PDA

View Full Version : What else could the USPSA classification system be like?



Alpha Sierra
12-21-2019, 05:20 PM
GJM suggested in a different thread that a discussion of how the USPSA classification system could be revamped to better reflect match performance. I think that's a fair question, so here it is.

I'm not going to detail here how the USPSA classification system works. I assume most reading this subforum already know that. If a reader unfamiliar with it wants some background, all one would want to know is here: https://uspsa.org/pages/about/classification

As a way to kick off the conversation I'll bring up a sport that has some important parallels to practical pistol competition: sporting clays. The relevant similarity is that both sports have non-standardized courses of fire. Sporting clays target presentations, like those of USPSA (and IPSC and IDPA) are never the same from match to match. even at the same range/shooting ground.

US Sporting Clays (governed by the National Sporting Clays Association) has no classifiers. Your classification depends on how you place within your class at every match you go.

The basic system works on the premise that it takes a certain number of "punches" (aka points) to move up in class. As the classifications go higher, the number of punches needed to get into them increase.

Punches are awarded on a sliding scale based on how many people were in your classification at a particular match. The more people in your class at a particular match, the higher the number of punches for the class winner and the deeper the punches go within class.

Examples:

1 to 2 shooters in a classification at a particular match = no punches for anyone
3 - 9 shooters in class, 1 punch earned by top score in class and all those who tie it
10 - 14 shooters in a class, 2 punches for high score and ties, 1 punch for second highest score and ties
15 - 29 punches in class, 3 punches for high score and ties, 2 punches for second highest score and ties, 1 punch for third highest score and ties
etc.....


Punches earned in a calendar year carry over one more year. You lose punches earned more than one calendar year back from where you are.

Having climbed up that system from E (the bottom) to A (3rd highest classification) I think the system is pretty fair and does what it's supposed to do: group shooters of similar ability. It's not perfect, and there are variations on that theme, but I think a similar system would better reflect overall competitive ability in USPSA compared to what we have now.

ranger
12-21-2019, 05:28 PM
I used to be a very active Sporting Clays shooter. No classification system is perfect, but I saw a lot of sandbagging where competitors would show up a major event with a low classification then miraculously shoot an extremely high score to dominate their class. Mix in side money betting drove the sandbagging even more. "Money corrupts..."

Eyesquared
12-21-2019, 05:54 PM
I think it would be interesting if USPSA classifications weren't basically earned for life. What if you automatically dropped a class if you didn't shoot in a given division for over a year? Or if you could drop a class by shooting under the % for your class? I know the common argument against this is that sandbaggers will take advantage but as someone who mostly just shoots club matches I think worrying about prize tables determined by class placement might be overblown for the vast majority of shooters who don't travel to major matches.

Alpha Sierra
12-21-2019, 06:20 PM
I used to be a very active Sporting Clays shooter. No classification system is perfect, but I saw a lot of sandbagging where competitors would show up a major event with a low classification then miraculously shoot an extremely high score to dominate their class. Mix in side money betting drove the sandbagging even more. "Money corrupts..."

I should have said this from the get go: there is no system that can account for sandbagging. None.

Yes, there's sandbagging in NSCA. Just like there is in every shooting sport. So I humbly suggest that we don't even consider the issue because it can't be designed out of the system. Any system.

TheRoland
12-21-2019, 06:49 PM
I think I missed the original discussion, but in USPSA is there really a big disconnect between classification and match performance? I'm aware that there are "paper GMs", but is the issue larger than that? The goal of divisions seems like it should be keeping lower divisions fair, so if someone wants to shoot against other GMs, that's fine.

Alpha Sierra
12-21-2019, 07:53 PM
in USPSA is there really a big disconnect between classification and match performance?

The question, as I see it, is what should classification measure?

Should it measure your match performance against that of others within your division? Or should it measure your ability to perform on standard exercises?

Or something else?

YVK
12-21-2019, 10:50 PM
I think I missed the original discussion, but in USPSA is there really a big disconnect between classification and match performance?

No. The biggest disconnect in USPSA is that JJ, Ben, Max, Nils are GMs, as is a bunch of guys who can't even come close to 85% of those guys match in and match out. Sure, you get an occasional underclassified shooter ( a B class guy taking fourth at the CO Nationals) and you have a number of aging GMs that can't do what they could anymore. That said, with all its imperfections, the classification generally reflects match placements. I also think that it is a better way to measure things than match performance. Match placement depends on who shows up and what mood they are in. Classifiers are just so much more consistent and reproducible. If USPSA wants the classifiers results to correlate even better with field courses results, the simple answer is to introduce classifiers set in a field course format. It would be a bitch to set those up accurately but it is not impossible. I just don't think it is needed.

Eyesquared
12-21-2019, 11:22 PM
No. The biggest disconnect in USPSA is that JJ, Ben, Max, Nils are GMs, as is a bunch of guys who can't even come close to 85% of those guys match in and match out. Sure, you get an occasional underclassified shooter ( a B class guy taking fourth at the CO Nationals) and you have a number of aging GMs that can't do what they could anymore. That said, with all its imperfections, the classification generally reflects match placements. I also think that it is a better way to measure things than match performance. Match placement depends on who shows up and what mood they are in. Classifiers are just so much more consistent and reproducible. If USPSA wants the classifiers results to correlate even better with field courses results, the simple answer is to introduce classifiers set in a field course format. It would be a bitch to set those up accurately but it is not impossible. I just don't think it is needed.

You have much more USPSA experience than me, but I thought that the issue of people aging or having their skills deteriorate so that they can't shoot to their peak classification isn't solely exclusive to GMs.

Alpha Sierra
12-21-2019, 11:35 PM
with all its imperfections, the classification generally reflects match placements. I also think that it is a better way to measure things than match performance.

Is there any objective evidence that it does?

YVK
12-21-2019, 11:37 PM
That's correct. It is just more noticeable, or attracts attention, when people have higher ranks next to their names. B class guys don't really win matches so if they lost a step and are finishing inside of a C class folks, nobody's noticing.

YVK
12-21-2019, 11:51 PM
That said, with all its imperfections, the classification generally reflects match placements. I also think that it is a better way to measure things than match performance.



Is there any objective evidence that it does?

In regards to the first sentence or the second? If the former, then I think yes. I shoot CO now, we don't have GMs in it locally, only Ms, but it is M class guys who win matches, not me with my A. Incidentally, higher % Ms win, not barely there paper Ms. When I shot Production, it was again GM or high Ms who were winning.

Here is a link to the Limited Nats results. Choosing Limited simply because it is a long established and popular division. Pretty telling as far as what letters are mostly seen at the top vs middle vs bottom.


https://practiscore.com/results/new/93422?q_division=2

GJM
12-22-2019, 07:35 AM
While I generally agree with your belief that the classification system generally works, nationals and area matches are mostly attended by stronger shooters, so you have selection bias at play.

YVK
12-22-2019, 11:07 AM
, nationals and area matches are mostly attended by stronger shooters, so you have selection bias at play.


If you're including me in that group too, I am gonna get emotional.


Regardless, I am not worried much about that. I think that a 500 participants match with some selection bias is still a better data than 70 participants local match with a low sample size issues.

bofe954
12-22-2019, 11:16 AM
I always wonder how many sandbaggers are really just people on a fast trajectory upwards and classification system is just too slow. I go to a weekly match and we only run a classifier every other week. Local weekend matches don't generally run classifiers. I tend to place worse percentage wise at majors so if those are tacked in my percentage will go down. Plus it's just easy to botch one. In production one C can drop you from a GM to M score. Literally a half inch difference in one shot...

When you consider a new guy practicing hard could probably change skill level from C to M over the course of a year, I think it unlikely they'd shoot enough classifiers to see that change.

As far as upper level guys showing up lower in major matches, could easily be a nasty malfunction or just a mistake. They percentages get pretty tight at well attended matches, doesn't take that much to dump you down.

I've often wondered if the system shouldn't be fixed vs always changing. You sit down with a calculator and figure out hitfactors for GM's. For example you say a GM should have 0.9 draw to an A on an open target at 10 yards, an 0.18 split, a 0.30 transition, and a 1.0 reload. Then you just set a number. I think the numbers of GM's would increase and majors would mainly just be a bunch of GM's duking it out. I've won quite a few plaques for lower class wins and honestly I never get any satisfaction from it. Who gives a shit if you're top C when you get stomped by 40 other competitors anyway?

Alpha Sierra
12-22-2019, 11:21 AM
I've won quite a few plaques for lower class wins and honestly I never get any satisfaction from it. Who gives a shit if you're top C when you get stomped by 40 other competitors anyway?

I sorta agree. I focus only on my placement within my division. I don't care about the overall and I don't care about B class either.

bofe954
12-22-2019, 11:37 AM
I sorta agree. I focus only on my placement within my division. I don't care about the overall and I don't care about B class either.

Overall gets useful if you're shooting something oddball and there aren't many people in your division. You won limited 10 at your local but you were the highest class guy anyway, maybe you compare yourself to the SS guys who are major but probably have the same number of reloads or something. You shoot SS minor, maybe you compare yourself to the production guys. Or maybe there is just a super consistent guy and you keep an eye on the percentage delta.

Clusterfrack
12-22-2019, 12:18 PM
I agree with YVK and GJM that the current USPSA classification system mostly works. It's interesting that even though the classification game is quite different from the match game, classification more-or-less reflects match placement. There's a lot of overlap, but the ranks pretty much work. GMs at the top, next Ms, and so on.

I've usually been able to shoot my classification percentage at State, and Section matches, so that tells me it's working ok for me. When a national-class GM shows up at an Area match, that pushes everyone's % down but I don't get butthurt about it. I think it's awesome to be able to compete in the same matches with the top players in the sport. Should we have a Pro Class for them? Sure, why not?

I have come to like the classification game, and I like it even more now that only one re-shoot is allowed. I never liked the multiple re-shoot policy. It felt like a cheap way to move up. Now, it amounts to one run for match points, and one for classification. Two games in one.

I don't think match placement at any one match tells us that much. My goal is to continue to improve.

I do think USPSA classification is meaningful. It requires a lot of work to move up, and is as good or better a test of of shooting skill as any set short drills could be. The new classifiers are even better.

Sandbaggers and Grandbaggers? I don't really care about GBs, but sandbaggers piss me off because they take away from the enjoyment of the sport for B and A level shooters.

Alpha Sierra
12-22-2019, 02:10 PM
I don't think match placement at any one match tells us that much.

I agree. Which is why NSCA requires more than 2 - 3 punches to move up even from the lowest to the next highest class. And the number of punches (class win/place/show points) needed to move up increase as you go up the levels. This schedule requires some pretty consistent performance to move up.

AA Class to Master Class: 20 punches
A Class to AA Class: 14 punches
B Class to A Class: 12 punches
C Class to B Class: 8 punches
D Class to C Class: 6 punches
E Class to D Class: 4 punches

45dotACP
12-22-2019, 02:36 PM
I feel like the USPSA classification system works for everyone except the guys at the very top level.

Ben, JJ, Nils...those guys are GM class...but they're barely the same thing as your local GM.

Overall a B class guy will still be worse than A class, who won't be as good as M class etc...

Sent from my moto g(6) using Tapatalk

cheby
12-22-2019, 03:08 PM
I think grandbagging and sandbagging is not a problem. GrandbRagging on the other hand, is:)

Clusterfrack
12-22-2019, 03:13 PM
I think grandbagging and sandbagging is not a problem. GrandbRagging on the other hand, is:)

:D What's the equivalent for M class shooters? MasterBragging? Masterbating?

The GMs in our area are solid shooters who deserve their classification, and are good dudes. I haven't run into the bragging thing around here.

Jim Watson
12-22-2019, 11:45 PM
I feel like the USPSA classification system works for everyone except the guys at the very top level

So do we need a Great Grand Master class?

Jeff Cooper wrote that in the early days A was about a half dozen shooters who would dominate a match. B was another half dozen who would duke it out if none of the As showed up. C was the top half of the mass, D the bottom half. (No Masters then.).

45dotACP
12-22-2019, 11:53 PM
Probably not. It just needs to be understood that a GM card...much like a black belt in a martial art, means you still get to learn.

Not every BJJ black belt can be a world class competitor. You'll probably see an enormous skill difference actually. But that doesn't mean they aren't a black belt.

Sent from my moto g(6) using Tapatalk

Clusterfrack
12-23-2019, 12:07 AM
Probably not. It just needs to be understood that a GM card...much like a black belt in a martial art, means you still get to learn.

Not every BJJ black belt can be a world class competitor. You'll probably see an enormous skill difference actually. But that doesn't mean they aren't a black belt.


Well said. I emphatically agree.

Jim Watson
12-23-2019, 01:18 AM
Does BJJ not have the multiple degrees of black belt common to the Orientals?

Eyesquared
12-23-2019, 01:24 AM
I would think most if not all GMs should have a decent idea where they stack up in the grand scheme of things. On the other hand I think accurate classification is actually more important for lower classes where people lack the experience in the sport to know what level they're at. As a newer shooter myself I find it very annoying that I am comparing myself to people who made a classification years ago and haven't shot that division since, or I am comparing myself with tons of unclassified shooters.

It may also be important for instructors trying to market, using their classification, to outsiders who don't really understand the nuances of the sport. I see lots of people who don't understand how hard it is to make master or GM, how the relative skill levels stack up within each classification, that not all GMs are competitive with the top GMs, etc.

45dotACP
12-23-2019, 09:01 AM
Does BJJ not have the multiple degrees of black belt common to the Orientals?It does...but there are only a handful of 10th Dan BJJ practitioners. It doesn't mean you're the best in the world because by that time you're pretty damn old, and remaining competitive in a combat sport is a younger man's game.

Sent from my moto g(6) using Tapatalk

Dismas316
01-25-2020, 03:23 PM
The 4 new classifier that were recently introduced (19-01-04) are in the right direction imo. I hope that starts a trend for classifiers going forward.

olstyn
01-25-2020, 03:41 PM
The 4 new classifier that were recently introduced (19-01-04) are in the right direction imo. I hope that starts a trend for classifiers going forward.

Classifiers with movement are IMO a good thing, and more likely to be predictive of real match performance than the mainly stand and shoot ones the sport has relied on in the past. El Pres, while definitely a valid test of shooting skill, is pretty much nothing like you'll ever see in a "real" stage. It'll be interesting to see what more classifiers with movement in them do for different shooters' classifications. You could have two people at the same classification level who got there in two very different ways; maybe one is great at old-style stand and deliver style classifiers and the other doesn't shoot as fast in that scenario but has a higher fitness level and thus moves much better. Ultimately, making M/GM level is still going to require a high level of dedication to training/practice either way, though.

Gio
02-03-2020, 11:59 AM
GJM suggested in a different thread that a discussion of how the USPSA classification system could be revamped to better reflect match performance. I think that's a fair question, so here it is.

I'm not going to detail here how the USPSA classification system works. I assume most reading this subforum already know that. If a reader unfamiliar with it wants some background, all one would want to know is here: https://uspsa.org/pages/about/classification

As a way to kick off the conversation I'll bring up a sport that has some important parallels to practical pistol competition: sporting clays. The relevant similarity is that both sports have non-standardized courses of fire. Sporting clays target presentations, like those of USPSA (and IPSC and IDPA) are never the same from match to match. even at the same range/shooting ground.

US Sporting Clays (governed by the National Sporting Clays Association) has no classifiers. Your classification depends on how you place within your class at every match you go.

The basic system works on the premise that it takes a certain number of "punches" (aka points) to move up in class. As the classifications go higher, the number of punches needed to get into them increase.

Punches are awarded on a sliding scale based on how many people were in your classification at a particular match. The more people in your class at a particular match, the higher the number of punches for the class winner and the deeper the punches go within class.

Examples:

1 to 2 shooters in a classification at a particular match = no punches for anyone
3 - 9 shooters in class, 1 punch earned by top score in class and all those who tie it
10 - 14 shooters in a class, 2 punches for high score and ties, 1 punch for second highest score and ties
15 - 29 punches in class, 3 punches for high score and ties, 2 punches for second highest score and ties, 1 punch for third highest score and ties
etc.....


Punches earned in a calendar year carry over one more year. You lose punches earned more than one calendar year back from where you are.

Having climbed up that system from E (the bottom) to A (3rd highest classification) I think the system is pretty fair and does what it's supposed to do: group shooters of similar ability. It's not perfect, and there are variations on that theme, but I think a similar system would better reflect overall competitive ability in USPSA compared to what we have now.

I have very few complaints about the USPSA classification system. I don't want to see it be purely based on match performance, because I think one of the best aspects of it is being able to compare yourself to the top shooters without having to commit your life to traveling around to major matches all the time. Also in my experience, shooters who shoot the way the classification system is designed tend to have their classification match their actual match performance. The examples that don't hold true to that are shooters who intentionally sandbag, or shooters who pay to reshoot classifiers over and over until they get a decent score.

I think USPSA can take some steps to ensure the classification system is as robust as possible though and minimize the outliers by taking a few proactive steps:
1. Only allow single attempts on a classifier per match. Allowing reshoots at all is bullshit, regardless of the reason.
2. Bump sandbanging shooters immediately for high performance at a level 2+ match up to any classification except GM. As the system works right now, there are too many rules for match bumps, and only the most populated divisions tend to have enough shooters to consider match bumps or classifying scores. I would exclude the GM classification from this except at nationals. GM's should need to shoot classifiers at a GM level in order to make GM. IDPA classification is broken because they give out match bumps to master left and right, and you have the majority of master class IDPA shooters who can't shoot a master score on the classifier.
3. Constantly re-evaluate the high hit factor on classifiers. They should have an algorithm built to do this on a continual basis as scores come in, with a way to throw out extreme outliers. As it stands now, we are lucky to get a HHF update once every 10 years, which results in broken "easy" and "hard" classifiers allowing shooters to cherry pick which ones they shoot or do well on.
4. Separate PCC classification from pistol classifications. A PCC GM should not automatically be a M in other divisions. They are completely different skills and I see a lot of pistol M's at matches with a ~60% classification score in that pistol division because they made PCC GM.

I actually think IDPA has a better classifier stage with their longer course. It's long enough it can't be hero'd/zero'd and it has a good balance of close range and distance shooting. Of course they screwed it up/watered it down by adding the 5x5, which is no where near as challenging to make master, and they give you a match bump at a level 2 match with no competition.