A testament to the fallibility of the USPSA classifier system.

[DonovanM]

In the past year my classification score has dropped by almost 4%. But yet, my actual match scores have increased significantly, on the order of 10-20% or something.

I've shot enough GM classifiers, that if the timing were different and they were grouped closer together I would have made Grandmaster last year. But yet here I am - a year later - without a single GM score on record, or even more than one in the last 10(!) months. I would be humiliated by that if I had made GM!

Just goes to show that the skills required to succeed at short classifier stages are fairly distinct from the skills required to succeed at matches and/or long field courses. That's all. And it goes without saying that my training focus shifted almost completely at one point from classifier skills to match skills.

And now for some opinions: There's nothing wrong with a shooter who is better at classifier stages than they are at long field courses, and can't, let's say, shoot 95% or better at Nationals. Because frankly that's all but 3 or 4 GMs in a single division.

If anything, I think this is a more "practical" and/or realistic training focus. A soldier or police officer who made GM in any capacity is literally the last single person on the planet I would ever want to get in a gunfight with.

But... but...

Also, an addendum: I think that under no circumstance should a classifier be allowed to be reshot that could not be applied for any other stage. GM can mean different things, but that just ruins the entire thing to some degree.

Edited April 15, 2013 by DonovanM

[TediousAndIneffectualMcGee]

Interesting. How closely does your major match performance mirror your performance on classifiers?

[DonovanM]

It took me 8 minutes to get trolled. You guys need to try harder.

[TediousAndIneffectualMcGee]

Ummmmm. No. I was legitimatley asking as I know some people's results vary from classifiers to Local matches to major matches.

Usually it's: Classifier>Local>Major but I've noticed that there are exceptions.

[Poppa Bear]

I agree that the skills needed to stand and shoot are very different from those needed to run a stage. Movement between arrays, around walls and through ports creates challenges that are hard to duplicate in a classifier. I can practice my draw, and practice my presentation to the first target. I can practice transitioning 24" from A zone to A zone to lock it into muscle memory. But i cannot practice transitioning from target to target when the stage might be set up close to what it was last week but it is still different spacing, or the port is changed into a door.

[midget]

Have you shot a major yet with enough GM's to get it listed as a classifier match? If so, how did you compare them them?

[DonovanM]

Have you shot a major yet with enough GM's to get it listed as a classifier match? If so, how did you compare them them?

No, which is why I gave such a huge range of values and said or something. I've been shooting with almost the same exact people, at the exact same clubs with fairly similar stages for a couple years now so it's pretty easy to get a rough estimate of relative performance gains, and I think it's actually a better estimate than could be gained from a comparison between 2 or 3 majors simply due to the greater number of variables (like the pressure and experience dealing with the pressure) and the smaller sample size.

Anyone I shoot with who has tracked my progress over the past year or so will tell you that a 10-20% relative performance gain is not an unfair estimation. I am not correlating this increase with any assumption of major match performance. I am shooting Area 1 in a month so we'll have to see then.

Edited April 15, 2013 by DonovanM

[latech15]

There is a local club here in Louisiana that advertises that they will allow you to reshoot the classifieds as many times as you want to. They (supposedly) take the first one for your match score and they will only send in your best one. Needless to say, they have a lot of paper tigers over there.

I disagree with the concept of shooting classifiers as many times as you want to, but it is specifically allowed in the classifier rule book.

[DonovanM]

it is specifically allowed in the classifier rule book.

Yeah, I really think that should be changed.

[Nimitz]

Our local club has an annual classifier match every Jul (5-6 stages) plus 2-3 more classifiers thrown in throughout the year for a total of 9-10 every year. This way there is ample opportunity for anyone interested in classifcation to do whatever is necessary. Might consider suggesting to your MD to hold an annual classifier match, it's great way to get a lot of competitiors to your local match ....and compete against folks you normally don't ...

[lugnut]

There can be no perfect system to measure one's performance at matches and/or classifiers. All matches are different as are classifiers. Some matches have lots of tough targets, movers, steel, etc... some don't. I don't see it as an issue. In general I do believe the system is fairly accurate. In fact, most of the time when a B shooter is competitive with an A or an A is competitive with a M is because they go for broke on the classifiers they shoot. If a top shooter can't compete with lower shooters- well maybe they shoot lots of classifiers over an over.

Major matches don't compare well with classifications IMO. A top GM will shoot classifiers well over 100% and push everyone's match % down as well.

[motosapiens]

I disagree with the concept of shooting classifiers as many times as you want to, but it is specifically allowed in the classifier rule book.

Why do you care? Does it hurt your feelings if some of the people you beat in matches are classified higher than you?

[DonovanM]

I disagree with the concept of shooting classifiers as many times as you want to, but it is specifically allowed in the classifier rule book.

Why do you care? Does it hurt your feelings if some of the people you beat in matches are classified higher than you?

WTF? Talk about a non sequitur.

[beltjones]

Our local club has an annual classifier match every Jul (5-6 stages) plus 2-3 more classifiers thrown in throughout the year for a total of 9-10 every year. This way there is ample opportunity for anyone interested in classifcation to do whatever is necessary. Might consider suggesting to your MD to hold an annual classifier match, it's great way to get a lot of competitiors to your local match ....and compete against folks you normally don't ...

Donovan shoots 2 - 3 classifier matches per year (with about 6 classifiers per classifier match), so it's not that.

[motosapiens]

I disagree with the concept of shooting classifiers as many times as you want to, but it is specifically allowed in the classifier rule book.

Why do you care? Does it hurt your feelings if some of the people you beat in matches are classified higher than you?

WTF? Talk about a non sequitur.

What part was hard to understand? I can explain further. Someone disagreed with the concept of re-shooting classifiers. The only effect of re-shooting (I can think of) is that a few people may classify higher than they would have otherwise, even tho their actual shooting abilities are unchanged. Why should someone care about that?

[Hi-Power Jack]

I wonder if using Area & the Nat'l Match is a better way of classifying shooters?

Most good shooters shoot at least 2 of these/year - if I shoot Area 6 & the Nat'ls

at 78%, shouldn't I be an A? Regardless of what my classifiers say?

Just a thought ...

[Thomas H]

Major matches don't compare well with classifications IMO. A top GM will shoot classifiers well over 100% and push everyone's match % down as well.

Yes, but...

...if you compare match finishes, the classification system works remarkably well. It is only if you expect classification percentages to equal match percentages that you have a problem.

Way back when, I did an analysis of a Nationals, and compared scores from different classes within the Production division---and not only were the classes clearly separated within the match, but there was an obvious and clear relation between times and scores for each class.

That means that the classification system is pretty darn accurate, in terms of giving relative skill.

This is not to say in any way that the classification percentages actually work that way, however. The same analysis also showed that most people did NOT necessarily shoot within their classification percentage at the match (amusing to see how many GMs did not shoot GM percentages at Nationals) ---but then again, I wouldn't expect them to do so.

Classification tests certain fundamental shooting skills. Matches test varying sets of those skills plus a number of other things. As such, expecting the numbers to match seems odd.

In response to Donovan's original post: So, you have solid fundamental shooting skills that you'd been practicing, and thus was classified fairly high, and finished fairly high in local matches. As you learned more about competition, you practiced competition elements more, and thus fundamentals less--which meant your classification percentages dropped slightly, but your match performance got better because your shooting ability didn't decrease much, but your stage planning and movement got better so your times dropped and your hit factor went higher.

What about this is a surprise? And how does that mean the classification system is broken? Are you beating GMs that you weren't beating before? Are you being beaten by shooters that you weren't beaten by before?

Using local matches to see your relative skill works up to a point. However, increasing your score 10-20% at a local match (especially depending on if any local GMs are around) is good, but doesn't mean much on a larger scale. Or are you in a club that has 2-3 GMs in your division, and you are now 10-20% closer to them, when you started at 85% of their score in the first place?

If you started at 65% of a GM, and now end up 85%, that is entirely expected of an M-class.

Your post, as an argument, would need to:

1) show your starting point, and where this 10-20% increase occurred (for example: 45% to 60% isn't a big deal, and you have a way to go; 75% to 90% of a GM is about what you should expect to be doing as an M),

2) show that this is compared to a decent set of reference points (for example: three GMs shoot in your division, and your scores relative to them have increased; another example: you normally beat the A-class guys by 5%, but now you are consistently beating them by 20%), and

3) that your original classifications were the result of a consistent set of scores, as opposed to a random number of high scores (interspersed with low scores) such that your scores, while consistent, have gone down slightly over time. (As opposed to a set of random scores where you still have low scores, but your high scores just don't have quite as many ones that are quite as high.)

Without a comparison point, a scale, and a classification that was based on consistency, your situation is no surprise. Happens to plenty of people.

[DonovanM]

I disagree with the concept of shooting classifiers as many times as you want to, but it is specifically allowed in the classifier rule book.

Why do you care? Does it hurt your feelings if some of the people you beat in matches are classified higher than you?

WTF? Talk about a non sequitur.

What part was hard to understand? I can explain further. Someone disagreed with the concept of re-shooting classifiers. The only effect of re-shooting (I can think of) is that a few people may classify higher than they would have otherwise, even tho their actual shooting abilities are unchanged. Why should someone care about that?

Because in a normal stage, if your gun malfunctions or you have a brain fart or something you're stuck with your score unless something out of your control happens and you get a reshoot. For some reason with classifiers this is treated differently. You can reshoot a classifier for basically whatever reason you wish. I don't see why classifiers shouldn't be as much a reflection of your on-demand ability to perform ONCE as any other stage is, and I believe the integrity of the classification system is compromised by that.

[Thomas H]

Just goes to show that the skills required to succeed at short classifier stages are fairly distinct from the skills required to succeed at matches and/or long field courses. That's all.

I'd disagree fairly strongly with that---simply because to succeed at major matches requires the skills necessary to succeed at short classifier stages PLUS other match skills.

You can have classifier skills without match skills. However, above a certain point, you can't have match skills without classifier skills.

And now for some opinions: There's nothing wrong with a shooter who is better at classifier stages than they are at long field courses, and can't, let's say, shoot 95% or better at Nationals. Because frankly that's all but 3 or 4 GMs in a single division.

I'd be really curious to see some data on that. It is true that there aren't many GMs who score 95%+ at Nationals. However, that isn't the same as saying there are only a few GMs who are not "better at classifier stages than they are at long field courses."

Again---expecting the percentages to match is ridiculous. However, observing the classification system and the relative finish levels at major matches, we can see that the system itself is mostly spot-on.

[DonovanM]

I wonder if using Area & the Nat'l Match is a better way of classifying shooters?

Most good shooters shoot at least 2 of these/year - if I shoot Area 6 & the Nat'ls

at 78%, shouldn't I be an A? Regardless of what my classifiers say?

Just a thought ...

At one point I came up with the idea of capping shooters at A class until their performance at an Area or Nationals match proved otherwise. I think this would be better than the current system, but the current system is OK.

[DonovanM]

Our local club has an annual classifier match every Jul (5-6 stages) plus 2-3 more classifiers thrown in throughout the year for a total of 9-10 every year. This way there is ample opportunity for anyone interested in classifcation to do whatever is necessary. Might consider suggesting to your MD to hold an annual classifier match, it's great way to get a lot of competitiors to your local match ....and compete against folks you normally don't ...

Donovan shoots 2 - 3 classifier matches per year (with about 6 classifiers per classifier match), so it's not that.

That is true... Creep.

[DonovanM]

Just goes to show that the skills required to succeed at short classifier stages are fairly distinct from the skills required to succeed at matches and/or long field courses. That's all.

I'd disagree fairly strongly with that---simply because to succeed at major matches requires the skills necessary to succeed at short classifier stages PLUS other match skills.

You can have classifier skills without match skills. However, above a certain point, you can't have match skills without classifier skills.

Yeah, I'll buy that.

However, if you track over a period of time two shooters who started out with equal short course skills, one further practicing their short course skills and the other their ability to apply their short course skills to long courses and other concerns for long courses, I would argue the latter would be closer to the top shooters in the sport than the former would, even if the former made Master and the latter was still in A class or something.

I'm not saying the current system sucks or is broken or something. It's pretty good. But I think it could be better.

And now for some opinions: There's nothing wrong with a shooter who is better at classifier stages than they are at long field courses, and can't, let's say, shoot 95% or better at Nationals. Because frankly that's all but 3 or 4 GMs in a single division.

I'd be really curious to see some data on that. It is true that there aren't many GMs who score 95%+ at Nationals. However, that isn't the same as saying there are only a few GMs who are not "better at classifier stages than they are at long field courses."

Again---expecting the percentages to match is ridiculous. However, observing the classification system and the relative finish levels at major matches, we can see that the system itself is mostly spot-on.

Yes. Expecting the percentages to match perfectly is silly.

However, if a GM can shoot 100% classifiers with regularity, but not score better than, say 90% at an Area match or something, aren't they better at classifiers than complete matches?

[bbbean]

The exception does not disprove the rule. We periodically see posts that suggest the classifier system is broken because in some cases, people's match perfomance differs from their classification percentage. Similarly, I personaly know tall people who aren't very good at basketball.

If you look at level II and level III matches, you'll see that classification turns out to be a pretty good predictor of where shooters finish. That is to say, GMs tend to finish on top and D class shooters tend to finish on bottom. Further, MOST shooters tend to finish with percentages that are in or close to their classification percentages. That doesn't mean that there aren't Ms beating GMs, or D's beating Cs, but as a general rule, if you tel me a shooter is a C class shooter, I'm pretty sure they're going to finish with a score between 35% and 55% of the top GM at the match.

The classifier system also performs another important task - it's the reason most shooters join USPSA. There aren't 18000 shooters competing in level II and level III matches. Most USPSA members are shooting club matches where there may or may not be GMs to compare scores with, and the stages may or may not be level III quality. The classification system gives them a way to measure their progress over time, and a way to see how they stack up against the best of the best without traveling to 2-3 day matches and paying $100-300 entry fees plus travel expenses.

Personally, my classifier scores have proven an excellent predictor of where I finish at major matches. In the handful of cases where they haven't, it's been due to identifiable factors (i.e. I just tanked BitB, but my practice regimen has been VERY slack this winter). I know that when I practice, clear distractions, have my equipment working, and focus, I can turn in a solid B class performance just like I can turn in a solid B class classifier. If I get distracted or lazy, I turn in an it'll do C class performance. Hopefully I'll bump those up to A and B by the end of this year.

[DonovanM]

Major matches don't compare well with classifications IMO. A top GM will shoot classifiers well over 100% and push everyone's match % down as well.

Yes, but...

...if you compare match finishes, the classification system works remarkably well. It is only if you expect classification percentages to equal match percentages that you have a problem.

Way back when, I did an analysis of a Nationals, and compared scores from different classes within the Production division---and not only were the classes clearly separated within the match, but there was an obvious and clear relation between times and scores for each class.

That means that the classification system is pretty darn accurate, in terms of giving relative skill.

This is not to say in any way that the classification percentages actually work that way, however. The same analysis also showed that most people did NOT necessarily shoot within their classification percentage at the match (amusing to see how many GMs did not shoot GM percentages at Nationals) ---but then again, I wouldn't expect them to do so.

Classification tests certain fundamental shooting skills. Matches test varying sets of those skills plus a number of other things. As such, expecting the numbers to match seems odd.

In response to Donovan's original post: So, you have solid fundamental shooting skills that you'd been practicing, and thus was classified fairly high, and finished fairly high in local matches. As you learned more about competition, you practiced competition elements more, and thus fundamentals less--which meant your classification percentages dropped slightly, but your match performance got better because your shooting ability didn't decrease much, but your stage planning and movement got better so your times dropped and your hit factor went higher.

What about this is a surprise? And how does that mean the classification system is broken? Are you beating GMs that you weren't beating before? Are you being beaten by shooters that you weren't beaten by before?

Using local matches to see your relative skill works up to a point. However, increasing your score 10-20% at a local match (especially depending on if any local GMs are around) is good, but doesn't mean much on a larger scale. Or are you in a club that has 2-3 GMs in your division, and you are now 10-20% closer to them, when you started at 85% of their score in the first place?

If you started at 65% of a GM, and now end up 85%, that is entirely expected of an M-class.

Your post, as an argument, would need to:

1) show your starting point, and where this 10-20% increase occurred (for example: 45% to 60% isn't a big deal, and you have a way to go; 75% to 90% of a GM is about what you should expect to be doing as an M),

2) show that this is compared to a decent set of reference points (for example: three GMs shoot in your division, and your scores relative to them have increased; another example: you normally beat the A-class guys by 5%, but now you are consistently beating them by 20%), and

3) that your original classifications were the result of a consistent set of scores, as opposed to a random number of high scores (interspersed with low scores) such that your scores, while consistent, have gone down slightly over time. (As opposed to a set of random scores where you still have low scores, but your high scores just don't have quite as many ones that are quite as high.)

Without a comparison point, a scale, and a classification that was based on consistency, your situation is no surprise. Happens to plenty of people.

Good post.

I want to address #3 in your list of prerequisites. In the time when I was close to making GM, my classifier scores were NOT consistent, but the system didn't allow that inconsistency to effect my classification score all that much. Fortunately, the timing was off so I didn't get a couple more of them closer together which would have ended up with me making GM. It easily could have happened, and I'm glad it didn't. Fast forward to today and I seem to almost be able to shoot 85+% classifiers on demand. (See here) Almost, not quite. This alone is such a night and day difference in terms of actual match performance - it's the difference between going hero or zero all the time and occasionally connecting, to just shooting solidly and consistently - but yet my classification percentage is lower. I don't think it should be that way. If I had made GM last year, and today I was a much better shooter even though I hadn't more than one GM classifier in recent memory... well that would just be weird, and worth addressing. That's all.

You are correct to question my experimental conditions for the conclusions I made. I really have no way to nail down some set percentage that I either improved or got worse by. I can't really show my starting point, as performances - especially mine - are so variable. All I can do is give you the list of all of the results from all of the club matches I've shot and told you to look at the long-term trends, which I would beg you not to do because it would be a colossal waste of time and not at all appropriate for this level of discussion. But anyone who's shot with me can tell you from a purely objective standpoint that I've improved greatly in the past year. I know, that is about as scientific as anthropology, but I thought it was enough to warrant discussion.

[DonovanM]

The exception does not disprove the rule.

That is all this is really, just an exception. I think the system could be tweaked a bit to compensate for it though, and make it more dependent on consistency and less forgiving of inconsistency. That would be awesome, and ratchet up the difficulty a bit for inconsistent shooters (like me) to reach the higher classifications.