/- 30 fps would be within 3 standard deviations.
But if another group of cartridges yield anywhere from 1640 to 1760 fps, velocites between 1680 to 1720 would fall in the 1st standard deviation while velocites between 1660 to 1740 would be in the 2nd standard deviation and the last group (1640 to 1760) would be in the 3rd standard deviation. So the velocites in the 3rd standard deviation and half of the 2nd standard deviation are all 'out of spec' - don't meet the 1700 fps +/- 30.
I don't know if any of that will make sense, but a low standard deviation is a good thing.0 -
KX500 is right that when it comes to ammunition, lower standard deviation is desired. You want less variability (or inaccuracy) in the velocity of your rounds. A high standard deviation indicates less accuracy/more variability -- and therefore sloppy quality control in manufacturing.
It's important to note that the standard deviation is not chosen arbitrarily. It is calculated from the data that comes from test-firing and measuring the velocity of each round. In order for the mean (average) and standard deviation to be accurate, you have to test-fire thousands of rounds, but for the sake of simplicity, let's just use five test-shots.
Ammunition 1
1000 fps
1250 fps
1500 fps
1750 fps
2000 fps
Average: 1500 fps
Standard Deviation: 395.28 fps
Ammunition 2
1450 fps
1475 fps
1500 fps
1525 fps
1550 fps
Average: 1500 fps
Standard Deviation: 39.53 fps
All other things being equal, which of these ammunitions is better? Both have the same average velocity (1500 fps). However, when ammunition 1 was test-fired, the velocity of the rounds varied widely. Some were as high as 2000 fps, while others were as low as 1000 fps. Although the average is 1500, it isn't too likely that most of your rounds are going to fire at that velocity. Now let's look at ammunition 2. It's average velocity is also 1500 fps, but when you look at results of the test firing, the measured velocities are pretty close to 1500 fps -- the fastest is only 1550 fps, and the slowest is only 1450 fps. If you want your rounds to be firing close to 1500 fps, you should go with ammunition 2.
The standard deviation is a reflection of the variability of your velocity. The velocity of ammunition 1 varies widely -- on average, each round fires at a speed of 1500 fps plus or minus 395.28 fps. Ammunition 2 varies less in its velocity. On average its rounds fire at 1500 fps plus or minus only 39.53 fps.
I hope that clears things up for you.
quote:Originally posted by KX500
The spec is actually 1700 fps plus or minus 30 fps. If the measured muzzle velocities are all in spec, then all that fall within plus or minus 10 fps would all be within 1 standard deviation, those that fall within +/- 20 fps would be within 2 standard deviations and those that fall within +/- 30 fps would be within 3 standard deviations.
Are you saying that 30 fps is the standard deviation? If 30 fps is the standard deviation, then any variability within 30 fps (whether it is plus or minus) is within 1 standard deviation. The same error appears in your second example.
Jacqueline
www.gratuitouslylongdomainname.net
"Necessity is the plea for every infringement of human freedom. It is the argument of tyrants and the creed of slaves." -- William Pitt (1783)0 -
FreudianSlippers - No I didn't mean to imply that 30 fps is the standard deviation in my first example. Standard deviations in my examples were 10 fps for the first and 20 fps for the second.
+/- 30 fps was simply the spec limits or control limits.0 -
quote:Originally posted by KX500
FreudianSlippers - No I didn't mean to imply that 30 fps is the standard deviation in my first example. Standard deviations in my examples were 10 fps for the first and 20 fps for the second.
That makes more sense. If the standard deviation is 10 fps, then 20 fps would be 2 standard deviations from the norm (z=2), and 30 fps would be 3 standard deviations from the norm (z=3).
Jacqueline
www.gratuitouslylongdomainname.net
"Necessity is the plea for every infringement of human freedom. It is the argument of tyrants and the creed of slaves." -- William Pitt (1783)0 -
Here is a very good article on Standard Deviation written by Ken Howell for Precision Shooting Magazine.
http://www.precisionshooting.com/
July 2004 Vol. 52 N0. 30
Please sign in to leave a comment.
Comments
10 comments