Turbulent Turtle
F-TR competitor
FroggyOne2 said:
You think he's in for a shock? (big smile)
Calculating SD?
- Thread starter thefitter
- Start date
You think he's in for a shock? (big smile)
Dennis,
definitely a him! (My mother was a fan of Laurence Olivier the British actor in his wartime films.) Sadly the diminutive version of the name, Laurie, started to become feminised over here in the UK in the late 1980s. That's what I reckon as when I worked in marketing around then I started getting mail addressed to Ms Laurie Holland which always amused me. If this was preparatory to a meeting I think some people got a surprise alright at being greeted by an overweight, follickly challenged middle-aged male with decidedly non-PC attitudes. I've now progressed to retired old fart which is great as you can be rude about all sorts of things and people and usually get away with it!
Laurie
Keith Glasscock
Gold $$ Contributor
That was good for a laugh...
I agree that a statistically valid system for making reasonably accurate predictions would require quite a few bullets over the chronograph. By the nature of the beast, at least 40, and much better 100 rounds would have to be fired in as nearly identical conditions as possible.
That would be an expensive proposition. Not only in terms of barrel wear and components expended, but a climate controlled shooting range would be a couple dollars more than most of us can afford.
For example, here are the numbers from my 200 grain load. I only did a 10 shot string and it shows some of what Laurie is explaining.
2599
2601
2606
2608
2602
2608
2608
2598
2603
2604
Now, lets count each velocity:
2598 - 1
2599 - 1
2600 - 0
2601 - 1
2602 - 1
2603 - 1
2604 - 1
2605 - 0
2606 - 1
2607 - 0
2608 - 4
Now, I have an ES of 10 and an SD of 4 on this string, but I have a significant mode at 2608. The median would be 2605 and the mean (average) is 2603. This does not describe a bell curve - at least these 10 shots don't. The result at 1000 yards would be something like this (excluding all other variables)
4 shots in one location (we'll call this 0 for measuring)
1 shot .7" lower
1 shot 1.3" below
1 shot 1.6" below
1 shot 2" below
1 shot 2.3" below
1 shot 3.0" below
1 shot 3.3" below
So the total group would have been 3.3" high based solely on velocity variation. That is smaller than the size of the X ring at that range. As a matter of fact, if the ES is less than 15, the load will plot well inside the X ring vertically (G7 of .335). In reality, this small error would get lost in the noise of mirage, vertical winds, shooter errors, and BC variations.
As with so many other cases, getting the tune right and improving shooting skills will improve scores and group sizes much more than changing the ES or SD by a very small percentage of the total velocity.
I guess I'll have to go back to my old-fashioned way of doing things. If the vertical and score are good at 1000 yards, the load must be good.
I agree that a statistically valid system for making reasonably accurate predictions would require quite a few bullets over the chronograph. By the nature of the beast, at least 40, and much better 100 rounds would have to be fired in as nearly identical conditions as possible.
That would be an expensive proposition. Not only in terms of barrel wear and components expended, but a climate controlled shooting range would be a couple dollars more than most of us can afford.
For example, here are the numbers from my 200 grain load. I only did a 10 shot string and it shows some of what Laurie is explaining.
2599
2601
2606
2608
2602
2608
2608
2598
2603
2604
Now, lets count each velocity:
2598 - 1
2599 - 1
2600 - 0
2601 - 1
2602 - 1
2603 - 1
2604 - 1
2605 - 0
2606 - 1
2607 - 0
2608 - 4
Now, I have an ES of 10 and an SD of 4 on this string, but I have a significant mode at 2608. The median would be 2605 and the mean (average) is 2603. This does not describe a bell curve - at least these 10 shots don't. The result at 1000 yards would be something like this (excluding all other variables)
4 shots in one location (we'll call this 0 for measuring)
1 shot .7" lower
1 shot 1.3" below
1 shot 1.6" below
1 shot 2" below
1 shot 2.3" below
1 shot 3.0" below
1 shot 3.3" below
So the total group would have been 3.3" high based solely on velocity variation. That is smaller than the size of the X ring at that range. As a matter of fact, if the ES is less than 15, the load will plot well inside the X ring vertically (G7 of .335). In reality, this small error would get lost in the noise of mirage, vertical winds, shooter errors, and BC variations.
As with so many other cases, getting the tune right and improving shooting skills will improve scores and group sizes much more than changing the ES or SD by a very small percentage of the total velocity.
I guess I'll have to go back to my old-fashioned way of doing things. If the vertical and score are good at 1000 yards, the load must be good.
Keith Glasscock
Gold $$ Contributor
I had to run the ballistics for each shot individually with a fixed launch angle. Using JBM ballistics, you have to run the calculations for 1000 yard zero on the first shot and take note of the calculated elevation. Now, go back and put that elevation into the same name box on the inputs page. Uncheck the box that says "elevation correction for zero range."
Next, tell it to calculate again to make certain everything works.
Then, go back, and change the velocity without changing anything else. The 1000 yard trajectory should show a change in impact height.
It is a rough estimate, but it works.
Next, tell it to calculate again to make certain everything works.
Then, go back, and change the velocity without changing anything else. The 1000 yard trajectory should show a change in impact height.
It is a rough estimate, but it works.
Laurie commented: “What might be interesting in all this would be for a forum member who is a statistician or an engineer who uses this sort of data professionally to say what would be regarded as a good sample size in this sort of exercise.â€
I think I can qualify as someone who used ES and SD data every day as a professional. As I have, hopefully, explained below my answer to Laurie’s query would be: the sample size can never be too large.
(Laurie…one of my duty stations was RAF Chicksands, home of the
FLR-9 direction finding “elephant cage†with which you are probably familiar. Photos exist on the internet….see: http://www.bing.com/images/search?q=flr-9+photos&id=335E6A420847A558AD3348F64337C764FDFF5034&FORM=IQFRBA#x0y126. For details about the site, see: http://en.wikipedia.org/wiki/RAF_Chicksands.)
In my military career, I and my subordinates, peers and superiors relied on some fantastic gear to pin-point electronic transmitter target locations by “shooting†direction-finding bearings. Since the individual bearing “shots†naturally had an angular dispersion left or right or exactly on the target (think ES), results from different DF sites were triangulated. Our gear took hundreds of bearing “shots†at the target transmitter every second and gave us a “confidence†factor….expressed as SD…. that the target was at the indicated position. The larger/smaller the SD, the larger/smaller our confidence in the results indicating a true location.
The bearings and target location were good for only that particular point in time due to all the atmospheric and environmental conditions imaginable which continually affect electronic transmissions. The DF “shots†on a target could and would reflect a slightly or significantly different location from one time period to another.
So. Despite the hundreds of samplings per second, the target location had to be constantly refined over a period of time…..hours, days, weeks, months.
And so it goes with shooting bullets. The speed readings you get today will be different than those you get tomorrow due to the variables. Shoot and develop loads that have satisfactory, repeatable ES and are accurate under your shooting conditions. I suggest three separate outings of about 20 shots each with similar results.
BTW. My opinion of SD as it relates to shooting bullets is that it is pretty useless info except for putting a grin or frown on your face. The real tale to be told re shooting bullets (or shooting DF bearings) lies within the ES, especially if your target is beyond 300 yards. It’s the high and low velocities that can ruin your vertical past 300.
Frank B.
I think I can qualify as someone who used ES and SD data every day as a professional. As I have, hopefully, explained below my answer to Laurie’s query would be: the sample size can never be too large.
(Laurie…one of my duty stations was RAF Chicksands, home of the
FLR-9 direction finding “elephant cage†with which you are probably familiar. Photos exist on the internet….see: http://www.bing.com/images/search?q=flr-9+photos&id=335E6A420847A558AD3348F64337C764FDFF5034&FORM=IQFRBA#x0y126. For details about the site, see: http://en.wikipedia.org/wiki/RAF_Chicksands.)
In my military career, I and my subordinates, peers and superiors relied on some fantastic gear to pin-point electronic transmitter target locations by “shooting†direction-finding bearings. Since the individual bearing “shots†naturally had an angular dispersion left or right or exactly on the target (think ES), results from different DF sites were triangulated. Our gear took hundreds of bearing “shots†at the target transmitter every second and gave us a “confidence†factor….expressed as SD…. that the target was at the indicated position. The larger/smaller the SD, the larger/smaller our confidence in the results indicating a true location.
The bearings and target location were good for only that particular point in time due to all the atmospheric and environmental conditions imaginable which continually affect electronic transmissions. The DF “shots†on a target could and would reflect a slightly or significantly different location from one time period to another.
So. Despite the hundreds of samplings per second, the target location had to be constantly refined over a period of time…..hours, days, weeks, months.
And so it goes with shooting bullets. The speed readings you get today will be different than those you get tomorrow due to the variables. Shoot and develop loads that have satisfactory, repeatable ES and are accurate under your shooting conditions. I suggest three separate outings of about 20 shots each with similar results.
BTW. My opinion of SD as it relates to shooting bullets is that it is pretty useless info except for putting a grin or frown on your face. The real tale to be told re shooting bullets (or shooting DF bearings) lies within the ES, especially if your target is beyond 300 yards. It’s the high and low velocities that can ruin your vertical past 300.
Frank B.
Frank,
many thanks for your input. Your view above is what I've always thought about the matter - but many people seem to be much more impressed by SD than ES.
RAF Chicksands rang a faint bell, (from a mention in the novel 'The Fourth Protocol' by Frederick Forsyth, I suspect) but I'd not heard of the 'Elephant Cage', so your links were very interesting. Many thanks again.
Laurie
Keith Glasscock
Gold $$ Contributor
I think that people would be more directed toward ES if they fully understood what SD is in a normal distribution.
Roughly (instead of exactly):
68% of shots will have a velocity of + or - a SD (if SD is 4, then the ES for this group would be 8 )
95% of shots will have a velocity of + or - 2 SD (ES of 16 in above example)
99.5% of shots will be within + or - 3 SD (ES 24 in above example)
Lets go backward here-
68% will be +or- 4 fps
27% will be more than 4 but 8 or less fps above or below the average
4.5% will be more than 2 but less than or equal to 12 fps from the average
.5% (1 in 200) will be more than 12 fps from the average
So, for any 20 shot string, we would expect 13 to 14 to be within 4 fps, 5 or 6 more to be within 8 fps, and 1 to be between 8 and 12 fps above or below the average.
I would rather shoot a 20 shot string for ES. If the SD is truly correct (and useful), then the ES should be between 16 and 24 (likely 20 or so)
The chances of randomly selecting all 20 rounds from a box of 100 that are within 1 SD - reasonble.... The chance of selecting 10 that are really quite close - even better.
To me, the best way to determine the good loads from the bad is to use them in competition and see how they perform over a reasonably large number of rounds (say 100).
Roughly (instead of exactly):
68% of shots will have a velocity of + or - a SD (if SD is 4, then the ES for this group would be 8 )
95% of shots will have a velocity of + or - 2 SD (ES of 16 in above example)
99.5% of shots will be within + or - 3 SD (ES 24 in above example)
Lets go backward here-
68% will be +or- 4 fps
27% will be more than 4 but 8 or less fps above or below the average
4.5% will be more than 2 but less than or equal to 12 fps from the average
.5% (1 in 200) will be more than 12 fps from the average
So, for any 20 shot string, we would expect 13 to 14 to be within 4 fps, 5 or 6 more to be within 8 fps, and 1 to be between 8 and 12 fps above or below the average.
I would rather shoot a 20 shot string for ES. If the SD is truly correct (and useful), then the ES should be between 16 and 24 (likely 20 or so)
The chances of randomly selecting all 20 rounds from a box of 100 that are within 1 SD - reasonble.... The chance of selecting 10 that are really quite close - even better.
To me, the best way to determine the good loads from the bad is to use them in competition and see how they perform over a reasonably large number of rounds (say 100).
Amen Busdriver!
The subject of SD and how some look at it as the holy grail of shooting stats always gets my attention on this forum. Since I have responded to the issue several times, I decided to look...for the first time in 7 years....at all of my past postings to check out my previous comments. There were several on the SD subject.
I think my posting of 07/01/09 best defined my position, to wit:
"Get your ES into the single digits and the low SD will follow."
Frank B.
The subject of SD and how some look at it as the holy grail of shooting stats always gets my attention on this forum. Since I have responded to the issue several times, I decided to look...for the first time in 7 years....at all of my past postings to check out my previous comments. There were several on the SD subject.
I think my posting of 07/01/09 best defined my position, to wit:
"Get your ES into the single digits and the low SD will follow."
Frank B.
Keith Glasscock
Gold $$ Contributor
Quick note:
I was reviewing the data I presented earlier and realized that there was no way that the impact data was correct. I re-ran the ballistics and updated the impact data.
I corrected the original post with the correct information:
The total vertical at 1000 yards for a 10 ES with my load would be 3.3"
For some reason, I had gotten half that number the first time....
I appologise for any misunderstanding caused.
I was reviewing the data I presented earlier and realized that there was no way that the impact data was correct. I re-ran the ballistics and updated the impact data.
I corrected the original post with the correct information:
The total vertical at 1000 yards for a 10 ES with my load would be 3.3"
For some reason, I had gotten half that number the first time....
I appologise for any misunderstanding caused.
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.
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.