Your Groups Are Too Small

[INTJ]

In my BR rifle I have both the 215s and 200s jammed into the lands .021. They really like it there, but don't do that in a hunting rifle. For hunting rifles I have found that .020 to .030 jump usually works for the 215s. On occasion they like to be a lot deeper.

Also, the 156s, like any super long for caliber bullet, can be fussy.......but it's so cool when they work.......

Seems like every generalization we find has lots of exceptions.

As far as a 30 cal, I have come to really like the 300 PRC. It's only a couple grains more capacity than a 300 Win, but all the rifles you find for it have fast twist barrels and long magazine boxes. That said, a 300 Winny with a 3.6+ magazine and an 8-9 twist barrel is a great option.

 

[dieselchessy]

@Flyjunky

If you listen to some of their other podcasts they discuss load development processes different employees use.

#37 has some of it.

 

[Ryan Avery]

Ok I’m curious. @Formidilosus experience seems to align with Hornady’s data in the podcast.

Then Rokslide has at least one sponsor that does load development. @Unknown Munitions . Reading the load thread they “tune” and look for “nodes”.

I believe @Ryan Avery actually shoots some of the rifles during load development for Unknown.

So is that a waste of money? Is Form/Hornady wrong?

It's a modified "ladder test" at UM.

I've never done a ladder test for my rifles, and I do just fine at longer ranges. I have several one-shot kills on animals at over a thousand yards with a .7ish rifle.

I asked @INTJ to come on here to show the difference between @Formidilosus, which is closer to the way I loaded ammo, and the way bench rest guys load ammo. Both are very good at their craft. But for hunting, even long-range hunting, we chase our tails a lot and practice even less.

 

[Flyjunky]

In my BR rifle I have both the 215s and 200s jammed into the lands .021. They really like it there, but don't do that in a hunting rifle. For hunting rifles I have found that .020 to .030 jump usually works for the 215s. On occasion they like to be a lot deeper.

Also, the 156s, like any super long for caliber bullet, can be fussy.......but it's so cool when they work.......

Seems like every generalization we find has lots of exceptions.

As far as a 30 cal, I have come to really like the 300 PRC. It's only a couple grains more capacity than a 300 Win, but all the rifles you find for it have fast twist barrels and long magazine boxes. That said, a 300 Winny with a 3.6+ magazine and an 8-9 twist barrel is a great option.

I could be wrong but I believe during original load development it was seated at .020 off. I think it’s somewhere at 3.705” but I’d have to go look at my notes to be sure.

I’ve mostly shot the 140 in the 6.5-284 but I was able to pick up several thousand 156’s so I want to give that bullet a good testing the next few years. I’ve heard some inconsistencies in regards to in game performance and that it needs a touch more speed. My new build will have a 28” with 7.75 twist which should be able to push the 156’s around 3075+\- .

 

[eric1115]

I used to shoot non-round robin five shot groups at 1000 yds during tuning day. While that worked okay, my record groups during the match shrunk when I started shooting the round-robin "ladders"
that I showed.

The issue with shooting one group fully and then the next one and so on at 1000 yds is the conditions are always changing. So how do we know that group three didn't just happen to catch a very good or a very bad condition? Shooting round robin mitigates that. It means that if a condition is really good or really bad, it's not likely that it will make one group look good or bad. It will affect them all, and I have seen that. Also, it takes a lot less rounds to shoot the three shot round robin tests that to shoot multiple 5-shot groups.

We find that 10-shot groups don't tell us anymore about the load that 5-shot groups. 10-shot groups are more about the conditions. Remember, during the match we'd shoot 5 and 10 shot groups so we are getting a lot of feedback on our tuning approach. There are many matches where the same rifle is used for both classes, and the 5-shot groups are frequently around the same size as the 10.

Of course, there are also matches where the 10-shot groups are strung out horizontally across the target as well. That sucks, especially when you saw the condition change in the middle of your string, and then you have to decide if you stop, adjust, or keep going. I promise you that more than half the time, whatever you pick, is wrong--at least for me.......

Have you ever loaded and shot a poorer initial group to see if it scales? Like I'd be curious to see if a .4 moa 3 shot grows to .8 with the remaining 7 shots, but maybe the next test group the 3 test shots happened to be the outside edges at .8 and the last 7 all infill those first 3 for the exact same group size but the 3 shot test shows very different results. I hope that makes sense.

For a field rifle whose primary purpose to make a first round impact at a distance you don't know ahead of time, how do you view the balance between absolute maximum precision vs durable, consistent load that doesn't require frequent re-zeroing, re-validating, etc due to frequent changes in pursuit of smallest possible group size? Or am I not thinking about this in the correct terms?

 

[INTJ]

It's a modified "ladder test" at UM.

I've never done a ladder test for my rifles, and I do just fine at longer ranges. I have several one-shot kills on animals at over a thousand yards with a .7ish rifle.

I asked @INTJ to come on here to show the difference between @Formidilosus, which is closer to the way I loaded ammo, and the way bench rest guys load ammo. Both are very good at their craft. But for hunting, even long-range hunting, we chase our tails a lot and practice even less.

Chase your tail.....Bah!!! I have CAUGHT my tail........

You and I have talked a lot about how we don't need sub 1/2 MOA rifles to hunt with at long range. We have both said that when we get to .7 with our own hunting rifles we are happy.

I like 3-4 three shot groups of .7 better that are roughy to the same POI. I suppose a I MOA 10-shot group would work as well. I really like to validate the load at 300 yds--as long as I don't have to slog through knee high snow. However, customers want to see sub 1/2 MOA......

My only issue here is when guys tell me that it's not statistically possible to shoot well inside of that if we are willing to keep up with the tune. That's where statistics fail us, as my targets show.

 

[eric1115]

You're equally likely to put the 3 red bullets out of the sleeve as you are to pull the 3 yellow bullets, right?

How do you control for that chance? Is that where looking for the 2 adjacent charge weights with similar poi comes into play?

Clearly you're successfully adjusting the tuning of your load as your rifle changes, and I'm very interested in the process of sorting out the signal from the noise.

 

[INTJ]

Have you ever loaded and shot a poorer initial group to see if it scales? Like I'd be curious to see if a .4 moa 3 shot grows to .8 with the remaining 7 shots, but maybe the next test group the 3 test shots happened to be the outside edges at .8 and the last 7 all infill those first 3 for the exact same group size but the 3 shot test shows very different results. I hope that makes sense.

For a field rifle whose primary purpose to make a first round impact at a distance you don't know ahead of time, how do you view the balance between absolute maximum precision vs durable, consistent load that doesn't require frequent re-zeroing, re-validating, etc due to frequent changes in pursuit of smallest possible group size? Or am I not thinking about this in the correct terms?

I don't shoot poorer groups on purpose. We do sometimes track shots at 1000 yds, but those are 5 and 10 shot groups. In general, the more shots the larger the group, though like I mentioned I have shot many 10-shot groups that were better than my 5-shot groups.

I think you can see what you're asking if you shot a couple 10-shot groups. Inside of every 10-shot group is a very tight three-shot group.

For a field rifle I want a forgiving tune with as high a BC bullet that will shoot decently. I think it'a more important to practice wins skills that tune a hunting rifle below .7 MOA.

When I shoot's round robin I shooters each series in order. I have a port level ammo holder and I line up all the ammo. It's usually Bare, Red, Green, Yellow.

 

[gbflyer]

I’ve come to like 10 round groups on my sporting rifles. I’d never really thought about it much until I listened to the podcast and read some folks who do way more shooting than me. It’s way less frustrating than 2 touching and 1 out. If it grows to 1.5 or even 2, still all good.

Appreciate those willing to share.

 

[seand]

I’m not convinced that ladder testing or OCW can’t find a “node”, but I am convinced that the resolution needed to be sure of the conclusion is way higher than is being done by almost anyone. As in 10+shots at each charge weight. Or, alternativel, multiple ladder tests to verify which isn’t normally done either. I think the proof is assumed to be that the rifle shoots great at range after the ladder test, in reality it’s a really accurate rifle/load and is most likely to shoot great at charge weights outside of the assumed “node”. Which is really great, but not a lot was learned. More shooting is always good even if we are chasing our tails.

The validity of “my rifle shoots XX MOA” is pretty funny with no reference to round count. It sells rifles as even GAP does 3-round testing, as that is what customers expect. it makes the shooter confident I guess that’s great as long as they aren’t counting on it for practical field shooting. But the variability in field positions usually totally overwhelms the variability in mechanical accuracy anyway so even that doesn’t matter that much.

A lot of folks aren’t really understanding the data presented in the Hornady podcast. It’s pretty funny to see folks get twisted up about it, which was I think expected by the Hornady crew.

What do I tell my friends about this group? Do I say it was a 0.75moa rifle (based on 3,5 or 8 shots) , or a 1.2 MOA rifle (based on 13 shots)? It’s obviously not really a 0.75moa rifle all the time.

 

[dieselchessy]

What do I tell my friends about this group? Do I say it was a 0.75moa rifle (based on 3,5 or 8 shots) , or a 1.2 MOA rifle (based on 13 shots)? It’s obviously not really a 0.75moa rifle all the time.

You tell them its a "0.75 MOA rifle all day, if I do my part". The "If I do my part" portion is very important. Its the classification that allows your claim to not be an outright lie.
I thought everybody knew that by now.

 

[Flyjunky]

Just like before, here is the shot-by-shot breakdown, followed by the tabular breakdown of 2, 3, 4, 5, 10, 20, 30-shot groups

View attachment 500056

View attachment 500057

Median 2-shot group = 0.273 MOA
Median 3-shot group = 0.465 MOA
Median 4-shot group = 0.540 MOA
Median 5-shot group = 0.578 MOA
Median 10-shot group = 0.802 MOA
Median 20-shot group = 0.802 MOA
Median 30-shot group = 0.813 MOA
Overall 34-shot group = 0.8927 MOA

People should be out shooting and figure out what occurs with themselves and their “system”, but below is how my real world groups compare to the statistical scale

View attachment 500058

As should be expected, there is variance when scaling from 2-shot groups, likely due to shooter errors as well as the previously explained suppressor on/off cycles, small differences in environmentals & lighting, etc. The other differences between the tables is quite small from a big picture standpoint. I’ve gone through this exercise again and again, and my experience is that the statistical scale factors are quire close to what I observe shooting in the real world (barrel after barrel and various kinds of factory ammo).

In terms of group zero location, below is a table and some graphs showing the same thing. I broke the graphs out into elevation offset and windage offset since we are bound by our scopes ability to adjust back to zero. The thicker black line shows the respective median elevation and windage location for the overall 34 shot group, the green dash lines are +/-0.1 mil from that 34-shot median location, and then the scatter plot is the zero location of each 3, 4, 5, and 10-shot group.

You can see that all 3-shot groups are within 0.1 mil of the 34-shot group median elevation, with the exception of one group that is right at +0.1 mil. All of the other 4, 5, and 10-shot groups are within 0.1 mil of overall group location, and most are within 0.05 mils of overall group location. Similar results for the windage as well.

View attachment 500059

View attachment 500060

View attachment 500061

Below is a graphic that shows how far the running group count “zero” location is relative to the total 34-shot group location.

View attachment 500062

Again, just sharing data. My real world experience is that statistics is a real thing, 5-shot groups are sufficient for field shooting in terms of zeroing & evaluating mechanical precision, and factory ammo is sufficient to produce 30+ shot sub-moa groups.

Well done, thank you.

 

[Lawnboi]

@TK-421

Would your opinion on the 5 round group being adequate be the same if say that 30 round total ended up at 1.5-2”?

Iv noticed that my rifles that don’t shoot as well take more rounds to zero. And my rifles with good barrels on them take much fewer rounds zeroing to build confidence. Zeroing meaning getting to a point that I’m confident enough in that zero to March it out.

You have obviously put your money where your mouth is with the aggregate groups, which are awsome, and imo can be achievable from a good shooter, shooting good equipment. Heck a group over 1moa out of my match gun be it 3 or 12 shots is unacceptable to me. That said I also have rifles that don’t shoot like that, rifles I firmly believe I am outshooting.

 

[pyrotechnic]

@TK-421

Would your opinion on the 5 round group being adequate be the same if say that 30 round total ended up at 1.5-2”?

Iv noticed that my rifles that don’t shoot as well take more rounds to zero. And my rifles with good barrels on them take much fewer rounds zeroing to build confidence. Zeroing meaning getting to a point that I’m confident enough in that zero to March it out.

You have obviously put your money where your mouth is with the aggregate groups, which are awsome, and imo can be achievable from a good shooter, shooting good equipment. Heck a group over 1moa out of my match gun be it 3 or 12 shots is unacceptable to me. That said I also have rifles that don’t shoot like that, rifles I firmly believe I am outshooting.

I'm going to take a stab at this and say that if you your adjustment resolution (0.1 mil or whatever) is smaller than the uncertainty of your zero location then you will need to shoot larger groups to center them to the resolution of your optic adjustments.

So if you shoot 5 shot 2 MOA groups you might need to increase the number of shots to say "I am centered to the nearest 0.1 mil" whereas if you shot tighter groups you wouldn't need to.

@TK-421
You know that Tangent was just constantly shifting its zero enough to get all your other errors to center.

 

[gbflyer]

Just like before, here is the shot-by-shot breakdown, followed by the tabular breakdown of 2, 3, 4, 5, 10, 20, 30-shot groups

View attachment 500056

View attachment 500057

Median 2-shot group = 0.273 MOA
Median 3-shot group = 0.465 MOA
Median 4-shot group = 0.540 MOA
Median 5-shot group = 0.578 MOA
Median 10-shot group = 0.802 MOA
Median 20-shot group = 0.802 MOA
Median 30-shot group = 0.813 MOA
Overall 34-shot group = 0.8927 MOA

People should be out shooting and figure out what occurs with themselves and their “system”, but below is how my real world groups compare to the statistical scale

View attachment 500058

As should be expected, there is variance when scaling from 2-shot groups, likely due to shooter errors as well as the previously explained suppressor on/off cycles, small differences in environmentals & lighting, etc. The other differences between the tables is quite small from a big picture standpoint. I’ve gone through this exercise again and again, and my experience is that the statistical scale factors are quire close to what I observe shooting in the real world (barrel after barrel and various kinds of factory ammo).

In terms of group zero location, below is a table and some graphs showing the same thing. I broke the graphs out into elevation offset and windage offset since we are bound by our scopes ability to adjust back to zero. The thicker black line shows the respective median elevation and windage location for the overall 34 shot group, the green dash lines are +/-0.1 mil from that 34-shot median location, and then the scatter plot is the zero location of each 3, 4, 5, and 10-shot group.

You can see that all 3-shot groups are within 0.1 mil of the 34-shot group median elevation, with the exception of one group that is right at +0.1 mil. All of the other 4, 5, and 10-shot groups are within 0.1 mil of overall group location, and most are within 0.05 mils of overall group location. Similar results for the windage as well.

View attachment 500059

View attachment 500060

View attachment 500061

Below is a graphic that shows how far the running group count “zero” location is relative to the total 34-shot group location.

View attachment 500062

Again, just sharing data. My real world experience is that statistics is a real thing, 5-shot groups are sufficient for field shooting in terms of zeroing & evaluating mechanical precision, and factory ammo is sufficient to produce 30+ shot sub-moa groups.

I believe that is the most “nerded - out” thing I have ever seen in my life regarding group size, and I don’t mean that in a bad way. I certainly wouldn’t wish to be shooting against you at a match.

Thanks for sharing.

 

[Macintosh]

@TK-421 , thanks for this. You and I also talked about this via PM a while back, and my targets had some issues outside of the topic, but I still struggle with the concept. I think I have similar questions to a few of the folks just above^^, and I am not sure if that's what you are saying or not.
--Looking at your last aggregate target (all 30+ rounds), what is preventing your 5-round group from being my cherrypicked 5 rounds that are all far-right of the group center (or far-top of the group center, etc), and therefore their center would be off a full click or more from actual center of your extreme spread?
--If you did the same excercise in a gun that held 1.8moa (about double your aggregate .88moa), wouldnt the center of those same 5 cherrypicked rounds potentially be off double the amount from center of cone as in your target?
If so, it seems very reasonable to me that as the extreme spread increases, so must the group size needed to find the center of that cone with X% confidence? Is that what you are saying, or are you saying something different?

 

[Macintosh]

Thank you, I think that answers my question. I want to make sure I'm understanding correctly, as I am not even close to a statistics guy. Is it correct to capture what you are saying as: the reason your 5-shot group doesn't wind up being "the 5 worst shots all grouped together as far from cone-center as possible", is because the probability of any 5 consecutive shots landing there together is so low as to be "swamped" by the far higher probability of them landing pretty close to within the normal larger-group distribution...i.e. it's theoretically possible, but the probability of it happening 5 times in a row is so low that it's hardly worth considering?

To quantify this a bit and maybe learn something, looking at your chart, there is a section listed at the far Right of the bell curve at 1.7%, then .5%, then .1%---so in a normal distribution any one shot has a 2.3% chance of landing that far from group center, correct? So is the probability of that happening 5 times in a row 2.3% x2.3%x2.3%x2.3%x2.3%...or .0000006% chance of all 5 shots in one group landing there together?

And, do shooting groups fall into a Normal distribution bell curve like this? I'd be really curious to see the actual quantified bell curve of shot distribution for a 30-round group from a 1.25 or 1.5moa rifle and see what the actual % odds are of having a 5-round group land more than .1mil (i.e. measureably off) and more than .2mil (or whatever we deem significant for a normal-ish hunting distance) from cone center. I dont know how to plot that. Regardless, at that point I think I'm geeking out on maths, not shooting!

 

[mtnwrunner]

I believe that is the most “nerded - out” thing I have ever seen in my life regarding group size, and I don’t mean that in a bad way. I certainly wouldn’t wish to be shooting against you at a match.

Thanks for sharing.

Makes my head hurt.

Randy

 

[Macintosh]

"weaponized math"

 

[wind gypsy]

Thank you, I think that answers my question. I want to make sure I'm understanding correctly, as I am not even close to a statistics guy. Is it correct to capture what you are saying as: the reason your 5-shot group doesn't wind up being "the 5 worst shots all grouped together as far from cone-center as possible", is because the probability of any 5 consecutive shots landing there together is so low as to be "swamped" by the far higher probability of them landing pretty close to within the normal larger-group distribution...i.e. it's theoretically possible, but the probability of it happening 5 times in a row is so low that it's hardly worth considering?

To quantify this a bit and maybe learn something, looking at your chart, there is a section listed at the far Right of the bell curve at 1.7%, then .5%, then .1%---so in a normal distribution any one shot has a 2.3% chance of landing that far from group center, correct? So is the probability of that happening 5 times in a row 2.3% x2.3%x2.3%x2.3%x2.3%...or .0000006% chance of all 5 shots in one group landing there together?

And, do shooting groups fall into a Normal distribution bell curve like this? I'd be really curious to see the actual quantified bell curve of shot distribution for a 30-round group from a 1.25 or 1.5moa rifle and see what the actual % odds are of having a 5-round group land more than .1mil (i.e. measureably off) and more than .2mil (or whatever we deem significant for a normal-ish hunting distance) from cone center. I dont know how to plot that. Regardless, at that point I think I'm geeking out on maths, not shooting!

I'd have to watch/listen to the Hornady video again but it seems that @TK-421's data shows better and more consistent accuracy (distance from POA) than the data provided by Hornady. His results are impressive. It'd take me some time to conclusively wrap my head around what the take aways should be in comparison to what Hornady is saying. It does seem that he and his equipment are just better than most and less likely to vary both in precision and accuracy.