patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
What is the differance between Es and sad ?what is a good number to get them to ?and how do I get to a good number?
-
If you are being asked to change your password, and unsure how to do it, follow these instructions. Click here
Es and sd
- Thread starter patrick021
- Start date
Help Support Long Range Hunting Forum
FearNoWind
Well-Known Member
Es (Extreme Spread) is the difference between the highest and lowest muzzle velocity in any test data group. If your MV is 2900 -2950 your ES is 50.
SD (Standard Deviation) is a avalue assigned to the dispersion in a group of numbers as measured from its mean. Put more directly, SD is the square root of the variance in a data set.
https://www.google.com/search?q=SD+...UIBygB&biw=1231&bih=895#imgrc=EWj_RcUArXWHWM:
The greater the ES, the higher the SD.
The answer to " how do I get to a good number?" in each of those categories should now be self evident.
I generally work to obtain a SD below 12 before I begin serious fine tuning of any load.
SD (Standard Deviation) is a avalue assigned to the dispersion in a group of numbers as measured from its mean. Put more directly, SD is the square root of the variance in a data set.
https://www.google.com/search?q=SD+...UIBygB&biw=1231&bih=895#imgrc=EWj_RcUArXWHWM:
The greater the ES, the higher the SD.
The answer to " how do I get to a good number?" in each of those categories should now be self evident.
I generally work to obtain a SD below 12 before I begin serious fine tuning of any load.
dsculley
Well-Known Member
ES = extreme spread. This is the difference between the two extremes (fastest and slowest) of a string of shots through your chronograph. The only thing this tells you is that difference for that string on that day.
SD = Standard deviation. This is the expected deviation for a given confidence interval based on a sample of the population. The sample is the string of shots. Based on the standard deviation and assuming a normal distribution (bell curve) you can expect 68% of your shots to fall within plus or minus 1 standard deviation, 95.5% to fall within plus or minus 2 standard deviations and 99% to fall within plus or minus 3 standard deviations.
Standard deviation is a much better indicator of your actual muzzle velocity.
Using a random number generator, I got the following numbers (assume they are your 5 shot string).
2839
2859
2887
2832
2847
The average is 2852.8, extreme spread is 55, but SD is 21.59. So based on this sample, you can expect velocity to range between 2788 and 2917.6. That is a very large spread, and much greater than the ES of 55 for the sample.
I just used random numbers generated between 2800 and 2900 so you can see that the predicted values were very close to my parameters. Pay more attention to your SD and remember that if you want to predict your velocity range for 99% of your shots, add 3 SD's to your average for your max, and subtract 3 SD's for your minimum.
The way you get the numbers lower is through precision reloading. You have to try and make every round you load the same. Pay attention to neck tension, annealing the necks when needed. Make sure your charges are the same. You may want to weigh a sample of your brass to see if you need to sort your brass (you can use the same statistical sampling for checking the SD of your brass weight/volume). Test your seating depth as well as your powder charge to determine the best load for your rifle. Read all you can on here, there are some very good people here who have written a lot of good information. Use the search feature or just browse the reloading section. You can learn a lot.
This can be a lot of fun, and you can get as deep into as you wish. Or, you may be satisfied to make plinking rounds. How precise you want to be is up to you.
It all starts with your case and how you handle it. Primer selection, powder selection and load amount, bullet type and seating depth, making your ammo concentric - all of these are part of the puzzle. Have fun and keep trying to get better. Since you posted a question (and a good one) I think you want to learn. Just keep asking questions.
FWIW,
Dennis
SD = Standard deviation. This is the expected deviation for a given confidence interval based on a sample of the population. The sample is the string of shots. Based on the standard deviation and assuming a normal distribution (bell curve) you can expect 68% of your shots to fall within plus or minus 1 standard deviation, 95.5% to fall within plus or minus 2 standard deviations and 99% to fall within plus or minus 3 standard deviations.
Standard deviation is a much better indicator of your actual muzzle velocity.
Using a random number generator, I got the following numbers (assume they are your 5 shot string).
2839
2859
2887
2832
2847
The average is 2852.8, extreme spread is 55, but SD is 21.59. So based on this sample, you can expect velocity to range between 2788 and 2917.6. That is a very large spread, and much greater than the ES of 55 for the sample.
I just used random numbers generated between 2800 and 2900 so you can see that the predicted values were very close to my parameters. Pay more attention to your SD and remember that if you want to predict your velocity range for 99% of your shots, add 3 SD's to your average for your max, and subtract 3 SD's for your minimum.
The way you get the numbers lower is through precision reloading. You have to try and make every round you load the same. Pay attention to neck tension, annealing the necks when needed. Make sure your charges are the same. You may want to weigh a sample of your brass to see if you need to sort your brass (you can use the same statistical sampling for checking the SD of your brass weight/volume). Test your seating depth as well as your powder charge to determine the best load for your rifle. Read all you can on here, there are some very good people here who have written a lot of good information. Use the search feature or just browse the reloading section. You can learn a lot.
This can be a lot of fun, and you can get as deep into as you wish. Or, you may be satisfied to make plinking rounds. How precise you want to be is up to you.
It all starts with your case and how you handle it. Primer selection, powder selection and load amount, bullet type and seating depth, making your ammo concentric - all of these are part of the puzzle. Have fun and keep trying to get better. Since you posted a question (and a good one) I think you want to learn. Just keep asking questions.
FWIW,
Dennis
patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
I'm having a hard time figuring sd how did you come up with 21.59 for the sad?
patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
Sorry sd
dsculley
Well-Known Member
I used the standard deviation function in Excel. I just ran it again and got the same result. Many chronographs will calculate it for you.
My example was an extreme range. Statistics can be used in lots of ways in shooting. If you don't have it, get the Berger reloading manual and read Bryan Litz's section on statistics.
My point was that extreme spread is valid only for that string. Standard deviation models the entire population and can be used to determine your expected extreme range over the entire population.
Dennis
My example was an extreme range. Statistics can be used in lots of ways in shooting. If you don't have it, get the Berger reloading manual and read Bryan Litz's section on statistics.
My point was that extreme spread is valid only for that string. Standard deviation models the entire population and can be used to determine your expected extreme range over the entire population.
Dennis
If you don't have excel here is a quick online calculator. Just input speeds separated by comma and hit calc button
https://www.easycalculation.com/statistics/standard-deviation.php
https://www.easycalculation.com/statistics/standard-deviation.php
Great information you guys !! And thanks for the link. I to have been trying to figure out this exact issue. Thank you ! Thank you and thank you !!
patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
I shot today my Es was 20 with a sd of 8.66 my groups were around 3 in at 418 yds I am tiring to get my groups as small as I can
Sounds like you're getting there. Good work
patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
Thanks
dsculley
Well-Known Member
Now that you have learned a bit about statistics, you can start applying it to your target and not just MV. You can take the group you shot with the SD of 8.66.
If your target doesn't have graph lines, you can take a square and draw a horizontal line under and vertical line beside your group. Measure the vertical and horizontal distance of each shot from the lines and record them. EG- I just started working on load development for a new build. I did this with my best group and got the following:
Vert Horiz
Shot 1 0.443 0.773
Shot 2 0.542 0.180
Shot 3 1.005 0.385
(Only 3 shots, not really enough but will have to suffice for this demonstration.)
Subtract the smallest from the largest for each and you get Vert spread of 0.562 and Horiz spread of 0.593. It is a bit under but we well call it a 0.6" group. Now that sounds promising doesn't it? Based on this, many would call this a 0.6" gun.
Now lets apply a bit of statistics. Using these figures, I calculated the SD for both vertical and horizontal. and got .030" for each. 99% of the shots should fall within an area 3 SD's above the average and 3 SD's below the average. For this example I don't need to compute the averages because all I am concerned about is the total spread, which is the SD x 6 or 0.30 x 6 = 1.80. 1.80" doesn't sound nearly as good does it? As I said earlier, 3 shots is not really enough to be statistically significant. I will keep this information and combine it with future shots to keep adding to the sample. Now suppose that I fire 7 more rounds and keep them withing the 0.6" original group. I used some random numbers and recomputed base on a ten shot group that stayed within the spread of this group and got SD of 0.195. 6 x 0.195 = 1.17. Now that is sounding more promising!
The reason for this exercise is to show that the same principles that you use to evaluate muzzle velocity can be used in other practical ways as well. It is also a bit of an eye opener when you start applying this to your targets.
There is another observation you can make when using this. See how many shots are above the average and how many are below the average. If your hand loading technique is good and your use a good, repeatable shooting position so that your shooting technique is "precision" then you should see as many shots above as below the average. If not, something is "out of control". (Pulled shots, unsteady shooting position, loose scope, loose action screws, etc)
FWIW,
Dennis
If your target doesn't have graph lines, you can take a square and draw a horizontal line under and vertical line beside your group. Measure the vertical and horizontal distance of each shot from the lines and record them. EG- I just started working on load development for a new build. I did this with my best group and got the following:
Vert Horiz
Shot 1 0.443 0.773
Shot 2 0.542 0.180
Shot 3 1.005 0.385
(Only 3 shots, not really enough but will have to suffice for this demonstration.)
Subtract the smallest from the largest for each and you get Vert spread of 0.562 and Horiz spread of 0.593. It is a bit under but we well call it a 0.6" group. Now that sounds promising doesn't it? Based on this, many would call this a 0.6" gun.
Now lets apply a bit of statistics. Using these figures, I calculated the SD for both vertical and horizontal. and got .030" for each. 99% of the shots should fall within an area 3 SD's above the average and 3 SD's below the average. For this example I don't need to compute the averages because all I am concerned about is the total spread, which is the SD x 6 or 0.30 x 6 = 1.80. 1.80" doesn't sound nearly as good does it? As I said earlier, 3 shots is not really enough to be statistically significant. I will keep this information and combine it with future shots to keep adding to the sample. Now suppose that I fire 7 more rounds and keep them withing the 0.6" original group. I used some random numbers and recomputed base on a ten shot group that stayed within the spread of this group and got SD of 0.195. 6 x 0.195 = 1.17. Now that is sounding more promising!
The reason for this exercise is to show that the same principles that you use to evaluate muzzle velocity can be used in other practical ways as well. It is also a bit of an eye opener when you start applying this to your targets.
There is another observation you can make when using this. See how many shots are above the average and how many are below the average. If your hand loading technique is good and your use a good, repeatable shooting position so that your shooting technique is "precision" then you should see as many shots above as below the average. If not, something is "out of control". (Pulled shots, unsteady shooting position, loose scope, loose action screws, etc)
FWIW,
Dennis
patrick021
Well-Known Member
- Joined
- Nov 17, 2012
- Messages
- 147
Could you post a picture of what you are talking about where I could better understand what you are saying thanks
dsculley
Well-Known Member
Sorry that I haven't gotten back sooner, but between work and some severe weather last week got really crazy. I will try to get some pics/explanations posted tonight.
Dennis
Dennis
