[WiskyT]

My Chrony chronograph will only display individual shot velocity unless a pushbutton switch is plugged into the jack on it's side. Some may not have this switch or are debating buying or making one. There may be some chronographs out there that do not have any capacity to calculate standard deviation. I forgot my switch the other day and had no ability to get SD or average and once the chrono is shut off, all data is lost. So I simply wrote down each individual shot velocity.

I found this calculator and you simply type in each shot and it will calculate useful data for you. I'm sure there are others, but this one is the first one I clicked on and it worked well.

http://easycalculation.com/statistics/standard-deviation.php

 

[XDRoX]

When I teach Stats I make my students find SD by hand a few times before I show them that their calculators can do it for them. Man are they happy when they see the calculator do it for them. Doing it by hand sucks.

I may be mistaken but I think Excel has a SD function as well.

 

[WiskyT]

When I teach Stats I make my students find SD by hand a few times before I show them that their calculators can do it for them. Man are they happy when they see the calculator do it for them. Doing it by hand sucks.

I may be mistaken but I think Excel has a SD function as well.

Now that you mention it, I found that Open Office Spreadsheet has a million different statistical functions. Since I never used it, it almost seems easier to enroll in a community college than to try to figure out how to use it. But, in 5 minutes of playing around, I got it to do the calculations. At this point, my system of napkins and online calculators seems easier, but maybe I should look into this spreadsheet thing.

 

[XDRoX]

I've always uses Excel to record all my reloading data. Nothing fancy, just a simple record of what I load and how it shot. The columns just make everything so easy to record and track.

 

[F106 Fan]

Excel is even more friendly when I can import data from my CED M2 Chronograph.

In this case, the SD and ES are part of the imported data but I could easily arrange for Excel to make the calculations. I suppose, with a little more effort, I could get a graph of velocity versus powder charge for several different loadings. I could get real clever and plot average velocity as well as min/max to get some kind of 'band' of velocity.

Open Office or LibreOffice will do something similar but I have never used them for this type of application.

Richard

 

[IndyGunFreak]

Now that you mention it, I found that Open Office Spreadsheet has a million different statistical functions. Since I never used it, it almost seems easier to enroll in a community college than to try to figure out how to use it. But, in 5 minutes of playing around, I got it to do the calculations. At this point, my system of napkins and online calculators seems easier, but maybe I should look into this spreadsheet thing.

I was gonna suggest this.

I use Open Office for all sorts of simple mathematical calculations. I also use it to calculate my costs on reloading (per round, per 100, per 1k).

You can do a lot of things w/ modern spread sheet software.

 

[Three-Five-Seven]

Extreme spread tells the story as well as, if not better than, standard deviation. An outlier will change SD but a little bit. However, an outlier will also loose a big match. So, for our purposes, extreme spread is probably a better metric.

It is a simple matter to detect the fastest and slowest rounds in a group and to subtract the latter from the former. If extreme spread is single digits, you've done all you can to maximize ballistic consistency.

 

[Boxerglocker]

I found this calculator and you simply type in each shot and it will calculate useful data for you. I'm sure there are others, but this one is the first one I clicked on and it worked well.

http://easycalculation.com/statistics/standard-deviation.php

This is the one I have bookmarked too... works well.

 

[WiskyT]

Extreme spread tells the story as well as, if not better than, standard deviation. An outlier will change SD but a little bit. However, an outlier will also loose a big match. So, for our purposes, extreme spread is probably a better metric.

It is a simple matter to detect the fastest and slowest rounds in a group and to subtract the latter from the former. If extreme spread is single digits, you've done all you can to maximize ballistic consistency.

I don't know that I disagree with you. SD is done because it makes some people happy, and it is free to do. For most people, really, even ES is not as important as "Low". Most shooters need a minimum velocity, for PF, for penetration and expansion.

But, a tight ES and SD do give a certain satisfaction. At least for me, when I see that my ES and SD are the same as factory ammo, or better, I know that my ammo is good enough.

 

[Khufu]

I don't know that I disagree with you. SD is done because it makes some people happy, and it is free to do. For most people, really, even ES is not as important as "Low". Most shooters need a minimum velocity, for PF, for penetration and expansion.

However, with standard deviation you can attempt to predict how often a "low" will occur.

For example, if minimum power factor is 6 standard deviations below the average velocity, there is roughly a 1 in 1Million chance that ay given round will be below the minimum PF. The the minimum PF is only 1 standard Deviation below the average (mean) there is about a 15% chance that any given round will be below the minimum PF.

You cannot use extreme spread to predict this.

 

[WiskyT]

However, with standard deviation you can attempt to predict how often a "low" will occur.

For example, if minimum power factor is 6 standard deviations below the average velocity, there is roughly a 1 in 1Million chance that ay given round will be below the minimum PF. The the minimum PF is only 1 standard Deviation below the average (mean) there is about a 15% chance that any given round will be below the minimum PF.

You cannot use extreme spread to predict this.

I'm not a statistics guy, so bear with me here. Let's say the minimum for PF is 1000fps. I shoot a big sample and the average is 1150 with a SD of 20. None of the shots ever goes below 1090. It seems to me that the odds of going below minimum are basically nil. The only way it could go below minimum is if there was some type of freak occurrence that statistics couldn't predict, like the powder measure malfunctioning.

 

[Khufu]

I'm not a statistics guy, so bear with me here. Let's say the minimum for PF is 1000fps. I shoot a big sample and the average is 1150 with a SD of 20. None of the shots ever goes below 1090. It seems to me that the odds of going below minimum are basically nil. The only way it could go below minimum is if there was some type of freak occurrence that statistics couldn't predict, like the powder measure malfunctioning.

Here is how to look at it:

1150-1000 = 150

150/20 = 7.5

So you are 7.5 standard deviation above the minimum. That means you have less than a one in a million chance of it falling below the minimum.

Of course, this assumes that you are loading correctly. If you forget powder or something, that isn't the same probability curve.

 

[Kentguy]

All I can say is that I’m glad you folks posted that on-line calculator. I seem to remember taking some sort of statistical analysis class of some kind in college but hell that was over 38 years ago… now-a-days I can’t remember what Idid 10 minutes ago let alone that many years back! :supergrin:

Very nice tool - thanks

 

[opus1776]

I could get a graph of velocity versus powder charge for several different loadings. I could get real clever and plot average velocity as well as min/max to get some kind of 'band' of velocity.

Richard

For the first part of the quote, I do this as part of my load development. As for the second part, I go even one step further.....

I have created an excel sheet so that I can generate a bar histogram that has counts on the y-axis and velocity on the x-axis(in bins of 25 ft/sec). Then using the calculated standard deviaton and average from the data set, I also overlay a normal distribution function on top of the histogram. This way I can get an easy visual of my load development. Usually I have at least 20 data points for a histogram.


Yes, for those of you who are wondering I am an Engineer. :wavey:


----------------------------------------------------------------
"Nothing tastes as good as skinny feels" K. Moss
You can NEVER be too rich or too thin...
Life is not a journey, but a series of unplanned detours....
Perfection: is not a goal---it's a demanded expectation.

 

[shotgunred]

It seems that with that much information I might have to admit the problem is me instead of the load.