Jump to content
Brian Enos's Forums... Maku mozo!

Changes To The USPSA Classfication System


Maximis228

Recommended Posts

8 minutes ago, Asianjedi said:

So I was bored and went through the Top 20 list for GMs, Ms, and As in Carry Optics for El Prez 99-11.

Here is what I came up with:
10.2594

9.0176

8.5131

8.4126

8.4108

8.3731

8.2467

8.0023

7.8516

7.9895

 

The average is 8.5076. The current HHF is 10.75.
So, in theory, does this mean the HHF for El Prez could go down significantly based on this rationale? 
 

 

Not enough data in CO to warrant changes yet (I would imagine). Not many top shooters have made the switch outside of a nationals match.

Link to comment
Share on other sites

  • Replies 74
  • Created
  • Last Reply

Top Posters In This Topic

So here is what I see unless I'm just ignorant. It is common practice for A and above shooters to take the hero-zero approach to classifiers. In fact they have to do that if they wish to advance their classification. 

 

So along comes Joe, A Shooter, and he shoots 95% on a certain classifier because he had a lucky hero/zero performance. He had shot the same classifier 9 times before and obtained a variety of scores ranging from 40 to 85% and averaging 75%. 

 

As I understand the historical and current approach to setting the HHF for the classifier, the 95% number may well be used to determine the classifier HHF but the 9 "failures" are not even part of the analysis. This would appear to set an arbitrarily high bar because it used a pure 1 in 10 lucky result to set the standard.

 

Why wouldn't it make a lot more sense to use all the scores for a given classifier to determine a statistical distribution of those scores and then divide that distribution into sections to set a given percentage of shooters into each classification? For example  only, the top 0.5 percent of the distribution would define GM shooters and so on. Further as an example, the lower 60% (or so) of scores would encompass B,C, and D shooters with those further divided as appropriate.

 

It doesn't make statistical sense to me to use "lucky" scores to set the bar and ignore statistical results.   

Edited by Brooke
Link to comment
Share on other sites

1 hour ago, Brooke said:

So here is what I see unless I'm just ignorant. It is common practice for A and above shooters to take the hero-zero approach to classifiers. In fact they have to do that if they wish to advance their classification.

 

No, they could just practice and actually get better.  Some people do take that route.

 

I do agree that taking the highest-ever-shoot scores and using them is probably not the best idea, though.

 

That being said, with a couple of glaring exceptions, most of the classifier HHFs for the long-existing divisions (NOT CO and PCC) seem to be working just fine, judging by the distribution of match finishes (by classification) that occur at various Level III matches. 

Link to comment
Share on other sites

Why are most classifiers shot standing still?  I know a lot of people that shoot great stationary, but when you add movement, they don't do so good.  They have bad knees and can't move, or are just getting old.  They are classified higher then match results show.  I think is a better representation when they use your average for a major as a classifier.

Link to comment
Share on other sites

A significant number of USPSA members never shoot at Major competitions. Which of course begs the follow-up question; If they never shoot at a major match, do they still need a classification?

 

Classifiers have to be set up exactly as per the stage diagram, so the more shooting positions, the longer it takes to set them up correctly. 

Link to comment
Share on other sites

23 hours ago, Asianjedi said:

So I was bored and went through the Top 20 list for GMs, Ms, and As in Carry Optics for El Prez 99-11.

Here is what I came up with:
10.2594

9.0176

8.5131

8.4126

8.4108

8.3731

8.2467

8.0023

7.8516

7.9895

 

The average is 8.5076. The current HHF is 10.75.
So, in theory, does this mean the HHF for El Prez could go down significantly based on this rationale? 
 

 

That's just thee scores those guys happened to shoot on that classifier, not the top 10 scores ever shot on it ever.

I have a feeling the HHFs are going to go up, except for on the impossible fixed time 40 yard classifiers which I bet no one has ever shot a 100% on.

Link to comment
Share on other sites

  • 4 weeks later...
On Thu Jul 20 2017 at 6:47 AM, Thomas H said:

 

No, they could just practice and actually get better.  Some people do take that route.

 

I do agree that taking the highest-ever-shoot scores and using them is probably not the best idea, though.

 

That being said, with a couple of glaring exceptions, most of the classifier HHFs for the long-existing divisions (NOT CO and PCC) seem to be working just fine, judging by the distribution of match finishes (by classification) that occur at various Level III matches. 

Your idea of proof that the classification as it is currently, works as noted by how major match results shake out is called selection bias.

 

Said another way....how many paper GMs and M's are going to sign up for a major match knowing full well they are going to get their asses handed to them by A and B classed shooters?

 

The paper GM's and M's self-select NOT to attend major matches.

 

I know or knew of a few paper M's whose sole purpose to get their M card was so that they could pad their resumes as an instructor and teach pistol and CCW classes.

 

Why piss away a weekend at a major match and show the world your paper M status when you could be getting paid like $3,000 or more to teach a CCW class that same weekend (to the rubes who don't know any better)?

 

We could just get rid of the classification system entirely and shoot head's up within a division.

 

;-P

Link to comment
Share on other sites

35 minutes ago, Chills1994 said:

We could just get rid of the classification system entirely and shoot head's up within a division.

 

Which is what they generally do in IPSC, I understand that their classification system (ICS) is not widely used. If USPSA were to remove the classification system then there may be a financial hit.

Link to comment
Share on other sites

1 hour ago, Chills1994 said:

[snip]

 

The paper GM's and M's self-select NOT to attend major matches.

 

I know or knew of a few paper M's whose sole purpose to get their M card was so that they could pad their resumes as an instructor and teach pistol and CCW classes.

 

 

So....your argument that the classification system is broken is based on what, then?  I mean, if the data we have for matches shows that in the main the classification system works, what exact data do you have supporting the opposite idea?

Particularly since "selection bias" doesn't really fit----since the point of the classification system is so that people have other folks with similar skill to shoot against, and major matches is where there are enough shooters for that to become important....so what other data would you like to use?  Other than major matches, where relative classification IS rather the point?

 

In addition---your argument about the classification system being broken is based on....the idea that people who classify wrongly don't shoot matches? 

 

Huh.  That's a new one.

Link to comment
Share on other sites

49 minutes ago, BritinUSA said:

 

Which is what they generally do in IPSC, I understand that their classification system (ICS) is not widely used. If USPSA were to remove the classification system then there may be a financial hit.

It's still like a buck fifty for the activity fee and then like another buck fifty to get the classifier score submitted, right?

 

Just scrap the classifier system and then make the activity fee $3 .

 

USPSA HQ still gets their "per capita" dues paid in from each of the "franchises" to the same total amount.

 

 

Link to comment
Share on other sites

15 minutes ago, Thomas H said:

 

So....your argument that the classification system is broken is based on what, then?  I mean, if the data we have for matches shows that in the main the classification system works, what exact data do you have supporting the opposite idea?

Particularly since "selection bias" doesn't really fit----since the point of the classification system is so that people have other folks with similar skill to shoot against, and major matches is where there are enough shooters for that to become important....so what other data would you like to use?  Other than major matches, where relative classification IS rather the point?

 

In addition---your argument about the classification system being broken is based on....the idea that people who classify wrongly don't shoot matches? 

 

Huh.  That's a new one.

Here, read this:

 

https://en.m.wikipedia.org/wiki/Sampling_bias

 

Since you used the key statistical buzzword of "distribution", you can only infer things about a total population when the sample is chosen at random.

 

The people who shoot major matches aren't random and hence subject to the aforementioned sampling bias .

 

 

Link to comment
Share on other sites

47 minutes ago, Chills1994 said:

It's still like a buck fifty for the activity fee and then like another buck fifty to get the classifier score submitted, right?

 

Just scrap the classifier system and then make the activity fee $3 .

 

USPSA HQ still gets their "per capita" dues paid in from each of the "franchises" to the same total amount.

 

 

I believe the point is that for the 'masses,' they only reason they're paying members is to keep track of their classification scores/rank. If you do away with classifiers, many wouldn't bother paying to be a USPSA member. Could be big bucks. 

Link to comment
Share on other sites

1 minute ago, TrackCage said:

I believe the point is that for the 'masses,' they only reason they're paying members is to keep track of their classification scores/rank. If you do away with classifiers, many wouldn't bother paying to be a USPSA member. Could be big bucks. 

Can't people track their progress via the various classifier  calculator websites/apps?

 

And/or the other kinda accepted drills?

 

free of charge???

Link to comment
Share on other sites

Hey I was all about making A class. Now I'm looking forward to having time to devote to "making M" next year when I'm shooting again.

 

Classifications fuel the competitive nature in us to keep showing up and improving.

 

"my calculator says..." isn't as satisfying to thecaveman inside.

Link to comment
Share on other sites

I think it is so funny that people always say "well most Ms and GMs are just shooting hero or zero on classifiers, and they just got lucky enough to string 6 of them together" or the even more ridiculous "paper Ms and GMs don't shoot majors cause they know they will get stomped, so it is biased." I am definitely not the best GM I know, and I shoot 90+ on most any classifier out there, barring some calamity (bad draw, blown reload, etc). I would actually argue the opposite, that most Ms and GMs can pretty routinely shoot very high scores on classifiers. 

 

That is beyond silly. There are 'grandbaggers' and 'sandbaggers' at every classification level (C level shooters in B, B shooters in M, Ms who should be Gs, etc), and yet when you go to majors... the scores shake out almost exactly according to everyone's class with few exceptions. Not to mention, the classification system gives people a good motivator to chase that next letter like MM said. I do think that a huge number of local-match-only shooters would drop their USPSA memberships and quit paying their money if there was no classification system.

Link to comment
Share on other sites

Just my luck that they are upping HF's as I attempt to get an initial classification! 

 

Guess I'm delegated to Charlie class until I can put in a few hundred hours of dry fire! 

 

 

 

 

Edited by Covfefe
Link to comment
Share on other sites

3 hours ago, Gooldylocks said:

I am definitely not the best GM I know, and I shoot 90+ on most any classifier out there, barring some calamity (bad draw, blown reload, etc). I would actually argue the opposite, that most Ms and GMs can pretty routinely shoot very high scores on classifiers. 

 

you're not the best GM i know either, but you're a damn good consistent shooter that practices and takes the sport seriously..... big surprise, your scores reflect that.

 

I join you in taking exception to the thought that you have to go hero/zero to move up. The better method is practice the necessary skills to shoot more consistently at a high level. those skills will also help in field stages.

Link to comment
Share on other sites

15 hours ago, Chills1994 said:

Here, read this:

 

https://en.m.wikipedia.org/wiki/Sampling_bias

 

Since you used the key statistical buzzword of "distribution", you can only infer things about a total population when the sample is chosen at random.

 

The people who shoot major matches aren't random and hence subject to the aforementioned sampling bias .

 

 

 

I'm award of what sampling bias is.  I'm also aware that the "total population" considered depends on what question you are trying to answer.    (The total population of any statistical analysis is not "automatically everyone."  As such, taking a random sample of the total population is not automatically "a random sample of everyone.")

 

If you are attempting to make an argument that classifications don't work in terms of people being compared to each other in a match format, major matches are the place in which you will be able to tell that sort of thing.  Smaller matches (in general) don't have sufficient comparisons for statistical analysis.  (In addition problems due to statistical outliers which occur in most real distributions will also skew the data and make the results invalid.)  Not to mention that smaller numbers of stages make single stage errors major points of failure for the match, which also makes the data less reliable.

 

In this situation, we actually have the opposite of sampling bias---since our data is literally everyone in major matches, our sample (regarding whether or not the classification system works in major matches) is pretty much close to 100% of the total population for this question.

 

If, in the only venue in which relative classifications are important, the classification system is obviously working, then....what exactly are you trying to argue?

I note:  attempted condescension is actually merely amusing when you don't seem to understand the question at hand.  Particularly when 1) your argument against it is "people like that don't shoot major matches" (which is the situation where this might actually be important), and 2) you can't seem to provide any data at all to support your contention.

 

Also, saying that USPSA should scrap the classification system, but that people could still use classifier calculators to track themselves is hilarious.

 

Gooldylocks said it correctly:  " There are 'grandbaggers' and 'sandbaggers' at every classification level (C level shooters in B, B shooters in M, Ms who should be Gs, etc), and yet when you go to majors... the scores shake out almost exactly according to everyone's class with few exceptions."

 

As such, saying that the classification system is broken when it does precisely what it is supposed to do at major matches is.....odd.

Link to comment
Share on other sites

1 hour ago, Thomas H said:

 

I'm award of what sampling bias is.  I'm also aware that the "total population" considered depends on what question you are trying to answer.    (The total population of any statistical analysis is not "automatically everyone."  As such, taking a random sample of the total population is not automatically "a random sample of everyone.")

 

If you are attempting to make an argument that classifications don't work in terms of people being compared to each other in a match format, major matches are the place in which you will be able to tell that sort of thing.  Smaller matches (in general) don't have sufficient comparisons for statistical analysis.  (In addition problems due to statistical outliers which occur in most real distributions will also skew the data and make the results invalid.)  Not to mention that smaller numbers of stages make single stage errors major points of failure for the match, which also makes the data less reliable.

 

In this situation, we actually have the opposite of sampling bias---since our data is literally everyone in major matches, our sample (regarding whether or not the classification system works in major matches) is pretty much close to 100% of the total population for this question.

 

If, in the only venue in which relative classifications are important, the classification system is obviously working, then....what exactly are you trying to argue?

I note:  attempted condescension is actually merely amusing when you don't seem to understand the question at hand.  Particularly when 1) your argument against it is "people like that don't shoot major matches" (which is the situation where this might actually be important), and 2) you can't seem to provide any data at all to support your contention.

 

Also, saying that USPSA should scrap the classification system, but that people could still use classifier calculators to track themselves is hilarious.

 

Gooldylocks said it correctly:  " There are 'grandbaggers' and 'sandbaggers' at every classification level (C level shooters in B, B shooters in M, Ms who should be Gs, etc), and yet when you go to majors... the scores shake out almost exactly according to everyone's class with few exceptions."

 

As such, saying that the classification system is broken when it does precisely what it is supposed to do at major matches is.....odd.

 

Ouch, reading this took me back to painful memories of Statistics 101 :wacko:. Think I'll go shooting now.

 

Edited by Bkreutz
Link to comment
Share on other sites

Did I say that the classification system is "broken"?

 

Those were your words that you are trying to put in my mouth, Slick.

 

And you still have no clue what sampling bias is.

 

What percent of production division shooters  all across USPSA shooters are GM, M, A, B, C, and D?

 

Let's say, hypothetically, they shook out like this across all of USPSA (in production division discounting all the unclassified shooters EDIT because they may be brand new shooters or more than likely are already classified in another division and it makes it tougher to compare...how do define unclassified in Production?  That could be a ton of people....no ceiling effect there):

 

GM:  7%

M:     8%

A;      22%

B:      33%

C:      20%

D:      10%

 

Do you get that same distribution of classes at Production Nat's?

 

If the answer is "No", then basing anything off of or coming to some conclusion from major match results is wrong.  Major match results are not a true (random) sample of the whole population of (classified) USPSA members.

 

 

 

 

Edited by Chills1994
Link to comment
Share on other sites

For example, at the 2016 Production Nat's, there were 369   Production shooters.  26 of them were unclassified.

 

So that drops the number of production classified shooters down to 343.

 

The break down by class is as follows:

 

GM:  41..... 11.9%

M:     63......18.3%

A:     72.......20.99%

B:   101.......29.44%

C.     54.......15.7%

D.     12.........3.5%

 

If across all of  USPSA, the classifications of Production division shooters do NOT look anything like that distribution , then the Production Nat's results are not a representative sample and you should absolutely NOT make any inferences or make any conclusions about the results.

 

 

 

 

 

Link to comment
Share on other sites

On 8/11/2017 at 6:13 PM, Chills1994 said:

It's still like a buck fifty for the activity fee and then like another buck fifty to get the classifier score submitted, right?

 

Just scrap the classifier system and then make the activity fee $3 .

 

USPSA HQ still gets their "per capita" dues paid in from each of the "franchises" to the same total amount.

 

 

I don't feel that USPSA offers enough value without the classification system for me to want to continue running "USPSA" matches if I had to send them $3.00 per shooter.

Link to comment
Share on other sites

11 minutes ago, davidb72 said:

I don't feel that USPSA offers enough value without the classification system for me to want to continue running "USPSA" matches if I had to send them $3.00 per shooter.

 

I wouldn't worry, the classification system isn't going anywhere and this is just another "forum conversation" that has no meaning (except to a few:blink:)

Edited by Bkreutz
Link to comment
Share on other sites

18 hours ago, Chills1994 said:

Major match results are not a true (random) sample of the whole population of (classified) USPSA members.

 

 

No, they aren't.  I agree.

 

However, since that isn't the question at hand, it does show rather clearly that you don't seem to understand the problem, and also makes clear why you keep saying "sampling bias" when it doesn't apply.  It also shows that you ignored what I said about how the total population for this question is not the total population of everyone who shoots USPSA.

 

Since you don't seem to actually understand the statistical analysis you are attempting to discuss, we can stop here.  Good day.

 

As a last comment regarding the topic of the thread:  " As such, saying that the classification system is broken when it does precisely what it is supposed to do at major matches is.....odd. "

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...