USPSA classifications.

[Ron Ankeny]

Do any of you guys know how USPSA came up with the method they use to trim the data set on classifiers? For instance, here are some open classifiers from the past year, 82.4, 84.7, 79.5, 77.6, 76.5, 78.6, 71.9, 40.2, 73.2, 80.5, 70.4, 87.4, 39.5, 97.9, 80.0, 79.3, 75.1, 67.0, and 84.1. Using traditional evaluation and measurement methods, be it averaging everything, removing outliers, massaging the data set with a statistical analysis pack, etc. there is no way I can get the data any higher than the upper 70’s. USPSA manages to get this mess to average out to 86%. Out of 20 classifiers there are only two over 85% and one of those is obviously so high as to skew the data set.

The revolving window and flagging system often removes perfectly valid scores in favor of scores that are suspect at best and even if the scores are valid, they represent the shooter’s very best efforts, not what the shooter could reliably perform under similar circumstances. This is especially true of shooters who just break over into the next class up, especially Master class. A low end Master will have scores 6% below their average tossed (perfectly valid scores) but USPSA will keep scores that are two classifications higher. I think that’s why we see so many M class shooters mixed in with A class shooters at bigger matches. It isn’t the shooter grandbagging, it’s the system.

So, why does USPSA use a system that obviously places shooters in a position where they can not be expected to reasonably perform? Why is it so upwardly biased? I am not complaining I am just curious.

[kdj]

So, why does USPSA use a system that obviously places shooters in a position where they can not be expected to reasonably perform? Why is it so upwardly biased?

A cynical response would be "Because most people like to hold the highest rank they possibly can" and making people feel good about them selves is the purpose of most classification systems these days

Truthfully, why not do it this way since it probably does make more people happy than any alternative. Is there any other reason to have a classification system?

Kevin

[vincent]

I'm not sure it is totally upward biased if you look at both side of the coin.

Basically I think it can be upward biased once you get to A class. This is in part because A class and M class are only 10 percentage points wide. Plus in M class you can't have too high of a score but anything lower than 75% is disregarded. This allows people in A and M class to "go for it" on local classifiers because if they over-extend and tank the score is much more likely to be disregarded.

However the opposite can be true for B and C class shooters. First off C class is 20 percentage points wide and B class is 15 points wide. This opens up the available score ranges, and the possiblility of a low score hurting your efforts is very real. For example in B class you can have classifiers at 50% count. You can't just let it all hang out and hope for a good performance because the bad performances will hurt you.

In the end I'm not sure how much it matters since most people I know look at overall results anyway.

For reference:

Scores are questioned that are more than 15% above your current class or more than 10% below your current class.

GM 95 to 100%

M 85 to 94.9%

A 75 to 84.9%

B 60 to 74.9%

C 40 to 59.9%

D Below 40%

[shred]

"Getting out of B is 90% of the way to Master"

The way the HHFs are now, Ron could have shot exactly the same ability level each time and still got the above score set.

Perfectly-normal-A-shooter shoots at a perfectly A-class ability level and can get back %'s from USPSA that range from 50% to 100%

Until they fix that (I have hopes for the 03-series), any other bias is secondary to me.

Btw, I notice more long-time non-serious shooters stuck in B than A or M-- the "allowed low" on B is much lower and it's much harder to get rid of one or two bad performances, especially if you're inconsistent, or get crappy-HF classifiers occasionally.

[MoNsTeR]

My first intuition is that they didn't really think about it, and don't want to change it now.

It also occurs to me that sandbagging is the bigger problem, and an upward-biased system helps fight it. Given how easy it is to raise one's class versus how hard it is to lower it, this seems the obvious answer.

[MasterLefty]

Plus in M class you can't have too high of a score but anything lower than 75% is disregarded.  This allows people in A and M class to "go for it" on local classifiers because if they over-extend and tank the score is much more likely to be disregarded.

Actually it is 80% for Master. I spend a lot of time at the local matches playing and trying some different things. (In fact it had a few of my local shooting buddies worried that I wasn't in my usual form this year.) I don't know that I just 'go for it' on classifiers though.

As a Master, I fall under the old protected status. I cannot lose my classification, as such I sometimes play with a classifier or the whole match. However, I could care less that Sedro has to keep 33 months worth of classifiers to maintain my M average. It wouldn't bother me if they showed that my current average was 78.780, high average is 90.743 and my classification is Master. If I show up and shoot to my ability, I am a Master class competitor.

I know that there were abusers to the system and that is why it was probably changed from working your average on the best 6 of the last 8 classifiers regardless of the result. Would I like to see my 100% and 99.2% roll off the books? You bet, at least until I am able to consistantly run at that level. However at my current rate it will take 4 years and by that time they may have determined that my average is above 95%.

Back to the topic. Our current system is biased to getting everyone moved up, unless you are just super consistant within you classification.

Kenny

[h4444]

That's why I want to win my cards in Limited.....and don't care as much about the others.

I like Limited the best and want those cards to come from direct head to head competition with the best shooters in the world.

Other divisions,,,,I'm not as concerned with,,,,,if I go A or higher in another division through classifers,,,,,I'm not going to sweat it too much.

Limited, I want to win those.

H4444

[redmist10]

So, just for point of clarification for newbies like myself....

Say you're in C class and you have a 57%...you shoot a couple of good classifiers back to back that push you to 60.1% and you jump to B class.

Then your next eight plus classifiers are your "normal" or slightly below normal performance and you're current percentage drops to 50% but your high percentage is 60.1%...you're still a B-class shooter?

Just trying to make sense out of it all. Sucks to have to keep a 42% and drop a 55% that were shot within a few months of each other just because the 55% isn't in the last 8 classifiers.

[Jake Di Vita]

That is correct.....the only way you can move down in classification is to petition USPSA with good cause. Once your percentage goes into the higher class, you can't go back.

[XRe]

Just curious, since I'm sort of in this boat, now - what if I haven't been classified for 5 years, or so?? I'm sure I could go do a search or read a bunch on the USPSA site (if I can get to the info), but.... I was classified as an M in Open, and A in Limited (L-10 and Production didn't exist at that point). Am I going to end up in A for L-10 "auto-magically" ???

[SRT Driver]

But the vast majority of shooters are B/C class in all divisions.

Just speaking for Production, 78% gets you the top 3.5% nationally but 22% below 100%. (93 of 2614).

The other divisions are close but not far off.

So is there an upward bias? I don't know but the "average" shooter is a high C class shooter.

[eric nielsen]

What Vincent & Shred said are completely true. It's harder to get out of C class and B class because of the allowable low scores.

A lot of my 'low' allowable scores were because I blew a speed-shoot type reload trying to score 95 or higher, and ended up with a 65 or something like that, on a speed shoot that takes like 5.5 seconds to complete with a smooth mag change. No room for wasted time at all.

A-class is a short trip for many shooters, for the reasons mentioned, also because now a 99% or what-have-you will not be flagged as 'too high.' For a B-shooter anything above 90 is flagged - unless you call or write to Val.

I used to write a nice note on the classifier sheet but I believe now that Sedro Wooley just gets a Classifier Report - electronically - is that correct?

[Chriss Grube]

Classifiers basically mean squat. It is a system that MIGHT show your ability. You want to find out where you stand go shoot a few big matches where the big boys are playing. Those are the scores that mean something. There are lots of paper GM's and M's that go to the big shows and get crushed by lower classed shooters. I say throw the whole system out and shoot head to head. Of course that will never fly in todays world because somebody's ego will get crushed because they didn't win a prize. Most decent shooters can practice a lot for a year and make it up to M maintaining that level of dedication is the hard part. Take a look at last years Nats percentages from Open. Nobody outside the top 10 shot a master class score. You have GM's shooting B class scores. The system is a joke and either needs eliminated or a total overhaul.

[lynn jones]

Just curious, since I'm sort of in this boat, now - what if I haven't been classified for 5 years, or so?? I'm sure I could go do a search or read a bunch on the USPSA site (if I can get to the info), but....   I was classified as an M in Open, and A in Limited (L-10 and Production didn't exist at that point). Am I going to end up in A for L-10 "auto-magically" ???

<{POST_SNAPBACK}>

yup!

[driver8M3]

the other obvious problem with classifiers is that real matches (club, sectional, area and national matches) have very few stages that actually resemble a classifier (stand here, shoot 6, reload, shoot 6 more). thats understandable (b/c there is no way to you can measure out and standardize big field courses), but it's a lot easier to get good at stand-and-shoot stages than 32 round field courses with lots of running, windows, doors, turners and swingers.

[XRe]

Am I going to end up in A for L-10 "auto-magically" ???

<{POST_SNAPBACK}>

yup!

<{POST_SNAPBACK}>

Probably going to be a little while before I have to worry about moving up to M, then....

[lynn jones]

Am I going to end up in A for L-10 "auto-magically" ???

<{POST_SNAPBACK}>

yup!

<{POST_SNAPBACK}>

Probably going to be a little while before I have to worry about moving up to M, then....

<{POST_SNAPBACK}>

yeah, my classification in revolver is about 50% with four classifers reported. as soon as the're six or so i'm in A class.

[Flexmoney]

The real bias in classifiers is that the shooter treats them different than other stages...enter the trying zone (also know as the go fast, see nothing, and miss a lot...zone)

[Jake Di Vita]

Totally echo what Flex said.

I think the vast majority of the problem would sort itself out if people shot the classifier like any other stage.

[SRT Driver]

I think the intent is to group everyone together that has the same basic skill level.

But I think we all know you can't test everything.

The reason we get crushed by the upper 1% is that they are THAT good!! Just like any sport..the top are strides better than the rest.

And at major matches, what is usually "enough" gets pushed up to 150%.

[SRT Driver]

After some more thought, I think the classification system is reasonable.

If you shoot your card at a major match..or close.. that's good. Is it possible all of the time..no, at least in my case.

Like Flex said when trying comes into play, that's when the nuts come loose. Everyone wants to win but not everyone has the capability to deal with the mental aspect of being in contention at major match and the pressure associated with it.

Some are really good at that. Especially if they have been in the same position in another sport and can carry it over.

My point is, the classification system measures your basic skill. It's up to you to take it to the next level mentally.

Is it possible to treat a major match like a local match? For some, I suppose but for mortals, like me, that spend a significant amount of time and money to attend..it's always in the back of my mind.

If I could learn to let that go, there would not be any excuses. Other than the satisfaction that I performed the best I could and still got crushed, I can't ask for anything more.

[Jake Di Vita]

"Is it possible to treat a major match like a local match? For some, I suppose but for mortals, like me, that spend a significant amount of time and money to attend..it's always in the back of my mind."

I do the opposite here. I treat every match (including local matches) as if they were nationals.

[Gwhiz]

Major matches need to be counted 2 to 3 times more than classifier stages. Us old fat guys can do ok stand and shoot, but 30+ round field courses are another thing altogether.

If we are going to recognize classes, leave us where we belong.

[Flexmoney]

At a Major match, the shooter will almost always shot below their class. Unless they ahve a great match.

The reason is simple math.

At a major match...you are scroed off of the ONE shooter who has 100%.

On the classifers, the idea is that the high hit factor that you are compared against...it is an average of the top ten scores turned in. (Plus, the hhf's might not get adjusted up as often as they should.)

[shred]

FWIW, the upward bias stops around M (since you can't go over 100%). You've gotta be both good, consistent (or wildly inconsistent) and a bit lucky (or alternatively, just be really good) to get the G.