PDA

View Full Version : Upsets



pax
28-12-2005, 11:30 PM
Rd 1:
Moulthun beat Wohl, Obst beat Goldenberg, Moylan beat Canfell, Song drew with Xie (with black), Caoili drew with Johansen (also with black). That's a very exciting start to a 15 board swiss (in contrast to the Junior U18 where every game went to rating).

Lets see - that's a 3.5/8 start to players rated under 2150 ACF (against average opposition around 2350) - not too shabby!!

PHAT
29-12-2005, 12:05 AM
What was the competitive index on this particular junk round, starter ;)

Phil Bourke
29-12-2005, 12:27 AM
Not surprised about Song gaining a draw with Xie, he did beat me at Coffs this year :)
Seriously, he is a talented junior, and it will be interesting to see how he goes later today against Johansen.
Johansen's draw was a good dogfight, and congrats to Caoili for her effort.
The top two seeds had hard fights for the full point as well, it augers well for the rest of the tourney.
One good thing about a 7 round swiss, puts the pressure on the players to get the wins, no weak arsed 10 move draws :)

Garvinator
29-12-2005, 12:53 AM
One good thing about a 7 round swiss, puts the pressure on the players to get the wins, no weak arsed 10 move draws :)
except that this tournament is a 11 round swiss.

Phil Bourke
29-12-2005, 04:53 AM
True :) More speed, less haste :)
Should still keep the players trying for the wins though, as a couple of draws would see you a point off the pace and maybe not getting an opportunity to peg the leader/s back.

ursogr8
29-12-2005, 06:24 AM
What was the competitive index on this particular junk round, starter ;)

Competitve index for this round (1) of the Championship was 278.8 (using the first rating quoted).

This is essentially a DIVISIONALISED tournament and, as such, no junk round would be expected. Hence the UPSETS should come as no surprise.

starter

Kevin Bonham
29-12-2005, 02:21 PM
Competitve index for this round (1) of the Championship was 278.8 (using the first rating quoted).

This is essentially a DIVISIONALISED tournament and, as such, no junk round would be expected. Hence the UPSETS should come as no surprise.

starter

But four points of upsets from 15 games for an average rating difference of 278.8 is surprising.

278.8 is a mean performance expectancy of about .17 for the lower rated player per game, or 2.55 for the 15 lower rated players combined.

Of course the difference between 2.55/15 and 4/15 is not statistically meaningful. However it is notable that three of those considered not worthy of a place by various juniorphobes and sundry restrictivists and antisubjectivists on the board were off the mark in round 1!

ursogr8
29-12-2005, 03:32 PM
But four points of upsets from 15 games for an average rating difference of 278.8 is surprising.

Barry finished the thread discussion on the criteria to declare a result a 'surprise'. I will leave him to rule on your measure (if he wishes, of course).


278.8 is a mean performance expectancy of about .17 for the lower rated player per game, or 2.55 for the 15 lower rated players combined.
Agreed.


Of course the difference between 2.55/15 and 4/15 is not statistically meaningful.
Agreed.

However it is notable that three of those considered not worthy of a place by various juniorphobes and sundry restrictivists and antisubjectivists on the board were off the mark in round 1!

I don't know what any of these words mean < juniorphobes and sundry restrictivists and antisubjectivists>, but I don't think any included me. ;)

regards
starter

ps To MODs, my original response to Matt's question posted in this thread because that is where the request was posted.

Kevin Bonham
29-12-2005, 04:48 PM
I don't know what any of these words mean < juniorphobes and sundry restrictivists and antisubjectivists>, but I don't think any included me. ;)

No, they didn't.

ursogr8
30-12-2005, 07:56 AM
Competitve indices for the rounds (1, 2) of the Championship were 278.8, 196 .


starter

Frank Walker
30-12-2005, 08:28 AM
Yesterday, I saw a kid drawing against a WFM.
He was up a pawn though, I think...

Frank Walker
30-12-2005, 08:29 AM
Devraj B. vs Shannon O.

pax
30-12-2005, 08:58 AM
Fewer upsets in the Championship in Rd 2.

Jay Lakner beating Kegless was the biggie (have a look at the neat finish). James Obst beating Arianne is a bigger upset on paper, but actually didn't surprise me so much. The GM draws on 1 and 3 definitely don't count as upsets!


[White "Canfell, Greg"]
[Black "Lakner, Jay"]


1.e4 c5 2.Nc3 Nc6 3.g3 g6 4.Bg2 Bg7 5.d3 Rb8 6.f4 b5 7.Nf3 d6 8.O-O Bg4 9.h3 Bxf3 10.Qxf3 Nd4 11.Qd1 e6 12.Ne2 Ne7 13.c3 Nxe2+ 14.Qxe2 O-O 15.Bd2 b4 16.Be1 bxc3 17.bxc3 Qb6 18.Rf2 Qa5 19.Qc2 Rb6 20.g4 Qa3 21.f5 Bf6 22.Qd2 exf5 23.exf5 g5 24.Re2 Rfb8 25.Rc1 Rb1 26.Rc2 h6 27.Kh2 Kg7 28.Bg3 Qa6 29.Re3 h5 30.Qe2 h4 31.Bf2 d5 32.Bf3 Qd6+ 33.Kg2 d4 34.cxd4 Bxd4 35.Rxe7 Qg3+

auriga
30-12-2005, 09:59 AM
Fewer upsets in the Championship in Rd 2.
[White "Canfell, Greg"]
[Black "Lakner, Jay"]


neat finish!

bobby1972
30-12-2005, 10:31 AM
Does Any One Have The Rest Of The Ly-wohl Game

Garvinator
30-12-2005, 11:49 AM
Does Any One Have The Rest Of The Ly-wohl Game
you can download it from the pgn file from the aus champ website: www.*******2006.com

pax
30-12-2005, 11:58 AM
you can download it from the pgn file from the aus champ website: www.*******2006.com

The pgn only goes to move 18 {moves unreadable}. I hope they can reconstruct it at some point.

Trizza
30-12-2005, 12:35 PM
The pgn only goes to move 18 {moves unreadable}. I hope they can reconstruct it at some point.

The full game is in the round 2 bulletin which can now be downloaded.

Some nice tactics in the wins by Obst and Lakner.

ursogr8
30-12-2005, 01:17 PM
Competitve indices for the rounds (1, 2, 3) of the Championship were 279, 196, 160 .


starter

Careth
30-12-2005, 02:15 PM
The Sam Grigg- Angela Song match in the Juniors ended a draw, while Grigg is over 600 points less in rating than Angela! Congrats Sam! :clap:

Kevin Bonham
30-12-2005, 03:33 PM
Fewer upsets in the Championship in Rd 2.

Jay Lakner beating Kegless was the biggie (have a look at the neat finish). James Obst beating Arianne is a bigger upset on paper, but actually didn't surprise me so much. The GM draws on 1 and 3 definitely don't count as upsets!

Both the upset wins were again by U2150s who according to some on this BB do not deserve a place, and also there was Pyke's draw with Smirnov, so another good round for the U2150 brigade.

DoroPhil
30-12-2005, 03:43 PM
...who according to some on this BB do not deserve a place...

"25% rule" is one of the ways to see who did not deserve a place in the tournament afterall.

So, those who are on 1 point or less after 5 rounds and/ or on 2.5 points or less after 11 rounds shouldn't have been allowed to play.

pax
30-12-2005, 05:51 PM
"25% rule" is one of the ways to see who did not deserve a place in the tournament afterall.

So, those who are on 1 point or less after 5 rounds and/ or on 2.5 points or less after 11 rounds shouldn't have been allowed to play.

That's rubbish, of course since virtually every Swiss will have players who score under 25%. It is perfectly possible for a player who totally deserves their place to have a bad tournament.

Kevin Bonham
30-12-2005, 06:38 PM
Also if you had sufficiently many players who supposedly didn't deserve a place admitted then some of those would score 3 just by beating the others. That is why I think the only useful measure is the performances of those who supposedly shouldn't be in against those who supposedly should.

I think the state champions should be in irrespective, but it is useful to look at the performances of those in under the improving junior or equivalent proficiency rules and see how they go when not playing each other. It is also useful to look at the results of those state champs who would have otherwise needed the improving junior or equivalent proficiency rules to get in.

Kevin Bonham
30-12-2005, 06:46 PM
The Sam Grigg- Angela Song match in the Juniors ended a draw, while Grigg is over 600 points less in rating than Angela! Congrats Sam! :clap:

Another draw with a similar rating difference (of interest to me because the lower-rated player is a Tasmanian and this would have to be her best result so far.)

5 1205 Eustace, Sophie (1) - Horton, Nina (1) 466

Anything could have happened at the end of this because Nina missed 50...Rg1+! winning then Sophie missed 52.Rf4+ likewise.

Kevin Bonham
30-12-2005, 09:28 PM
Yet again the U2150 brigade did well today with wins for Pyke vs Caoili and Moylan vs Humphrey. The only other upset win on ratings was Smerdon over Schmaltz.

ursogr8
31-12-2005, 06:51 AM
Competitve indices for the rounds (1, 2, 3, 4) of the Championship were 279, 196, 160, 139 .


starter

Oepty
31-12-2005, 11:16 AM
Another draw with a similar rating difference (of interest to me because the lower-rated player is a Tasmanian and this would have to be her best result so far.)

5 1205 Eustace, Sophie (1) - Horton, Nina (1) 466

Anything could have happened at the end of this because Nina missed 50...Rg1+! winning then Sophie missed 52.Rf4+ likewise.

Kevin. I have only looked at it on the screen but isn't white stilling winning after 52. Rd6+. Can't white bring the queen closer while checking and then eventually work the rook into play so there will be a checkmate or win of material.
Scott

Kevin Bonham
31-12-2005, 12:33 PM
Kevin. I have only looked at it on the screen but isn't white stilling winning after 52. Rd6+. Can't white bring the queen closer while checking and then eventually work the rook into play so there will be a checkmate or win of material.

It appears not. If White tries this, Black can at the appropriate moment play ...Kb3 and then block White queen checks on the diagonal with ...c4.

Re the "upsets" in the Champs, I notice that the following are currently the top-rated females on the ACF list:

2293 Caoili
2184 Sorokina
2148 Koshnitsky
2128 Berezina
2093 Moylan
2072 Nguyen
2071 Eriksson
(etc)

As many of the top female players are relatively inactive (the above seven played a whole two rated games between them in the December 05 period), their ratings may not be all that accurate and so upsets involving top female players (either winning or losing against rating) are to be expected.

ursogr8
03-01-2006, 07:35 AM
What was the competitive index on this particular junk round, starter ;)


hi Matt

All this talk of upsets.
He is a small challenge for you...>>>
Find the first upset in the
round 1 cross-table (http://www.*******2006.com/results.html) of the U18 Juniors currently in play.


And if you solve the first one, before your atomizer runs out of water, then have a look at the cross-table for the U18 Girls event and tell me if it is sorted in rating order.

regards
starter

Garvinator
03-01-2006, 08:48 AM
then have a look at the cross-table for the U18 Girls event and tell me if it is sorted in rating order.
from a quick look, what it seems to be doing in each score group is placing those who have white higher up than those who have black in each round. Well that is what I saw from looking at round 6 pairings and crosstable.

pax
03-01-2006, 09:06 AM
from a quick look, what it seems to be doing in each score group is placing those who have white higher up than those who have black in each round. Well that is what I saw from looking at round 6 pairings and crosstable.

No, the crosstables and standings are sorted in tie-break order, which in all events is:
-sum of progressive scores
-buchholz
-median buchholz

If all tie-breaks are equal, it is sorted in starting rank order.

Garvinator
03-01-2006, 09:11 AM
No, the crosstables and standings are sorted in tie-break order, which in all events is:
-sum of progressive scores
-buchholz
-median buchholz

If all tie-breaks are equal, it is sorted in starting rank order.
as i said, a quick check. Must just have been that for the first few boards in round 6, the white player has the higher tiebreak in each of the boards.

ursogr8
03-01-2006, 11:50 AM
from a quick look, what it seems to be doing in each score group is placing those who have white higher up than those who have black in each round. Well that is what I saw from looking at round 6 pairings and crosstable.

hi raggy''

HNY btw

Now listen my sun-affected mate; my post was directed at Matt. This was deliberate since his mind works on a different plane from other posters (a recent episode on 'Where is "sweeneys deleted posts thread?' indicated that I understand the workings of his mind far better than the big 2). If for one moment I thought you were going to respond to my post I would have been less oblique.


Of course the cross-table was not sorted in rating order.
Put 2 and 2 together and you will see I was highlighting the terrific performance of Mexicans, Sally Yu and Susan Sheng.

Stay cool

regards
starter

ursogr8
03-01-2006, 12:03 PM
Competitive indices for the rounds (1, 2, 3, 4, 5) of the Championship were 279, 196, 160, 139, 128 .


starter

Garvinator
03-01-2006, 12:11 PM
hi raggy''

HNY btw

Now listen my sun-affected mate;
in fact, i havent been that sun affected as I have been in the air conditioned tournament hall during the afternoons:)

ursogr8
03-01-2006, 03:40 PM
in fact, i havent been that sun affected as I have been in the air conditioned tournament hall during the afternoons:)

I guess I took my line on your weather from
http://chesschat.org/showpost.php?p=82085&postcount=1

ursogr8
03-01-2006, 09:52 PM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6) of the Championship were 279, 196, 160, 139, 128, 113 .


starter

ursogr8
06-01-2006, 08:03 AM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6, 7, 8 ) of the Championship were 279, 196, 160, 139, 128, 113, 90, 128 .


starter

ursogr8
06-01-2006, 09:21 PM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6,7, 8, 9 ) of the Championship were 279, 196, 160, 139, 128, 113, 90, 128, 61 .


starter

ursogr8
07-01-2006, 09:51 PM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6,7, 8, 9, 10 ) of the Championship
were 279, 196, 160, 139, 128, 113, 90, 128, 61, 113 .


starter

ursogr8
09-01-2006, 05:51 AM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 ) of the Championship
were 279, 196, 160, 139, 128, 113, 90, 128, 61, 113, 164 .


starter

Garvinator
09-01-2006, 10:16 AM
and now that you have measured it, how are you going to improve it?

Watto
09-01-2006, 10:41 AM
Competitive indices for the rounds (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 ) of the Championship
were 279, 196, 160, 139, 128, 113, 90, 128, 61, 113, 164 .


starter
I keep reading these new posts thinking I’m going to find about some new upset. But no, just another series of numbers… ! lol.

Doesn’t worry me, just amusing. It’s like this vegetarian restaurant I know- they have all sorts of ‘meat’ items on the menu. No matter how often someone tells you it’s only pretend chicken, lamb, fish etc you can’t help the feeling of anticipatory excitement (if you’re a meat eater)…

ursogr8
09-01-2006, 11:16 AM
I keep reading these new posts thinking I’m going to find about some new upset. But no, just another series of numbers… ! lol.

Doesn’t worry me, just amusing. It’s like this vegetarian restaurant I know- they have all sorts of ‘meat’ items on the menu. No matter how often someone tells you it’s only pretend chicken, lamb, fish etc you can’t help the feeling of anticipatory excitement (if you’re a meat eater)…

hi Watto

There has been previous debate, but no consensus, as to what defines an upset. Some examples of criteria discussed are 'a win where there was 400 rating points difference, etc'.

What would you define as an upset?

regards
starter

pax
09-01-2006, 11:38 AM
hi Watto

There has been previous debate, but no consensus, as to what defines an upset. Some examples of criteria discussed are 'a win where there was 400 rating points difference, etc'.


If that's the definition, then I imagine Cedric antolis in the U12 is very, very upset. He is rated over 1500, and has lost a string of games to triple digit rated players.

Watto
09-01-2006, 12:03 PM
hi Watto

There has been previous debate, but no consensus, as to what defines an upset. Some examples of criteria discussed are 'a win where there was 400 rating points difference, etc'.

What would you define as an upset?

regards
starter

Hi starter

Not sure why you’d ask me that question but this is my rather unauthoritative answer.

‘Upset’ isn’t a scientific term… it’s simply a result in which the clear favourite to win (if there is one) gets beaten. Hence the lack of consensus (although thinking about it, even if it was a scientific term there’d be a fair chance we’d get no consensus).

Of course people like to try to measure these things so in chess we turn to ratings. With the sub 1500ers whose ratings are often unreliable for whatever reason, players rated even 500 to 600 points below their opponent not infrequently come up trumps. Still, a 400 rating point difference would probably be fair for this group. Probably also for the 1600 to 2000ers, not that I’ve given this much thought. :)

With the top players whose ratings tend to be much more reliable (improving juniors and improving adults aside) and where 100 points can signify a real difference in strength, an upset could often involve quite a bit less than a 400 rating points difference. If Canfell had beaten Chandler it would have been a glorious upset in my book but of course there’s less than 200 rating points between them.

ursogr8
09-01-2006, 01:42 PM
and now that you have measured it, how are you going to improve it?
hi ''r''a''gg''y

I missed this post of yours in all the excitment of a new poster showing (indirect) interest in competitive indicies...well at least, the upsets part of the main-line thread discussion (elsewhere). ;)

First, are you sure you have not fallen for Kaitlin's misinterpretaion...see 'stitches' thread.

Second, I would have thought the CIs here demonstrate already that the Championships competitiveness is admirably improved by Divisionalisation.

regards, and keep posting
starter

ursogr8
09-01-2006, 01:44 PM
If that's the definition, then I imagine Cedric antolis in the U12 is very, very upset. He is rated over 1500, and has lost a string of games to triple digit rated players.

Given that 400 is close to the previous upper-bound of the envelope where we discussed consensus, then, sure, Cedric probably is. :( ;)

starter

ursogr8
09-01-2006, 01:47 PM
Hi starter

Not sure why you’d ask me that question but this is my rather unauthoritative answer.

‘Upset’ isn’t a scientific term… it’s simply a result in which the clear favourite to win (if there is one) gets beaten. Hence the lack of consensus (although thinking about it, even if it was a scientific term there’d be a fair chance we’d get no consensus).

Of course people like to try to measure these things so in chess we turn to ratings. With the sub 1500ers whose ratings are often unreliable for whatever reason, players rated even 500 to 600 points below their opponent not infrequently come up trumps. Still, a 400 rating point difference would probably be fair for this group. Probably also for the 1600 to 2000ers, not that I’ve given this much thought. :)

With the top players whose ratings tend to be much more reliable (improving juniors and improving adults aside) and where 100 points can signify a real difference in strength, an upset could often involve quite a bit less than a 400 rating points difference. If Canfell had beaten Chandler it would have been a glorious upset in my book but of course there’s less than 200 rating points between them.

Very erudite of you Watto.

And, your post makes me think of two American Presidents.
1 The one who wanted one-armed economists.
2 The second as in 'beating around the bush'. ;)

regards
starter

Watto
09-01-2006, 03:16 PM
Very erudite of you Watto.

And, your post makes me think of two American Presidents.
1 The one who wanted one-armed economists.
2 The second as in 'beating around the bush'. ;)

regards
starter

Hi starter. So I’m guessing I’m Truman's two-armed economist... and an ‘erudite’ George Bush? LOL.
There’s no straightforward answer about what constitutes an upset… to spend too much time working out strict measures for that seems a bit pointless to me.

Cheers
Watto
p.s. on the other hand, I'd just like to take this opportunity to say that measures have their uses and are to be greatly encouraged... ;-)