Sorry to say but your high school physics teacher stood in front of the class and told a bold face lie when they said, "For every action, there is an equal and opposite reaction." While this theory is true for extremely simple systems, it does not hold well for more complex systems like the human and the pistol in the shooting system.
I bring this up because of the number of posts where somebody states that they made this, that or the other change resulting in this, that or the other improvement or made their score worse. I would like to submit that the short term "improvements" might have occurred had no change been made.
Consider your typical match score. It is a combination of scores, rarely do you shoot the same score all match. And if you combine ten shoot groups, some are probably much better than others. And you shoot different total scores from match to match. simiply stated, there is some amount of normal variation in your scores.
My point is that you have to do some experiments with different techniques and equipment but short term improvements (or drops in socre) may not be a very good indicator of whether a given change really helps you or not.
Now there are all kinds of statistical charting techniques that I could describe to show determine if the change was "real" or significant.
But more important perhaps is to consider that if you make a change to a technique or equipment, you only really can say it is an improvement if you have sustained improvement. You have to give yourself a chance to really learn the new technique and/or get use to the new equipment.
The new what ever concept
Moderators: pilkguns, m1963, David Levene, Spencer, Richard H
Forum rules
If you wish to make a donation to this forum's operation , it would be greatly appreciated.
https://www.paypal.com/paypalme/targettalk?yours=true
If you wish to make a donation to this forum's operation , it would be greatly appreciated.
https://www.paypal.com/paypalme/targettalk?yours=true
-
Steve Swartz
OH NO-O-O-O-O!!!!
O.K., sounds like we're talking about Statistical Process Control, Process Variability Reduction, Design of Experiments . . . and all that other "Scientific Crap!"
No way, Pete- I'd rather just moly-coat my bullets and cryogenically treat my barrel thank you very much.
Who has time to take data and analyze it objectively?
Sheesh. The nreve of some people.
Steve Swartz
O.K., sounds like we're talking about Statistical Process Control, Process Variability Reduction, Design of Experiments . . . and all that other "Scientific Crap!"
No way, Pete- I'd rather just moly-coat my bullets and cryogenically treat my barrel thank you very much.
Who has time to take data and analyze it objectively?
Sheesh. The nreve of some people.
Steve Swartz
-
Fred Mannis
- Posts: 1298
- Joined: Sun Aug 29, 2004 8:37 pm
- Location: Delaware
-
Pär Hylander
- Posts: 35
- Joined: Tue Dec 28, 2004 5:16 pm
- Location: Sweden
Are we talking about Newtons classical physics or statitics? ;-)
With a scientific point of view you might be totally correct, but I think that a more pragmatic and usefull way is to just accept things that improve your performance without trying to analyse why in every small detail. Not seldom, the reason for improvement when changing equipment or technique is to be found between the ears of of the shooter.
With a scientific point of view you might be totally correct, but I think that a more pragmatic and usefull way is to just accept things that improve your performance without trying to analyse why in every small detail. Not seldom, the reason for improvement when changing equipment or technique is to be found between the ears of of the shooter.
What I see and what I wonder
Fred recently pointed out the concept of fun. That is why we do this IMHO. Some practices will yield good targets and strings of 5 or 6 deep 10's. On those occasions, what I have observed about my own condition is that I am not really thinking about anything...at all.....and the eyes trail to clarity on the front sight and the black hovers ever-so-steadily just above the blade and my finger naturally approaches my palm moving the trigger at the same time. As soon as I try to catalogue all those little elements in isolation.....a looser 9 sneaks in or, gasp, an 8. Other practice/match sessions, the start is sort of bleak and the progress comes after "settling in" for a couple dozen shots. And the session closes with some neat feats. What all this means to me outside of the statistical analysis demonstrates that the subconscious (vis a vis Lanny Basham) takes over, the "zone" dominates. If there are flaws in the system, i.e. deficient elemental techniques, consciously applying the reins and tweaking fundementals into line will yield better shots. The real trick to all this if one exists, centers on feeling center and staying there when you are there or moving toward it when you're not. The benefit of good practice is to facilitate either condition. When I am shooting well I am very good (but not nearly often enough yet) and when I am not shooting well......hey, this is still fun! CraigE Thanks for listening.
My backround is very much from the statistical prespective as Steve points but I did not go into a formal discussion of the statistical approaches because few would really care and it requires far more than these posts should involve.
I agree that the vast majority of us are really more interested in having fun, and experimenting with equipment is more fun than serious training and disciplined improvement in technique.
So most of us prefer to shoot ten shots with the rear sight narrow, then ten with the rear sight wide, then ten shots with center hold, then ten with sub-six, then ten with the guys pistol who just walked onto the range...
Any improvement is really just the chance occurance or at best the result of having concentrated more because that change required us to concentrate a little more temporarily.
With in a range of techniques, most differences are between the shooter's ears and what we are comfortable with. But my point remains, if you want to really improve, you have to experiment. But you must use a new technique or equipment change long enough to evaluate the change on a sustained basis. Otherwise, you will just be changing, not improving.
I agree that the vast majority of us are really more interested in having fun, and experimenting with equipment is more fun than serious training and disciplined improvement in technique.
So most of us prefer to shoot ten shots with the rear sight narrow, then ten with the rear sight wide, then ten shots with center hold, then ten with sub-six, then ten with the guys pistol who just walked onto the range...
Any improvement is really just the chance occurance or at best the result of having concentrated more because that change required us to concentrate a little more temporarily.
With in a range of techniques, most differences are between the shooter's ears and what we are comfortable with. But my point remains, if you want to really improve, you have to experiment. But you must use a new technique or equipment change long enough to evaluate the change on a sustained basis. Otherwise, you will just be changing, not improving.
-
Steve Swartz
The math-impaired should probably skip this post.
Pete, it's a classic DOE problem that can be represented by an n-dimensional response surface where
X1: finger position on trigger
X2: stance
X3: etc
.
.
.
Xn: the last important independent variable on the list
and
Y1: distance of shot center to center of target
Now, if we were to identify the important Xs (perhaps through a literature review and/or survey of world class coaches and shooters), and then simply collect data on all the variables while we vary the settings of the input variables, we would come up with the beta weights (slopes) of each variable (effect size) AND identify the technical "sweet spot" value for each variable.
Assuming equivariance and normality, the rule of thumb is we need about 10 data points (shots) for each treatment. Say 12 technique elements (stance, grip, etc.) and at least 3 levels or settings for each element, we would have a full factorial design of 36 total treatments and only need 360 shots total. Given that with RSM you don't need a full-factorial design we could probably get away with using Latin Squares or Taguchi; then again, the variance will be horrendous and the effect sizes small.
We could probably totally analyze our technique front to back find the optimal settings with 600 shots or so. In ten days we could have all the data we needed . . .
Next year I work on technique. This year was my physical conditioning year. I'll let you know next summer what I found out.
Steve Swartz
Pete, it's a classic DOE problem that can be represented by an n-dimensional response surface where
X1: finger position on trigger
X2: stance
X3: etc
.
.
.
Xn: the last important independent variable on the list
and
Y1: distance of shot center to center of target
Now, if we were to identify the important Xs (perhaps through a literature review and/or survey of world class coaches and shooters), and then simply collect data on all the variables while we vary the settings of the input variables, we would come up with the beta weights (slopes) of each variable (effect size) AND identify the technical "sweet spot" value for each variable.
Assuming equivariance and normality, the rule of thumb is we need about 10 data points (shots) for each treatment. Say 12 technique elements (stance, grip, etc.) and at least 3 levels or settings for each element, we would have a full factorial design of 36 total treatments and only need 360 shots total. Given that with RSM you don't need a full-factorial design we could probably get away with using Latin Squares or Taguchi; then again, the variance will be horrendous and the effect sizes small.
We could probably totally analyze our technique front to back find the optimal settings with 600 shots or so. In ten days we could have all the data we needed . . .
Next year I work on technique. This year was my physical conditioning year. I'll let you know next summer what I found out.
Steve Swartz
Steve, I would go with a Plackett-Burman Screening experiment if I were to get that formal. But that is definitely no fun at all.
I think we want to convey a simple message, experiments should have enough data to proof improvement, unless you just like messing around. And messing around helps drive a lot of equipment sales.
I think we want to convey a simple message, experiments should have enough data to proof improvement, unless you just like messing around. And messing around helps drive a lot of equipment sales.
-
Steve Swartz
Pete:
Yes, you are absolutely correct. However, my point is simply that (if you know what you are doing) you don't really need all that much "data" to run a complete, 100%, front to back analysis.
Pick something simple like "Which hold area works best for me." Spend a few training sessions getting comfortable switching back and forth between hold areas, so that you reduce the "learning effect" and other historical confounds. O.K., so now you're ready: How much data do you need?
Run a protocol where you shoot 3 x 10 shot strings. Hold all other technical elements constant as much as possible. Use 9 warm-up ("sighters") before each 30 shot data taking session. Run 3 x 30 shot sessions:
9 warmup (sub, center, 6 x 3)
10 sub, 10 center, 10 6
9 warmup (center, 6, sub x 3)
10 center, 10 6, 10 sub
9 warmup (6, sub, center x 3)
10 6, 10 sub, 10 center
O.K., now you have 30 shots each and have taken data on hold area and shot value for each combination. Run a simple difference of means test- o.k., get fancy and run an ANOVA- and voila, you have your answer. You may need to run the experiment twice.
But in three days you have figured out something that most people obssess over and flop around about for years.
Steve Swartz
Yes, you are absolutely correct. However, my point is simply that (if you know what you are doing) you don't really need all that much "data" to run a complete, 100%, front to back analysis.
Pick something simple like "Which hold area works best for me." Spend a few training sessions getting comfortable switching back and forth between hold areas, so that you reduce the "learning effect" and other historical confounds. O.K., so now you're ready: How much data do you need?
Run a protocol where you shoot 3 x 10 shot strings. Hold all other technical elements constant as much as possible. Use 9 warm-up ("sighters") before each 30 shot data taking session. Run 3 x 30 shot sessions:
9 warmup (sub, center, 6 x 3)
10 sub, 10 center, 10 6
9 warmup (center, 6, sub x 3)
10 center, 10 6, 10 sub
9 warmup (6, sub, center x 3)
10 6, 10 sub, 10 center
O.K., now you have 30 shots each and have taken data on hold area and shot value for each combination. Run a simple difference of means test- o.k., get fancy and run an ANOVA- and voila, you have your answer. You may need to run the experiment twice.
But in three days you have figured out something that most people obssess over and flop around about for years.
Steve Swartz
-
Fred Mannis
- Posts: 1298
- Joined: Sun Aug 29, 2004 8:37 pm
- Location: Delaware
-
Steve Swartz
Fred:
Well, actually, yes- I have run this experiment (and others like it for other elements of technique).
A major caveat is in order however- these results were for myself; and probably not generalizable. Not only that, the results will change over time and the experiment would need to be repeated occasionally as your skills develop.
Final caveat- the last time I ran the "aim-hold" test was a few years ago with a Free Pistol; so obviously I have not followed my own advice!
ON THE AVERAGE what I found was (rank from best to worst):
Max Size Avg Shot
Aim Area Hold Values
Center Hold: 2 1 2
6 Oclock: 1 3 3
Sub 6: 2 2 1
Center gave me the smallest average holding area, and was tied with sub six for the maximum deviation from center. Second best shot values (average distance from center). Six O'Clock was smallest maximum deviation during hold, and worst average hold and shot value. Sub 6 gave me o.k. holds, but shot values clearly better.
This makes sense, but would require a somewhat lengthy explanation . . .
YMMV
Steve Swartz
Well, actually, yes- I have run this experiment (and others like it for other elements of technique).
A major caveat is in order however- these results were for myself; and probably not generalizable. Not only that, the results will change over time and the experiment would need to be repeated occasionally as your skills develop.
Final caveat- the last time I ran the "aim-hold" test was a few years ago with a Free Pistol; so obviously I have not followed my own advice!
ON THE AVERAGE what I found was (rank from best to worst):
Max Size Avg Shot
Aim Area Hold Values
Center Hold: 2 1 2
6 Oclock: 1 3 3
Sub 6: 2 2 1
Center gave me the smallest average holding area, and was tied with sub six for the maximum deviation from center. Second best shot values (average distance from center). Six O'Clock was smallest maximum deviation during hold, and worst average hold and shot value. Sub 6 gave me o.k. holds, but shot values clearly better.
This makes sense, but would require a somewhat lengthy explanation . . .
YMMV
Steve Swartz