Fencing.Net - Your Guide to Fencing & Fencing Equipment

dekko · 04-06-2006, 02:53 PM

Quote:

Originally Posted by Beowulfman6

I can name at least one other epeeist in the Southeast who just earned his A this year.

It's ok, Frank refuses to admit there are fencers in NC. One day he will learn differently.

telkanuru · 04-06-2006, 06:07 PM

This seems to be an issue that's relevant only if people particularly care about ratings. Me, I just beat whoever's hooked up to the other reel...

So, the question is why do you care about your/other people's ratings?

rcmatthews · 04-06-2006, 06:49 PM

Thats very high and mighty of you. Its very in vogue to hate on ratings, but I don't really see why.

I think of ratings sort of like the preseasons rankings in other sports. They aren't going to tell you exactly where everyone will finish, but they do give you a pretty good idea of whos going to be decent.

Reasons to be concerned with your rating:
1) c or above gets you to div 1 events
2) it's an indicator of relative progress
3) better seeding into pools
4) it sets a goal for something to work towards

mrbiggs · 04-06-2006, 09:17 PM

Quote:

Originally Posted by telkanuru

So, the question is why do you care about your/other people's ratings?

I care about mine because it's a goal and because it helps with pool seedings.

I care about others' because it can give an approximation to how well they fence.

Fechter1 · 04-06-2006, 11:22 PM

Quote:

Originally Posted by rcmatthews

Reasons to be concerned with your rating:
1) c or above gets you to div 1 events
2) it's an indicator of relative progress
3) better seeding into pools
4) it sets a goal for something to work towards

Same here. (Except for #3... I'd actually like harder pools than I've been getting.)

bigdawg2121 · 04-07-2006, 01:17 AM

I'm not gonna go as far as Telk and say that ratings don't matter, b.c. to some extent the reasons given are all rather valid. The rub is that if there's ratings inflation 2,3, and 4 don't really mean much. Also when everyone has the high ratings (at national tournaments especially) it doesn't matter where you get seeded. Although really seeding only makes but so much difference anyway, you could seed me as a C at a Div 1 NAC and I could still be reasonably sure that I'm making it out first round of pools no worse than 3-3 if all goes horribly. You fence at the level you fence at and the pools generally coome out evenly distributed more or less.

rcmatthews · 04-07-2006, 01:42 AM

Yeah, but if you are seed as a C, you have more higher ranked people in your pool. Instead of you being the A, there is someone else, thus increasing the likelihood of a loss.

KD5MDK · 04-07-2006, 02:06 AM

Quote:

Originally Posted by rcmatthews

Yeah, but if you are seed as a C, you have more higher ranked people in your pool. Instead of you being the A, there is someone else, thus increasing the likelihood of a loss.

At a Div I Epee NAC, it's more like a pool of 3As, 3Bs and a C.

JEC · 04-07-2006, 02:14 AM

Quote:

Originally Posted by KD5MDK

At a Div I Epee NAC, it's more like a pool of 3As, 3Bs and a C.

Typically, with 2 fencers (A's) having Div. I National points.

bigdawg2121 · 04-07-2006, 02:26 AM

At the end of the day you're going to have two at least that many As Bs and Cs in your pool anyway, yeah it's a bit harder being seeded as the C but again, if it's a Div 1 NAC and I'm a strong Div 1 fencer (similar analogies for whatever level of tournament) the initial seed isn't going to mean that much. One of the As is going to be really good, one is going to be middle of the pack even with points, one is going to be a random A from somewhere, one B will be an underrated B that should have an A, the other two will be w/e and the C is the C. What difference does it make? I figure if I can't muddle through a first round pool that DE's probably weren't meant for me that day. I mean really, first round is just for seeding purposes it ain't that rough usually.

KD5MDK · 04-07-2006, 03:33 AM

Well, it's also for the 40% cut...

RoninX · 04-07-2006, 11:54 AM

So the general consenus seems to be acknowledgement that this occurs but no one seems that concerned about it.

My real problem with this system is actually the reverse situation from the one I initially posed. That in some areas that are initially lacking an As and Bs that compete locally frequently it gets very difficult to build up a rating base, and my money is on this being aa much more frequent skewing problem than the former.

keith · 04-07-2006, 12:09 PM

Quote:

Originally Posted by RoninX

My real problem with this system is actually the reverse situation from the one I initially posed. That in some areas that are initially lacking an As and Bs that compete locally frequently it gets very difficult to build up a rating base, and my money is on this being aa much more frequent skewing problem than the former.

Surely only a problem for the fencer who cannot earn a C and is prevented from getting into a DivI NAC?

.... and from previous threads on this topic I can only think of one place that this applies to

rcmatthews · 04-07-2006, 12:25 PM

Quote:

Originally Posted by bigdawg2121

At the end of the day you're going to have two at least that many As Bs and Cs in your pool anyway, yeah it's a bit harder being seeded as the C but again, if it's a Div 1 NAC and I'm a strong Div 1 fencer (similar analogies for whatever level of tournament) the initial seed isn't going to mean that much. One of the As is going to be really good, one is going to be middle of the pack even with points, one is going to be a random A from somewhere, one B will be an underrated B that should have an A, the other two will be w/e and the C is the C. What difference does it make? I figure if I can't muddle through a first round pool that DE's probably weren't meant for me that day. I mean really, first round is just for seeding purposes it ain't that rough usually.

This logic doesn't make sense to me. Sure, it might only be a marginal difference, but its still a difference. I don't think an NCAA basketball team would say, "Oh, well, were going to have to play hard teams anyway, so a number 12 seed is just as good as a number 1"

oiuyt · 04-07-2006, 12:36 PM

Quote:

Originally Posted by bigdawg2121

One of the As is going to be really good, one is going to be middle of the pack even with points, one is going to be a random A from somewhere, one B will be an underrated B that should have an A, the other two will be w/e and the C is the C.

The difference is instead of having a killer-A, being a solid-A, having a random-A, an underrated-B-that-should-be-A, and two B's and a C that one should beat, barring a fluke result, one receives a pool with a killer-A, another solid-A, a random-A, an underrated-B-that-should-be-A, and two B's that one should beat (barring flukes).

This means not having the bout against the C, which was an easy bout, and replacing it with a bout against another solid-A, in other words a 50-50 bouts (assuming one is a solid-A). You've lost about 1/2 of a win in the process.

Now let's say that you're a B who's overrated as an A vs. a B who's underrated as a C vs. a B who's correctly rated.

In the first case you have 3 bouts you should lose, 2 50-50 bouts, and a bout that you should win. You're expected to go 2-4 and be on the cusp of making the second round.

In the second case you have 4 bouts you should lose and 2 50-50 bouts. You're expected to go 1-5 and be done.

In the third case you have 4 bouts you should lose, 1 50-50 bout, and 1 bout you should win. Depending on how you do in the 50-50 bout, against a similar-level opponent, you might make it (you're fencing well that day) or you might not (you were fencing a bit below your norm).

Then add in random fluctuation of about a bout to each of those scenarios. Clearly case 1 leaves you in a significantly better position than case 2 or 3, and can absolutely make the difference between advancing and not out of that first round. The solid-A with low points type person DOESN'T have to worry about incestuous classifications, he's getting his without that and isn't improved by the phenomenon. Look just a bit below that level and it DOES definitely start becoming important.

-B

keith · 04-07-2006, 12:58 PM

Quote:

Originally Posted by oiuyt

The difference is instead of having a killer-A, being a solid-A, having a random-A, an underrated-B-that-should-be-A, and two B's and a C that one should beat, barring a fluke result, one receives a pool with a killer-A, another solid-A, a random-A, an underrated-B-that-should-be-A, and two B's that one should beat (barring flukes).

....snips the rest....

Which is perhaps al fair concern if the competition is a one off, don't think DivI (or any regular event) counts as that.

Any fencer who is competitive at that (or any) level is going to have the results of that ability (national points in this case). Now if you have national points and still have to hope for good 'match ups' in your first pool in order to progress, well......

Now every year there may be a bunch of ringers due to inappropriate ratings, but if they actually have the ability then *shrugs*. Besides not sure this is any more unfair than the antics of a weak A/B who decides to get serious and trains hard with a brilliant coach etc - now that is terribly unfair to the rest of his/her pool the first DivI they enter

RebelFencer · 04-07-2006, 01:04 PM

Quote:

Originally Posted by oiuyt

The difference is instead of having a killer-A, being a solid-A, having a random-A, an underrated-B-that-should-be-A, and two B's and a C that one should beat, barring a fluke result, one receives a pool with a killer-A, another solid-A, a random-A, an underrated-B-that-should-be-A, and two B's that one should beat (barring flukes).

This means not having the bout against the C, which was an easy bout, and replacing it with a bout against another solid-A, in other words a 50-50 bouts (assuming one is a solid-A). You've lost about 1/2 of a win in the process.

Now let's say that you're a B who's overrated as an A vs. a B who's underrated as a C vs. a B who's correctly rated.

In the first case you have 3 bouts you should lose, 2 50-50 bouts, and a bout that you should win. You're expected to go 2-4 and be on the cusp of making the second round.

In the second case you have 4 bouts you should lose and 2 50-50 bouts. You're expected to go 1-5 and be done.

In the third case you have 4 bouts you should lose, 1 50-50 bout, and 1 bout you should win. Depending on how you do in the 50-50 bout, against a similar-level opponent, you might make it (you're fencing well that day) or you might not (you were fencing a bit below your norm).

Then add in random fluctuation of about a bout to each of those scenarios. Clearly case 1 leaves you in a significantly better position than case 2 or 3, and can absolutely make the difference between advancing and not out of that first round. The solid-A with low points type person DOESN'T have to worry about incestuous classifications, he's getting his without that and isn't improved by the phenomenon. Look just a bit below that level and it DOES definitely start becoming important.

-B

You make my head hurt.

Dr. Pfleschbach · 04-07-2006, 01:07 PM

Quote:

Originally Posted by telkanuru

This seems to be an issue that's relevant only if people particularly care about ratings. Me, I just beat whoever's hooked up to the other reel...

So, the question is why do you care about your/other people's ratings?

You've got the right attitude. Ratings are just for seeding purposes. Eventually everything works out the way it should because fencing is the ultimate Darwinian experience.

oiuyt · 04-07-2006, 01:11 PM

Quote:

Originally Posted by keith

Any fencer who is competitive at that (or any) level is going to have the results of that ability (national points in this case). Now if you have national points and still have to hope for good 'match ups' in your first pool in order to progress, well......

Well, no. What you're looking at is how well do the top people do in the early rounds. Competitive at a given event is the middle of the pack. So absolutely my example is appropriate. We're not looking at whether or not being over/underrated can help/hurt your chances of winning a national event.

If Seth Kelsey got seeded as a C in Sacramento would it have a huge impact on his final result? Probably not. He'll have a bout in the first round that is harder than it should have been (perhaps he'll fence Eric Hansen instead of a C at his first ever D1 NAC). He'll seed into the second round lower than he otherwise ought to have (win or lose against Eric, it'll presumably hurt his indicators at least). He'll have a harder second round pool than he ought to, although to a lesser degree than the first pool was. He'll enter the DE tableau a few slots lower than he should. It'll end up having a small effect in final placement.

If a random high-level B enters does his classification have a significant impact on his final result? It absolutely can.

Mis-classification can have an impact of as much as half a bout up or down (so as much as a full bout if you count the swing from massively underrated to massively overrated; if you start at one end of the spectrum you can only gain one of these two directions). Does an extra half to full victory in the first round matter? Frequently. At any level. That's what seeding can do for you. Will it win/lose the tournament? No. Can it affect results? Absolutely.

-B