DrTodd13 Posted November 9, 2004 Report Share Posted November 9, 2004 Anybody have any thoughts on why it seems to be necessary to average 5 IMPs a board to win these 12 board ACBL IMP tourneys? DrTodd Quote Link to comment Share on other sites More sharing options...
McBruce Posted November 9, 2004 Report Share Posted November 9, 2004 Not that far out of line. A one-session IMP Pairs game will often be won with a score of about 3-4 IMPs per board. That's on 26 boards, so it is not surprising to see a 12 board game won by the pair that gets lucky most often. Quote Link to comment Share on other sites More sharing options...
hrothgar Posted November 9, 2004 Report Share Posted November 9, 2004 The basic problem is very simple: ACBL tournaments are intended as a tool to generate revenue.They certainly aren't designed as an objective measure of skill.You'll only be disappointed if you treat them as such... The most basic problem is that most tournaments feature too few boards to objectively measure skill. Coupled with this, there is no such thing as fair or balanced movements.Winning tournaments is more a matter of luck that skill. In order to win, you need to play against the "right" set of opponents. Finally, you have extreme ranges of sill levels. There's also a nasty little positive feedback loop at work. In order to win an event, you need a "big" score. In turn, this encourages players to adopt high variance strategies... Quote Link to comment Share on other sites More sharing options...
Brandal Posted November 9, 2004 Report Share Posted November 9, 2004 Anybody have any thoughts on why it seems to be necessary to average 5 IMPs a board to win these 12 board ACBL IMP tourneys? DrTodd Correct me if I'm wrong here,but the fewer tables the more "valuable" the occasional "badly sacrificed" 1400/1100 score will be? I mean,those who get "lucky" playing that particular pair on that particular board will almost have secured top 3 in one board in IMPs? I like playing imps at tables in Main,but for tourneys,what's the point playing 12 boards ending between 0 and 10 imps,which will happen if you meet the more solid pairs. My point being there is usually at least 1 "freak score" a board,maybe it reflects less in MP because you can't get more than 100%? I'll just stop here,I'm on thin ice Quote Link to comment Share on other sites More sharing options...
inquiry Posted November 9, 2004 Report Share Posted November 9, 2004 I am not going to get involved into a bid debate on how many imps one need to win a tourney.. (ACBL or otherwise). But, when these things are only 12 boards events, you bid two easly slams pick up 24 imps, opponents overbid to two hopeless slam against you another 24 imps, that is 48 in four hands. Normal bridge on teh other 8 hands were you average 1.25 imps per hand, and you are over 5 imps per borad. Nothing horribly odd about that..... luck of the draw, getting the hand people overbid against you. The last five acbl imp tourneys, this was the average imps the winner had (the imps needed to win would be lower, it would be something more than the second place pair)... 5.023.463.301.643.032.842.483.072.472.18 As you can see, the winning imp average here was only 2.94, right below what McBruce estimated would be the winning average, but a few of the low ones above are from individuals, where the scores are typically lower.. which probalby explains why the average wasn't within his estimated range. Quote Link to comment Share on other sites More sharing options...
sceptic Posted November 9, 2004 Report Share Posted November 9, 2004 Just out of curiosity, I really am out of my depth here, but a few things I find odd about ACBL tourneys. Why do the ACBL, not run longer tourneys, or is hrothgar right,they are just revenue generators When you look at the scores some of the JQKA listed people have, it would appear that they only have the J Q K or A because of the amount of tournaments they enter. as their scores are tiny it is just the quantity of them that add up to a lot of masterpoints. Also this is may cause offence but it is not intended to I am really curious thats all. Why do some experts adavanced etc etc advertise to play in ACBL if the partner pays? they seem to solicit in the partnership room waiting for some one to pay for them. (I go there to find a p but most expect others to pay) what does this say about the people that do pay for them ? are they just hoping to get a good score beacuse they expect the pard they are paying for to carry them? and what do the ones that solicit their skills get out of it in the long run? Quote Link to comment Share on other sites More sharing options...
inquiry Posted November 9, 2004 Report Share Posted November 9, 2004 This is an international field. Some people advertise to play if their partner pays for a very simple reason.. they don't have an account with any money in it. Maybe they don't ahve a credit card, maybe they don't have disposable income to spend multiple dollars a day, maybe they don't need or care about acbl masterpoints so there is little incentive for them to pay. Why 12 boards? That is a good question, but at 8 cents a board, that is a great price by any standand. I like 26 board things myself, real life or on line. And I would be willing to pay a whooping $2.50 for one of those things... (I am a big spender, hehehehe). As for AKQJ.. of course it those who have entered the most who will have the big face cards. After a while, it will the better the players from among those who have played the most. It is not an idiciation of ability so much as persistance. Ben Quote Link to comment Share on other sites More sharing options...
hrothgar Posted November 9, 2004 Report Share Posted November 9, 2004 Tournament design can actually be modeled as a sampling problem If people can provide some basic statistics it should be possible to derive the number of boards necessary to determine the top player with 95% certainty. I don't need the mean (it better be either 0 or 50%), however, I do need an accurate measure of Standard Deviation... Quote Link to comment Share on other sites More sharing options...
jikl Posted November 9, 2004 Report Share Posted November 9, 2004 Tournament design can actually be modeled as a sampling problem If people can provide some basic statistics it should be possible to derive the number of boards necessary to determine the top player with 95% certainty. I don't need the mean (it better be either 0 or 50%), however, I do need an accurate measure of Standard Deviation... The problem with this question is also related to table movements. Since we don't have real table movements as a TD to select from, it is impossible to calculate. For example, in an 8 round Indy (1 board rounds), I played with the same partner 3 times. I don't know how hard it would be to get "real" movements for TDs into BBO. I would love to run a 4 table Indy with 15 1 board rounds where everyone plays with every other person. Sean Quote Link to comment Share on other sites More sharing options...
uday Posted November 9, 2004 Report Share Posted November 9, 2004 Why do the ACBL, not run longer tourneys, or is hrothgar right,they are just revenue generators BBO runs the ACBL tournaments, using the same sort of franchise used by brick-and-mortar bridge clubs all across North America. That is, the ACBL provides a franchise -- BBO pays for it -- and regulations that each franchise must follow. The ACBL is not involved much beyond that. BBO tourneys are currently classified as ACBL club games. Why don't we run longer tourneys? I'm sure we will at some point, if there is demand. I dont know why we settled into this minimum-number-of-boards configuration, but here we are for now. ACBL tourneys on BBO are revenue generators in the sense that your job is a revenue generator. Your family, our servers, they appreciate being fed. The ACBL (ITA, LAND, SKY, etc.) tourneys contribute heavily towards keeping the rest of BBO free. Quote Link to comment Share on other sites More sharing options...
uday Posted November 9, 2004 Report Share Posted November 9, 2004 A better Indy movement is on my list. I have a movement (somewhere) from Cascade that seems clean. Now that Istanbul is behind us, I will have some time to look into this and other things that have been bugging us for a while. Quote Link to comment Share on other sites More sharing options...
jikl Posted November 9, 2004 Report Share Posted November 9, 2004 I reccomend one book, the massive tome by Hallen et al. It is the movement bible. Sean Quote Link to comment Share on other sites More sharing options...
KiwiBridge Posted November 10, 2004 Report Share Posted November 10, 2004 A true individual movement would indeed be magic. Let's stop all other requests for change so that Uday can work on this one. ;) Kiwi Quote Link to comment Share on other sites More sharing options...
jtfanclub Posted November 10, 2004 Report Share Posted November 10, 2004 There's also a nasty little positive feedback loop at work. In order to win an event, you need a "big" score. In turn, this encourages players to adopt high variance strategies... I think the issue is much simpler than that... I'd say well over half the people at the tourney are playing with a new partner. Maybe a sub, maybe somebody from the partnership desk, maybe somebody they met at the main bridge club. Now, when I play with such people (and I usually do, since I sub a lot), the results tend to be that on 10 boards out of 12, we get decent results- a +5 here, a -2 there, overall a bit above average. And on about two boards, we get a disaster -12 or so. There's a very simple reason: we haven't played together much before, we haven't ironed out the system, and we don't know each other's tendencies. We may both be too conservative, or too aggressive, and not find what the field finds. If you figure the established pairs get average on 10 boards and get 24 IMPs on the other two boards because of miscommunications by opponents, that already puts them 2 IMPs above average. Getting four of these miscommunications gets you 4 IMPs above average, and could easily with the tournament for somebody who is completely average, but has no misunderstandings. Note that increasing the number of boards isn't going to help much- it'll just double the number of swings. In fact, it may make it worse- in some long touneys, by the end it seems like over half the pairs have at least one sub. It might be interesting to play a tourney where there was no partnership desk. I'll bet it would cut the winners down by a lot. Quote Link to comment Share on other sites More sharing options...
hotShot Posted November 12, 2004 Report Share Posted November 12, 2004 I think the big average per board is a result of: 1) big differences in the playing and bidding ability of the field2) "random" partnerships without proper agreements3) tournament tactic, at imps and few boards you need big scores, so you play high risk and produce big scores often against you 4) unbalanced movement (if you play 2 or 3 boards / round and get the right victims you can produce many imp's) Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.