Classifier HHF Determination

[kneelingatlas]

PS: please don't use big words like "statistical inference", they intimidate me

[motosapiens]

You give three (actually five) possible methodologies a few of which I could agree with, but how could you justify using one on some classifiers and another on others? Above you said there is no reason that it has to be the same process for every classifier, but how could that be?

imho, the only important thing is the result (that most people shoot approximately their classification when shooting at a normal match pace). The process to get there is not really important, at least to me. It sounds like that process *is* important to you, and that's ok.

[kneelingatlas]

I understand your position, but how then would you select the method by which you would evaluate the scores on any particular classifier? Just pick one randomly, then try a different one if you don't like the results?

Edited November 8, 2013 by kneelingatlas

[Chris iliff]

Now I can't find where to place the mark on the table for my trigger guard in CM 13-05. Hmmm. This is a classifier that just keeps on giving, lol. What a cluster jack.

If USPSA would institute a board that dealt with classifiers and stuff, say a couple Area Directors or something, then this stuff wouldn't happen. Hold on,.........they do have such a board made up of Area directors that deal with just classifiers.

OK, I guess I should say IF USPSA would have let the board that is chosen as the "classifier committee" to actually pick and implement the new classifiers, establish a protocol, organize the effort, then maybe this wouldn't have happened.

But alas, we have such a board, they just weren't ever told anything about, well, .......anything. EXCLUDED from the project comes to mind. Wonder how this could even happen? I wonder Why?

Hmmmmm, just rcv'd my mini classifier book from Chris Keen, PM him for yours, and went through all the new diagrams. Yikes, I found some things fairly confusing. Gonna start a new thread for all that. Maybe we can list the problems and get them fixed before too many get tossed out due to irregularities.

Edited November 9, 2013 by Chris iliff

[kozy]

Hey Guys, I was wondering the same thing on how the HHF is calculated. I emailed Phil a week ago and he responded right away with he wasnt sure but CC'd John Amidon and Val at USPSA to chim in. Crickets since. I remailed all of them again asking for some clarification, please.

I would ask that you all email John Amidon asking for clarification also...Maybe they can post a notice up...

[sperman]

Val actually talked about it in the latest issue of Front Sight. I'll try and remember to post what she said when I get home tonight.

[Chills1994]

How in the world did I miss this thread?

I've only been asking for more transparency in the classifier system since ...oh...about 2008???

Remember when the old USPSA website had a page for the top twenty in each class per division? What a waste of server space/bandwidth!

I said back then that particular page should be done away with and replaced with a records page showing all the info on the top five runs on each classifier stage, by division.

I specifically remember Zhunter starting a thread after single stack became a legit division. He was asking why the SS HHF's were the same as the L-10 HHF's. His pee pee might have gotten smacked for that thread. Or his use of the popcorn eating emoticon. I can't remember which now.

Now that I have got that "what am I, chopped liver?" feeling, I am going to go sulk.

[sperman]

This is what Val said in Front Sight:

How do we calculate the high hit factors? These are determined by the scores submitted over time for each classifier, and are in fact an average of the highest scores submitted. They are always calculated for each division, so the HHF for Open will not be the same as the HHF for Production. This means you are competing against the highest scores - against the GMs the top dogs, the "best of the best" - in your division.

[kneelingatlas]

That sounds rather arbitrary... I wonder if there's a methodology in place for choosing which scores to average?

"Ooh that one was fast! Let's throw it in the HHF hooper"

[Racer377]

Then why is the HHF for "Can you count" the same HF Max Michel laid down? Not an average, and they used the same HHF for all divisions other than revolver?

and why are all the HHF in SSTK for almost all the 99 series EXACTLY 5% higher than those in PRD?

The explanation, particularly with the "They are always calculated for each division" claim, simply doesn't square with the numbers.

[Chills1994]

A few years back...,

Whoops, let me rewind a bit...

Somebody had sent me an email with an attachment, an excel spreadsheet with all the HHF's for production. In the main part of the email, this guy said something like "you should check out shred's thread on the classifier scores". It's still here somewhere on BE, complete with pretty graphs that shred put together.

At about the same time somebody had posted or sent me another email saying that the HHF's for open were set at let's say X. He went on to say that Limited was stepped down 10% from that, then production was stepped down another 10% from that. Sure enough, I checked out and did the math on about 5 to 10 classifiers, and that's how the HHF's broke down.

[Chills1994]

Then why is the HHF for "Can you count" the same HF Max Michel laid down? Not an average, and they used the same HHF for all divisions other than revolver?

and why are all the HHF in SSTK for almost all the 99 series EXACTLY 5% higher than those in PRD?

The explanation, particularly with the "They are always calculated for each division" claim, simply doesn't square with the numbers.

Depends on what your definition of "is" is.

Did someone take the HHF of CM whatever in open division and multiply that by 90% to calculate the HHF for Limited?

Then that statement is technically correct/true.

Has anybody actually shot classifier XYZ in real life, at such and such a hit factor that has proven to be the highest hit factor submitted to HQ, for that stage?

I don't know.

That is why I have been saying there needs to be a records page posted on the USPSA website.

Slight thread drift... I don't know about how things were at your high school when it came to athletics or sports. My school's gymn had several 4X8 sheets of plywood hung up about 12 feet off the floor. There were lines there for each track event:

100 meter dash..........john doe 11.2 seconds 1989

110 meter high hurdles...Jim Smith 13.8 seconds 1985

200 meter dash........Justin Case 24.3 seconds 1986

And so on and so on...

You get the idea.

[Nik Habicht]

This is what Val said in Front Sight:

How do we calculate the high hit factors? These are determined by the scores submitted over time for each classifier, and are in fact an average of the highest scores submitted. They are always calculated for each division, so the HHF for Open will not be the same as the HHF for Production. This means you are competing against the highest scores - against the GMs the top dogs, the "best of the best" - in your division.

I'm calling BS on that explanation.....

That's been the party line for a long time -- and I'd be much more likely to believe it, if so many of the recent HHFs hadn't been equivalent to national's or major match stage winner's HHFs....

[whitedog]

And again it comes back to this....why is the HHF a secret anyway? What's the point? To stop someone from grand bagging? Hell, that already happens anyway.

I'm concerned the process is not transparent, coupled with the secrecy, it just seems things are being manipulated or something.

If the previous HHF math posts are correct, the whole system is a sham. No credibility in the explanation or the process.

Just my view

[Chills1994]

It's been joked about here on this forum how a new CM _____ series will come out and POOF! All of a sudden like a month or two later there will be a much longer list of newly minted Master classed shooters published in Front Sight magazine.

I have always been of the mindset that the classifier system was and always has been a revenue generating stream, first and foremost.

I could not tell you how many temporary bans/time outs from this forum I have been subjected to because I brought up money, prize tables, or the classifier system.

Does USPSA HQ want to make money? Yeah, sure. I can't say that I blame them. You'll notice one of the first things USPSA HQ did once it bought the Steel Challenge was to give it a classification system. ( You'll also notice that 3 Gun Nation also added classifiers, and I am assuming a classification system. So it can be theorized that somebody there thinks that particular business model works.)

I'd be all in favor of just paying a $3 "activity fee" and scrapping the classifier system entirely. Run whatchya brung head's up within a division.

Edited November 16, 2013 by Chills1994

[whitedog]

I like the theory of a "rank" system, as it gives me some arbitrary gauge of my own progress. I believe many people do. That said....I stand by my view that the system is so convoluted and secret I believe the percentages have no bearing whatsoever. It's just a gauge.

[Chills1994]

go to a major match, heck go to several major matches, where you place in your division, that's your "rank"

[JakeMartens]

Really how many if the top gm's don't shoot their "rank" at nationals

[whitedog]

Chills....that's what I do. Have only shot 3 majors and did about the same in all 3. Amazingly mediocre. ?

[Chris iliff]

Really how many if the top gm's don't shoot their "rank" at nationals

Yeah, Jake is making a valid point, that argument is and always will be a wash with me. Not one person shot into the GM range at Production Nationals..........well except of course, Eric.

[Chills1994]

Looking at results from Nat's represents a confounding issue in terms of sampling. It would be like going to a gunshow and taking a survey of attendees on their thoughts on the Second Amendment.

And just in general that argument is not relevant to this thread, which is how the HHF's are derived.

It would be nice if the classifier CoF description had a box for text that said something like "Max Michel, A34XXX, set the HHF as a 9.8 when he shot it at the 2012 Limited Nationals,"

[JakeMartens]

Looking at results from Nat's represents a confounding issue in terms of sampling. It would be like going to a gunshow and taking a survey of attendees on their thoughts on the Second Amendment.

And just in general that argument is not relevant to this thread, which is how the HHF's are derived.

It would be nice if the classifier CoF description had a box for text that said something like "Max Michel, A34XXX, set the HHF as a 9.8 when he shot it at the 2012 Limited Nationals,"

According to USPSA HQ it is an average so a box like that would not be able to be done

"How do we calculate the high hit factors? These are determined by the scores submitted over time for each classifier, and are in fact an average of the highest scores submitted."

[motosapiens]

go to a major match, heck go to several major matches, where you place in your division, that's your "rank"

and for me and most of the folks I've traveled with, that rank has been very close to our current classification. That tells me that regardless of how the HHF is derived, it's at least approximately accurate.

Sure, there are some people that are already shooting at the next higher classification, but haven't had enough classifiers to get bumped. Sure, there are people who are no longer training and practicing as seriously as the year they got their A or M or GM card. But for most active competitors, the system seems to work fine.

[kozy]

Looking at results from Nat's represents a confounding issue in terms of sampling. It would be like going to a gunshow and taking a survey of attendees on their thoughts on the Second Amendment.

And just in general that argument is not relevant to this thread, which is how the HHF's are derived.

It would be nice if the classifier CoF description had a box for text that said something like "Max Michel, A34XXX, set the HHF as a 9.8 when he shot it at the 2012 Limited Nationals,"

According to USPSA HQ it is an average so a box like that would not be able to be done

"How do we calculate the high hit factors? These are determined by the scores submitted over time for each classifier, and are in fact an average of the highest scores submitted."

Jake, has HQ confirmed that they take a average? On the new classifiers, at least with 13-04 & 13-07, they are taking the highest in open div.

[JakeMartens]

That is from Front Sight, the article that Val wrote.