3 shot group vs ladder tests

[Kracken McGraw]

And others apparently....
They probably didt pick up on your "trick" because who would expect someone to do that? And why? Obviously your groups were inconsistent enough that guys thought they were different loads.

To make other people look foolish in his own mind is my opinion.

 

[old_heli_logger]

...some people's kids...

 

[seymour fish]

To make other people look foolish in his own mind is my opinion.

Dark Triad, self-proclaimed. When somebody tells you what they are, believe it.

 

[seymour fish]

I think you watched a very different video series than the rest of us.

“The rest of us” ? Your tells are obvious. When you confessed to manipulative “gotcha” behavior, we already knew. A fool and his cover story soon part

 

[JFrank]

Point, counter point, counter counter point.

I’ve seen a few tests posted that guys suggest seating is off or powder is obviously wrong but when I look at it,,,they all look the same so I wonder if I’m just too stupid to see the cat in the bush or what the F.
For myself .. short range OCW tests are the toughest to interpret.

 

[Varminterror]

And why?

Because shooting two replicants of the same “test,” one with an actual ladder (a manipulated variable) and one as a control, then asking in the post “can you tell which is which” would take twice as many rounds, and the ammo I did use for the “test” was simply left overs - it existed, so I used it to answer a question for which I had suspected I knew the answer for several years. Satisfying idle curiosity. No malice, and for their good will participation, I didn’t call “gotcha” to the respondents. But the social experiment is what is is - folks want to find patterns whether a pattern exists or not, so they did, when it didn’t.

 

[Evlshnngns]

Because shooting two replicants of the same “test,” one with an actual ladder (a manipulated variable) and one as a control, then asking in the post “can you tell which is which” would take twice as many rounds, and the ammo I did use for the “test” was simply left overs - it existed, so I used it to answer a question for which I had suspected I knew the answer for several years. Satisfying idle curiosity. No malice, and for their good will participation, I didn’t call “gotcha” to the respondents. But the social experiment is what is is - folks want to find patterns whether a pattern exists or not, so they did, when it didn’t.

Post the targets.

 

[FTR_Mojo]

I'm a bit disappointed by the personal attacks and vitriol being directed at Varminterror. He's been nothing but polite, honest and knowledgeable with his posts. I think he's made some incredibly interesting points about stats and the way our minds perceive them differently.
I for one have found this discussion fascinating and really helpful. There are heaps of golden nuggets of information!

I would however love to see those images you sent to the other forum! I'm really intrigued to see if Mr Wheeler is right in his idea that the gun was just shooting big. Or was it something else people thought they saw....

 

[Thauglor]

I've shot a lot of ladders for load dev. I've seen a lot of other peoples ladders. None look like that. They look more like this:

View attachment 1673339

I picked 44.5 from that ladder and went on to seating depth. I know, I know, it's a trash method and doesn't work. But it does. At least for the disciplines I participate in.

Where is your actual data? I doubt you shot in 0.01 grain increments to get sich a smooth line, and if you are interpolating with anything other than straight lines you are deceiving yourself

 

[LVLAaron]

Where is your actual data? I doubt you shot in 0.01 grain increments to get sich a smooth line, and if you are interpolating with anything other than straight lines you are deceiving yourself

.1 increments, just like the image shows.

 

[Steve Donlon]

Because shooting two replicants of the same “test,” one with an actual ladder (a manipulated variable) and one as a control, then asking in the post “can you tell which is which” would take twice as many rounds, and the ammo I did use for the “test” was simply left overs - it existed, so I used it to answer a question for which I had suspected I knew the answer for several years. Satisfying idle curiosity. No malice, and for their good will participation, I didn’t call “gotcha” to the respondents. But the social experiment is what is is - folks want to find patterns whether a pattern exists or not, so they did, when it didn’t.

That's why I don't put much validity in ladder tests. I have tried and would get different outcomes each time. I think you have to be a very consistent shot, and many of you are, but many are not. So I use other methods, like one step at a time.

 

[JFrank]

It’s hard to argue against a guy using the method he’s most comfortable with. As long as the end result matches the goal, then why not, there’s several ways around the barn.

 

[TC260]

I'm a bit disappointed by the personal attacks and vitriol being directed at Varminterror. He's been nothing but polite, honest and knowledgeable with his posts. I think he's made some incredibly interesting points about stats and the way our minds perceive them differently.
I for one have found this discussion fascinating and really helpful. There are heaps of golden nuggets of information!

I would however love to see those images you sent to the other forum! I'm really intrigued to see if Mr Wheeler is right in his idea that the gun was just shooting big. Or was it something else people thought they saw....

Agreed on all points.

It would be nice to see his "test" but to be honest I'm much more interested in replicating the test myself. When Hornady intially posted their info on large sample sizes, I understood what they were saying but it didn't have that much impact. It wasn't until I took the time to actually replicate the tests they were doing that the full impact of what they were saying really hit me. I know it's a highly unpopular topic so I never talk about it here but it completely changed the way I look at things. Given the high variability of small sample sizes I'm not surprised by Varminterror's "test" results.

 

[KMW 1954]

To the subject of posting "targets", I did that once long ago on a few different forums to try and get "opinions" of a load I was working on for a new rifle/cartridge combination. I finally got frustrated and quit responding to it because I was not getting any form of consensus. Had as many different opinions as I did responses. Everyone seemed to read them differently.

 

[CharlieNC]

Agreed on all points.

It would be nice to see his "test" but to be honest I'm much more interested in replicating the test myself. When Hornady intially posted their info on large sample sizes, I understood what they were saying but it didn't have that much impact. It wasn't until I took the time to actually replicate the tests they were doing that the full impact of what they were saying really hit me. I know it's a highly unpopular topic so I never talk about it here but it completely changed the way I look at things. Given the high variability of small sample sizes I'm not surprised by Varminterror's "test" results.

It's easy to get into apples vs oranges comparison. For "larger" groups where the shooter, equipment, conditions, etc account for 75% of the variability then it is difficult to assess the relatively minor impact of load optimization. On the other hand it is much more straightforward to detect differences in the benchrest world of consistent, small groups. It's the ability to distinguish signal vs noise, and the necessary sample size to detect the desired difference can be calculated for both of these worlds.

 

[TC260]

It's easy to get into apples vs oranges comparison. For "larger" groups where the shooter, equipment, conditions, etc account for 75% of the variability then it is difficult to assess the relatively minor impact of load optimization. On the other hand it is much more straightforward to detect differences in the benchrest world of consistent, small groups. It's the ability to distinguish signal vs noise, and the necessary sample size to detect the desired difference can be calculated for both of these worlds.

Rereading my post I don't see where I made any comparisons but point taken.

 

[BountyHunter]

And others apparently....
They probably didnt pick up on your "trick" because who would expect someone to do that? And why? Obviously your groups were inconsistent enough that guys thought they were different loads.

Oh snap!

 

[jelenko]

I know that is a good point of view, but in some ways a learning experience for everyone. I felt foolish thinking my 30-06 bullet was going faster at 1000 yards than a 6.5 Creedmoor. And it was but not for the reasons I thought. There is always some truth in feeling foolish. It's a good thing, because we learn.

Yeah, but not because someone is setting you up.