You can see in the attachment that the two groups I shot with 59 grains went in a 1/2 inch or less. The Sierra manual shows that load in the 2900 fps range out of their 26" barreled test rifle. My rifle has the common euro 60 cm barrel which is just about 23 11/16". I've reloaded 20 rounds in new Lapua brass and adjusted my scope a tad. I plan on shooting tomorrow, if I get the same 1/2 moa accuracy a third time I'll break out the chrony. If I'm in the 2800 fps range that will likely be the only load that rifle ever sees again so long as I have it.
Reloader 17 in 30 06
- Thread starter oldscot3
- Start date
Oldscot,
If you haven't played with statistics you may be unaware of something called Standard Error. In effect, it is the standard deviation of the mean. In other words, its a number that can be used to predict how much the centers of your groups will wander around if the number of shots in each group (sample size) is the same. It turns out to be equal to the population standard deviation divided by the square root of the sample size.
Below is a computer-generated 9 shot group with a bias to the left of center due to a simulated steady wind. Otherwise, it uses Excel's random number generator to generate the hole locations on each axis (bivariate normal distribution, meaning the vertical and horizontal are two independent bell curve distributions) and the standard deviations of each axis is the same as the other, so the group tends to be round when you make a big sample. The center of the group (the mean location, where the horizontal and vertical average values cross) is the red X. I then show the locations of the first three and second three and third three shots that made up the nine shot group. Each has its own red X. Notice how that X moves around. It's less than the group sizes change by a factor of the square root of three, but it definitely wanders, and I think that may be all you are seeing on your targets.
Also note the difference in the sizes of the second and third sets of three shots. It illustrates why taking just three shots can fail to reveal the true picture of a gun and load's precision and can fool us into thinking the combination is either better or worse than it actually is on average. The first group of nine makes the real precision of the weapon more apparent. Mind you, two nine-shot groups won't have the same exact size or mean location, either, but the probability is the differences will be much smaller than you see for any random pair of three-shot groups.
If you haven't played with statistics you may be unaware of something called Standard Error. In effect, it is the standard deviation of the mean. In other words, its a number that can be used to predict how much the centers of your groups will wander around if the number of shots in each group (sample size) is the same. It turns out to be equal to the population standard deviation divided by the square root of the sample size.
Below is a computer-generated 9 shot group with a bias to the left of center due to a simulated steady wind. Otherwise, it uses Excel's random number generator to generate the hole locations on each axis (bivariate normal distribution, meaning the vertical and horizontal are two independent bell curve distributions) and the standard deviations of each axis is the same as the other, so the group tends to be round when you make a big sample. The center of the group (the mean location, where the horizontal and vertical average values cross) is the red X. I then show the locations of the first three and second three and third three shots that made up the nine shot group. Each has its own red X. Notice how that X moves around. It's less than the group sizes change by a factor of the square root of three, but it definitely wanders, and I think that may be all you are seeing on your targets.
Also note the difference in the sizes of the second and third sets of three shots. It illustrates why taking just three shots can fail to reveal the true picture of a gun and load's precision and can fool us into thinking the combination is either better or worse than it actually is on average. The first group of nine makes the real precision of the weapon more apparent. Mind you, two nine-shot groups won't have the same exact size or mean location, either, but the probability is the differences will be much smaller than you see for any random pair of three-shot groups.
That's interesting, I may have to read it a few more times to absorb it all, but thanks for sharing it.
When firing this hunting rifle with its' relatively slender barrel, my goal is know where the first shot out of a cold barrel will impact. With that in mind, my three shot groups are really more like three, one shot groups fired at the same target. I spent between thirty and forty-five minutes on each. I shoot at my private setup where no one else is shooting or inconvenienced by me. I take one shot, get up and putter around with little chores, then return and and take another. I guess I could have just as well shot one round at three separate targets. I don't know how shooting nine might effect the results given my process, but I think it would be a good exercise to try. Again, the one target where impacts were a little lower and left, I can't account for given where the others printed, before and after those three, except that maybe statistically I need an even bigger sample. Maybe twenty or thirty rounds. This could take awhile, I better take a snickers with me.
In the meantime, I think I'll overlay three of my targets and see what they look like. Thanks again.
When firing this hunting rifle with its' relatively slender barrel, my goal is know where the first shot out of a cold barrel will impact. With that in mind, my three shot groups are really more like three, one shot groups fired at the same target. I spent between thirty and forty-five minutes on each. I shoot at my private setup where no one else is shooting or inconvenienced by me. I take one shot, get up and putter around with little chores, then return and and take another. I guess I could have just as well shot one round at three separate targets. I don't know how shooting nine might effect the results given my process, but I think it would be a good exercise to try. Again, the one target where impacts were a little lower and left, I can't account for given where the others printed, before and after those three, except that maybe statistically I need an even bigger sample. Maybe twenty or thirty rounds. This could take awhile, I better take a snickers with me.
In the meantime, I think I'll overlay three of my targets and see what they look like. Thanks again.