Fencing Incest Breeding As and Bs?

[telkanuru]

If by 64 you mean 32 since that's where points go to

Re: various critics- It seems to me that if someone has issues with rating inflation, they are looking at the system as a high rating makes you a good fencer as opposed to one where good fencers tend to have high ratings. Ratings are a nice side effect of being good, but not an end in and of themselves.

[keith]

Originally Posted by telkanuru

If by 64 you mean 32 since that's where points go to

the dangers of doing a quick google and trusting the results .

[Dr. Pfleschbach]

The worst case theoretical outcome from inaccurate seeding due to bogus A's and B's is that the top fencers will face each other too early in the DEs, but that's just not realistically going to happen. The guys in the middle will be more effected but it's not going to seriously slant the outcome of the tournament. The guys who are good will over all do better than the guys who aren't.

[D+F+P=Hadouken!]

Ratings do work. Just not perfectly. Look at most tournaments. The better rated fencers tend to gravitate toward the top of the results. Not always, but a majority of the time.

[JEC]

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 ... 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

I had proposed earlier (see: http://www.fencing.net/forums/post354832-16.html ) a system to improve seedings, which in the end, it is the primary reason why we have ratings, to seed properly.

Originally Posted by JEC

I think the easiest and most helpful change would be to add the number of times it has been re-earned.

i.e.: A05-8 vs A05-1. Clearly, somebody who has re-earned 8 times their A would be better than somebody who has earned only once.

For Div.I, it would mean that better seedings of pools would occur. However, a fluke B03-01 would still be seeded ahead of a C06-08. Nevertheless, the system would correctly identify a strong C06-08 from a C06-01, who would be seeded closer to last

[keith]

Originally Posted by JEC

For Div.I, it would mean that better seedings of pools would occur. However, a fluke B03-01 would still be seeded ahead of a C06-08. Nevertheless, the system would correctly identify a strong C06-08 from a C06-01, who would be seeded closer to last

... but the case under debate is a C06-52 vs an A06-01 (for example). With the poor C unable to get an A or even a B due to the lack of suitable competition in their area.

For DivI (or any national championship) all the usual suspects will have points and, bar the odd ringer, everyone else is there to get them warmed up. Accurately seeding the middle and bottom of the pack hardly seems worth the effort.

[mrbiggs]

Originally Posted by JEC

I had proposed earlier (see: http://www.fencing.net/forums/post354832-16.html ) a which in the end, it is the primary reason why we have ratings, to seed properly.

that would improve it, but it would require a huge amount of effort on the part of the USFA because they'd have to print and mail so many cards.

[oso97]

Originally Posted by mrbiggs

that would improve it, but it would require a huge amount of effort on the part of the USFA because they'd have to print and mail so many cards.

Thus proving the point that the entire system is antequated and we need to move into the 21st century with on-line membership and classification databases.

Hrmmm... where have we heard this before?

[bigdawg2121]

Hmm, Brad I see your point and I fully agree on an academic level. That being said, speaking purely as a competitive fencer and one who was very underrated for a very long time due to being in an area where getting higher ratings was nigh impossible, I don't care. If I can get into the tournament you can put whoever in my pool b/c if I'm good enough to get the results I want to get then who you put in my pool is a nominal difference at best. If not then it doesn't much matter anyway. Obviously there's an academic difference and at some point not too long ago there was an almost exponential seeding advantage for a higher rating at a large tournamentl, but really unless you're on track for a team somewhere (in which case you're ranked on pts anyway) the difference is relatively negligible I think.

[KD5MDK]

In events with >100 As, seeding perfectly is somewhat less important than in events with 1 A per pool, or with some pools having 2.

[Basquet]

BCAF, a NJ club, had only one A-rated fencer, a veteran epeeist, for quite a few years. At the Pittsburgh NAC in December 2005, two more BCAF veterans earned their A's. Since then, six epee tournaments have been held at BCAF and 3 new A's have been created (all BCAF fencers), for a total of 6, as well as two new B's (not counting renewed B's, but two brand-new B's), one of whom is a BCAF fencer. All in the past four months! So, yes, I would say that ratings breed ratings! And the word incestuous might be applicable in this case. Pretty soon the three (I think) remaining epee B's at the club will have their A's too, because they have all beaten all the A's at one time or another...

[mrbiggs]

Originally Posted by KD5MDK

In events with >100 As, seeding perfectly is somewhat less important than in events with 1 A per pool, or with some pools having 2.

Some will probably have points, though.

[tallntender]

Originally Posted by Basquet

BCAF, a NJ club, had only one A-rated fencer, a veteran epeeist, for quite a few years. At the Pittsburgh NAC in December 2005, two more BCAF veterans earned their A's. Since then, six epee tournaments have been held at BCAF and 3 new A's have been created (all BCAF fencers), for a total of 6, as well as two new B's (not counting renewed B's, but two brand-new B's), one of whom is a BCAF fencer. All in the past four months! So, yes, I would say that ratings breed ratings! And the word incestuous might be applicable in this case. Pretty soon the three (I think) remaining epee B's at the club will have their A's too, because they have all beaten all the A's at one time or another...

Well, I looked up the results of the most recent BCAF tournment on AskFRED. It was an A2 with 34 fencers. Five As and 10 Bs entered. Three As, four Bs and a D reached the final eight (he earned a C). Three of the top eight were from BCAF. The 2, 3T, 6, 7, and 8th places were from other clubs.

Two of the As who entered the tournament were defeated in the round of 16. Neither of those guys is a pushover. One of them had national points last year. Only two of the 10 Bs were from BCAF.

As leading to more As? Perhaps. Incest? I doubt it.

If you want an A it looks to me as though you can drive down to BCAF and earn it. . . .

[KD5MDK]

True, but we're not worried about inflated Rolling Points.

[Goldgar]

Originally Posted by tallntender

If you want an A it looks to me as though you can drive down to BCAF and earn it. . .

Sure, come on down! Or up, or over, or whatever. Classification-hungry fencers have been making for some very good competitions lately. And we BCAFers are all pushovers.

[tallntender]

Originally Posted by tallntender

If you want an A it looks to me as though you can drive down to BCAF and earn it. . . .

Dear Goldgar,

Sorry, the sarcasm didn't come through. What I meant was, "why don't you drive down to BCAF and earn it?"

The tournaments look legit to me.

[Goldgar]

Originally Posted by tallntender

Dear Goldgar,

Sorry, the sarcasm didn't come through. What I meant was, "why don't you drive down to BCAF and earn it?"

The tournaments look legit to me.

Not to worry, tallntender. I understood your sarcasm perfectly, and was more going along with it than trying to rebut it.

Basquet is perfectly right, of course: ratings do breed ratings. That's the way the classification system is designed to work, though it does so imperfectly. And I think it's cool that we have enough classified fencers in our club to practically guarantee that any well-attended competition can be an A1 or A2. It draws in more and better fencers. Heck, that most recent competition we're talking about was one of the smaller ones -- we've had at least one B3 this season, and a couple that weren't that far from A3 or A4, IIRC.

Is it incestuous? Well, sure we can have A-rated competitions composed of only club members, so there's a sense in which you could call it incestuous. That's not what we want, though, and it's not what's been happening. There are lots of fencers in this general area, coming to our tournaments. I'd be surprised to find that more than half the fencers in our epee competitions are from our club, if even that many.

Are our A's weak? Only competition results will tell. Are the A's in our club rolling over for the B's, to help them boost their ratings? Anybody who thinks that hasn't been here.

- Dirk

[Basquet]

Originally Posted by Goldgar

Is it incestuous? Well, sure we can have A-rated competitions composed of only club members, so there's a sense in which you could call it incestuous. That's not what we want, though, and it's not what's been happening. There are lots of fencers in this general area, coming to our tournaments. I'd be surprised to find that more than half the fencers in our epee competitions are from our club, if even that many.

While it's true that the strength of the aforementioned epee competitions is increasing enormously and pulling in fencers (especially loads of A-hungry B's) from farther and farther afield (a good thing, and not hard in a location within driving distance of many of the fencers in NY Metro, Philly, CT, DC and Baltimore and even farther afield if you want to make a weekend of it), the new A and B ratings are going to the guys with home-court advantage, overwhelmingly, right at the moment (check the results on the NJ division website, www.njfencing.org, starting in Jan.06). That's why the thread-starter's question about "incestuous breeding of new ratings" seemed to apply - jokingly, of course! And no, nobody has to roll over for anyone else's advantage...no way! It's just that they know each other so well, fencing-wise...mostly trained by the same people. And they are all pretty solid fencers, darn good on a good day, even the geezers...all it took was a critical mass of A's to increase the chances that on any given day enough would turn up and/or finish in the right bracket to keep the tournament up to A level, and the randomness that seems to characterize epee, and voila...the results speak for themselves! The ball has started rolling, and there's absolutely no stopping it now! Epee fencing in NJ is going to become very exciting. Watch out NYAC!

So by all means go try your luck at BCAF. It's worth the trip. At the very least you'll get in some great epee fencing, and the facility is nice. And the chances of your increasing your rating, whatever it is and if that's what you're after, get better and better with each tournament! Check the NJ schedule or AskFred for upcoming tournaments.

[Goldgar]

Originally Posted by Basquet

While it's true that the strength of the aforementioned epee competitions is increasing enormously and pulling in fencers (especially loads of A-hungry B's) from farther and farther afield (a good thing, and not hard in a location within driving distance of many of the fencers in NY Metro, Philly, CT, DC and Baltimore and even farther afield if you want to make a weekend of it), the new A and B ratings are going to the guys with home-court advantage, overwhelmingly, right at the moment (check the results on the NJ division website, www.njfencing.org, starting in Jan.06).

The data don't seem to support this statement. I went to the NJ website and tabulated all the classifications earned so far this year in Men's Epee or Mixed Epee events, in competitions held at BCAF. That includes one Philadelphia division event held at BCAF, since NJ Division posts those results as a courtesy.

A total of 25 new classifications were awarded (including updated classifications):

A .... 3
B .... 4
C .... 7
D .... 8
E .... 3

The distribution of those classifications among clubs was:

Club ............... Count
AFC ................. 1
BCAF ............... 7
BFC ................. 1
FAP ................. 1
Hillsborough ...... 1
Medeo ............. 7
NYFC ............... 2
PAFS ............... 1
SMMA .............. 1
Unatt .............. 1
USMAPS ........... 2

Given the relative representation of the clubs at these events, I don't think 7 new classifications out of the 25 can be taken as showing any home-court advantage.

Now, excluding all but the A's and B's from the data, the same query returns this:

Club ............... Count
BCAF ............... 4
Medeo ............. 2
NYFC ............... 1

It's a small sample, but that hardly looks like an 'overwhelming" majority to me.

Where you might have a case is if you include only those fencers who earned A classifications. There were 3 of them, and they were all from BCAF. That's certainly enough to raise an eyebrow. My personal belief is that all three of these fencers (Ferrone, Henry, and BrownGold) have been fencing very strongly for a long time -- BrownGold, for example, placed second at MidAtlantic Sectionals in 2004 -- and now there are enough A-rated competitions for that strength to show a tangible result. I'd also point out that two of those A's were not won over another BCAF fencer in the final bout -- BrownGold def. Port (Medeo), and Henry def. McGrath (Medeo). Alas, Ferrone def. Goldgar (BCAF) -- that bum!

So it's hard to say where a home-court advantage manifests itself. It's certainly possible, since (as you point out) "they know each other so well, fencing-wise...mostly trained by the same people." On the other hand, doesn't that supposed advantage work both ways? I'm inclined to agree that knowing each other well and fencing each other often tends to flatten out the technical disparities between two fencers, so there probably is some effect there. We'd really need a much larger sample to draw any strong conclusions about how much difference that makes in earning classifications. Let's hope our sample size keeps growing!

[Frank Pratt]

I think the biggest weakness in the classification sytem is the fact that a fencer's rating may only reflect one good/lucky day for that fencer. An "A06" fencer who has confimed his "A" a dozen times is not rated higher than the "A06" who was a "B" until he took 8th place at an A4 event because his opponent in the round of 16 (who was winning 14-3) withdrew after breaking his ankle.

I am another good example. I "earned" my "D" in foil at an unrated tournament with 15 fencers. I had been fencing for 5 years, and the other 14 fencers had been at it for 6 moths, at the most. After earning my D, I never "reaffirmed" it. Since that day, I have never even finished high enough to get an "E" in foil (despite higher seedings I have enjoyed since that day.) My "D" expired, but I still have an "E03" from that one lucky day back in 1999, even though I have only put on a lame once in the past 4 years.