Shootin4fun
Well-Known Member
Me too, always 5 except when wanting to tiptoe into something. A reasonable sample size, and one less variable to consider when analyzing results.
-
If you are being asked to change your password, and unsure how to do it, follow these instructions. Click here
Garmin Xero - update
- Thread starter Varmint Hunter
- Start date
Help Support Long Range Hunting Forum
Shootin4fun
Well-Known Member
What flavor
TapiocaWhat flavor
Shootin4fun
Well-Known Member
Not too surprising. Garmin seems to be cratering LR sales. A year ago your offer probably wouldn't have lasted an hour. Even priced same as MS, the buyer must desire to work with that unit configuration instead in order to gain 2 things: slightly better accuracy of measurement and nothing to affect grouping. I would have been that buyer back when I bought my MS. But now, if I were to go Doppler, I would spend the $600 for the Garmin.
Fortunately for MS, I don't think there will be enough second hand LRs on the market to disrupt their sales much.
Hard to know how all this is affecting old tech models sales like Chrony and rcbs.
Fortunately for MS, I don't think there will be enough second hand LRs on the market to disrupt their sales much.
Hard to know how all this is affecting old tech models sales like Chrony and rcbs.
Varmint Hunter
Well-Known Member
I paid $600 for my bluetooth LR. Add on the ext battery, tripod, hard case, inertia trigger, aiming device and you're well over $800. I sold it for $400.
Sometimes you've just got to bite the bullet, so to speak. LOL
Sometimes you've just got to bite the bullet, so to speak. LOL
Anyone still looking for one Impact Data Book has 8 in stock and ready to ship. They also should be getting another 20 in a week.
Justice1327
Well-Known Member
Discussion on standard deviation differences between labradar and Garmin
Last edited:
cowboy
Well-Known Member
Most complete, precise and easy to understand on the criticisms of SD's differences between the 2 units.
Thank you.
Thank you.
Dthomas3523
Member
Hello, did you ever get a response from Garmin? Curious to hear their opinion of why they chose STDEV.P if they defend their choice.
Thanks
KsKevin
Well-Known Member
Garmin is doing it right as it's based on the entire population of the data set (not a random sampling of the data set). The explanation is really needed from Lab Radar as to why they chose to do it incorrectly.
Attachments
Dthomas3523
Member
Every chrono except garmin is using sample. Labradar, Magneto, Fx, Oehlar....etc.
Reason being is that it's still considered a sample if you intend to make more ammo based on the recipe and process used for the ammo you tested. This logic has been run by several with graduate level statistics degrees and they all agree using sample is appropriate.
You can get into a debate about the philosophy of sample and population. Regardless, Garmin is the current outlier. And an outlier in the sense that everyone else for 20 years has been using STDEV.S. So, it would be logical the outlier should be explaining why the industry standards are wrong.
Last edited:
Dthomas3523
Member
Here is the general reasoning that most seem to agree with:
Since you're taking a sample of five rounds from each recipe group, the data collected from each group would represent a sample. In this case, it's appropriate to use the sample standard deviation to analyze the variation in the speeds within each group of five rounds.
The reason for using sample SD over population SD is rooted in the nature of your data. The population SD is used when you have data for an entire population, which means you have information for every single member of that population. However, in most practical scenarios, it's not feasible to collect data from every single item in a population. Instead, you collect a smaller subset and use statistics calculated from that sample to estimate the parameters of the larger population.
When you're analyzing the speeds of the five rounds within each group, you would calculate the sample standard deviation to understand the spread or variability in the velocities measured from those particular five rounds.
To summarize, in your scenario, use sample standard deviation for each group of five rounds because you're working with samples (not the entire population) and want to understand the variation within each specific group of rounds.
Since you're taking a sample of five rounds from each recipe group, the data collected from each group would represent a sample. In this case, it's appropriate to use the sample standard deviation to analyze the variation in the speeds within each group of five rounds.
The reason for using sample SD over population SD is rooted in the nature of your data. The population SD is used when you have data for an entire population, which means you have information for every single member of that population. However, in most practical scenarios, it's not feasible to collect data from every single item in a population. Instead, you collect a smaller subset and use statistics calculated from that sample to estimate the parameters of the larger population.
When you're analyzing the speeds of the five rounds within each group, you would calculate the sample standard deviation to understand the spread or variability in the velocities measured from those particular five rounds.
To summarize, in your scenario, use sample standard deviation for each group of five rounds because you're working with samples (not the entire population) and want to understand the variation within each specific group of rounds.
Dthomas3523
Member
This basically just comes down to semantics and philosophy.
Let's say you take 10 different recipes for a cartridge and make five rounds each. Each of those five rounds makes up the entire population of that ammo for now. You run them over your chrono....you can use population SD under the definition above. However, once you choose one or more recipe and start making more ammo as close to the same as you can, your first five rounds becomes a sample of a much larger population.
So, most chronograph manufacturers use STDEV.S since it is still fine to use knowing that you will very likely be making more ammo. As well as it adds in more variability in the form of N-1.
So, it boils down to a philosophical difference in whether you consider the "now" to be all that matters. Which there doesn't seem to be a consensus on. You'll find many graduate level or better professionals who will agree that you can use STDEV.S for your ammo testing since the population of at least some of your ammunition will become larger.
The most important part isn't actually using P or S. It's that you use the same to compare ammo. I.E. you wouldn't want to run 85.5gr recipe over a Labradar and 90.0 gr over a garmin and use the two outputs to compare the ammo as they are using different formula.
Let's say you take 10 different recipes for a cartridge and make five rounds each. Each of those five rounds makes up the entire population of that ammo for now. You run them over your chrono....you can use population SD under the definition above. However, once you choose one or more recipe and start making more ammo as close to the same as you can, your first five rounds becomes a sample of a much larger population.
So, most chronograph manufacturers use STDEV.S since it is still fine to use knowing that you will very likely be making more ammo. As well as it adds in more variability in the form of N-1.
So, it boils down to a philosophical difference in whether you consider the "now" to be all that matters. Which there doesn't seem to be a consensus on. You'll find many graduate level or better professionals who will agree that you can use STDEV.S for your ammo testing since the population of at least some of your ammunition will become larger.
The most important part isn't actually using P or S. It's that you use the same to compare ammo. I.E. you wouldn't want to run 85.5gr recipe over a Labradar and 90.0 gr over a garmin and use the two outputs to compare the ammo as they are using different formula.
