Match Length Proposal

[Ulti...@aol.com]

The goal of the organizers (the USBF) has always been to run an event that gives the team that plays the best, the best chance of winning, in a given amount of time allotted to run the event. I think the players in USBCs agree with that goal.

 

Reducing the number of boards played per day does not achieve this goal. IMO. It just wastes time.

 

The more boards played, the better chance that the team that plays the best will win.

But there are diminishing returns for each additional board played in a match. If there was a scale that showed probability of the team that plays the best winning = and the team that plays the worst losing, it would look something like our VP scale -- each additional board played is meaningful, but less meaningful than the previous one. At some point we must say "enough boards is enough boards" for every match.

 

Matches do not have to be the same length. Long ago we concluded that, in allotting the overall number of days to run the event, we should play fewer boards in early stages = more boards in late matches.

 

Only recently have the organizers trended toward playing the same number of boards in all K/O matches. This is a function of logistics: it is easier to run an event that does not end in the middle of the day. But note than ending (and starting) a match in the middle of a day is possible, even if there is some inconvenience involved. We can play some 1.5 day matches if we want to allocate our time that way. 

 

Mike

[Jeff Aker]

Perhaps it would help to formalize the aspects of Mike's email by establishing a priority among goals. Number one is what's in his first paragraph, to give the team that plays the best the best chance of winning. Number two is presumably to limit the length of the event. All the other factors are merely means of reaching these goals. I'm not sure how having shorter matches early in the event helps meet the first goal, but it may be necessary in meeting the second. As noted, midday starts and finishes are somewhat undesirable, but not really important compared to the other factors.

Jeff

Sent from my iPad

--
You received this message because you are subscribed to the Google Groups "International Team Trials Committee" group.
To unsubscribe from this group and stop receiving emails from it, send an email to usbf-ittc+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

The information contained in this electronic message is confidential, for information and/or discussion purposes only and does not constitute advice about, or an offer to sell or the solicitation of an offer to purchase, any security, investment product or service. Offers of securities may only be made by means of delivery of an approved confidential offering memorandum or prospectus, may be legally privileged and confidential under applicable law, and are intended only for the use of the individual or entity named above. We do not, and will not, effect or attempt to effect transactions in securities, or render personalized advice for compensation, through this email. We make no representation or warranty with respect to the accuracy or completeness of this material, nor are we obligated to update any information contained herein. Certain information has been obtained from various third party sources believed to be reliable but we cannot guarantee its accuracy or completeness. Our investment products involve risk and no assurance can be given that your investment objectives will be achieved. Past results are not necessarily indicative of future results. Any review, retransmission, dissemination, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. Email transmissions are not secure, and we accept no liability for errors in transmission, delayed transmission, or other transmission-related issues. This message may contain confidential, proprietary or legally privileged information. Neither confidentiality nor any privilege is intended to be waived or lost by any error in transmission.

[Gavin Wolpert]

I think that recent years have proved that the field is deep enough that the quarterfinal should not have have shortened matches.   The seeding point structure is flawed (no suggestions for improvement, IMO the flaw stems from the limited number of events available to include in calculation).   If the QF is shorter it adds an element of randomness which seems contrary to the goals of the event.  

[Robb Gordon]

What Michael wrote is exactly along my line of thinking as well. I am hard-pressed to imagine that a 96 board match would be significantly less "formly" than 120.

But the longer these events last the fewer people are able to play given other demands of life. Some people even still work or go to school.

The players who have responded thus far are basically top-level full time players to whom the extra days don't matter.

I wonder what the lower seeded players who are more likely to have time conflicts think.

I wonder (not being facetious) whether it matters what they think.

Robb Gordon 

--

[Ulti...@aol.com]

[Danny Sprung]

A very large +1 to to Robb here.  14 days for a trial?  To then qualify for a 13 day event in the same calendar year?  The message we are sending is clear; working people, students and those not particularly well off need not apply.

More and more, these events are set up to benefit the pros and sponsors; while they represent most of the best of today's American players, I wonder where a Norman Kay type would fit in.

In addition, when an event gets that long, stamina is being tested to an even greater extent than bridge skill (just my opinion there).  

Alas, what the lower seeds want probably doesn't matter.  Maybe it shouldn't.  But, then again, look at the big upset by the Ganzer team, and the 5 handed Gupta team lost by a mere 4 IMP's to the eventual winners.

Danny

[geoff hampson]

I do tend to like the longer duration of the trials, mostly because it is the best event that I get to play each year and I want it to be the fairest that it can be.  That said, I think that 14 days to qualify via double elimination is too much.  Cutting to 90 board matches would shorten the event to a more manageable duration and probably not reduce the odds of selecting the strongest team (much).  It is difficult to contend that you were unlucky to lose in TWO 90 board matches.  Starting a match half way through the day is meaningless (to me) and need not be a big concern.  I do think that a move to "Bridge Standard Time" would be worth consideration. (ie starting earlier on the west and later on the east etc)

   

[Bruce Rogoff]

I’m in general agreement with the thoughts expressed by Danny and Robb.  It seems to me we should be able to produce two topnotch teams in less time than it takes to stage the entire summer Olympics.

 

Will 120-board matches be more formful than 90 or 96 boards?  One would have to assume yes (slightly), though we’ve certainly had a few “surprise” teams come through the trials gauntlet anyway.  However, if you’re a top team that was “unlucky” enough to lose twice, perhaps your squad wasn’t one of the two best after all.

 

Does it bother anybody that 14 straight days of 60 boards is more grueling than the event we’re all trying to qualify for?  Some teams might suffer more than others from that format, but would perhaps do better playing three sets of 16’s each day…same as the BB.  Having played in a few international events, I’m positive there’s a difference.  Wouldn’t we want to select the teams that play the best in those conditions?  Our current format often produces some shoddy bridge (IMHO) as teams enter the second week of play.  Seems like we’re selecting our marathon team based on times in a 35-mile run.

 

Anyway, if more are in favor of the 14-day format I would advocate the more relaxed schedule, reducing the matches to 96 boards.  We could easily reduce to 11 days by playing 90-board matches throughout, after a 2-day RR.  My personal preference is a 12-day format:  3-day RR reducing to 16 teams, then 90-board matches throughout.  Losing USA1 finalist plays USA2 semifinal immediately.  If there are fewer than 8 R16 losers, the top non-qualifiers from the RR play a KO or RR to fill out the start of USA2.

 

Bruce

---

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2013.0.3495 / Virus Database: 4311/9764 - Release Date: 05/13/15
Internal Virus Database is out of date.

[Jeff Aker]

To all ITTC members,

 

This discussion has covered a lot of issues, and in my role as Chairman  I’d like to focus on that the number of boards played per day. The basic proposals have been 48 and 60 (with a few stray votes for 54 and 64). My sense of the discussion has been that the consensus is to stay with the current 60 boards per day. What is less clear is whether we should attempt to shorten the event by playing shorter matches in the rounds of 16 and 8 and I’d like to see further discussion on that issue.

We’re agreed (although I’m not sure we have actual proof) that the longer matches will tend to run more true to form and are more apt to select the best team, but shortening the event will, allow more teams to enter, although those are teams which are perhaps less apt to win. Presumably, however, we want a large entry, which is why we went to the open format in the first place.

 

Jeff

 

Jeff Aker

ja...@gargoylegroup.com

[Ulti...@aol.com]

It appears (correct me if I am wrong) that those who have not won a USBC favor shorter matches and those who have won a USBC favor longer matches. That speaks volumes.

 

Do shortening matches -decrease, remain the same or increase- the probability of sending the team that plays the best in the USBC to the WBF World Championship"?

 

I previously wrote:

The goal of the organizers (the USBF) has always been to run an event that gives the team that plays the best, the best chance of winning, in a given amount of time allotted to run the event. I think the players in USBCs agree with that goal.

 

I should add, the organizers should create CoCs that are fair to all, but give the team that is the best, the best chance to win.

 

That's why we have long matches rather than short ones; and that's why we have PPs for feeder events (measuring NABC performance, and some masterpoints) and PPs for past USBCs.

 

Mike

[Josh Sher]

Mike, I suppose in 1965 you would have argued against the voting rights act on the same grounds? That the legislation was supported by the folks who had been unable to win elections previously, and opposed by the folks who had been winning? Any set of rules favors some people, and it’s hardly a coincidence that the folks that benefit from the current rules tend to favor it and the folks that don’t benefit tend to oppose it. What we need to evaluate is what is the fairest set of rules, and drop this “you guys have not won anything, so your point of view is not valid” bullshit.”.

 

I think the following statements are clear:

 

A.      Longer matches are more likely to be won by the better team (at least in terms of how well they were playing now) than shorter matches.

B.      The benefit of each additional board has positive but decreasing benefit toward the end of determining the better team

C.      Full time bridge players, especially those who are getting paid, are much better able to play in a long event than those who have non-bridge jobs

 

Speaking for myself, 10 years ago I was in a job where I had 10 days vacation per year, in a company that was not flexible about leave without pay. Had I played in the team trials then, and actually won, I would have faced the choice between competing the world championships and quitting my job or keeping my job. Extremely long events create barriers to entry, and leaves some people who might be good enough to compete and maybe even to win unable to compete.

 

I think it is fair to stay that almost all the best players currently, are full time players. But it does not have to be so. And possibly, part of the reason it currently is so, is that non full time players often do not have sufficient free time to play in many of the best events…

 

I understand that there are competing values. The value of selecting the best team, is not identical to the value of selecting the team that played best in this event, if a potentially great team was dissuaded from even entering. I don’t think its 100% clear how to balance the two values. (I do think that shortening the number of boards played per day doesn’t  help with either value, and that’s how this thread started.) I think we want to have matches that are long as possible, without chasing away good players that happen to have a life and responsibilities outside bridge.

Josh Sher 

--------
The best things in life are nearest: Breath in your nostrils,
light in your eyes, flowers at your feet, duties at your
hand, the path of right just before you. Then do not grasp
at the stars, but do life's plain, common work as it comes,
certain that daily duties and daily bread are the sweetest
things in life. - Robert Louis Stevenson



Ah, but a man's reach should exceed his grasp,
or what's a heaven for? - Robert Browning

[Danny Sprung]

Mike:

I have not won a USBC.  I don't even play in them currently.  I am not in favor of shorter matches, just a shorter event.  14 days is too long.  

My two suggestions would leave the events in the more reasonable 10 day time frame.  Eliminating the repechage by either:

1. Granting the Olympiad trials winners the BB berth instead (4 identical USBCs every cycle)

2. Running USA1 and USA2 in parallel in the Olympiad+1 year.

This last proposal was fairly roundly rejected (seeding concerns, mostly).  But it seems very sound to me.  Look at the losing finalists in all of the past USA1 matches.  Have their been any 'fluke' teams?  I'd say no.  Is it possible that there would be a fluke team if that format was used in the future?  Perhaps.  

As I understand it, the repechage was not especially well balanced for seeding either.

Danny

[eric greco]

As a side note.. Is there any possibility of the U.S. sending two teams to the olympiad?  Things have changed over time and we now have many players who previously represented other countries. Thus leaving the the U.S. with far more expert players than any other one country in the world, than any other time in the past.

---

Date: Tue, 26 May 2015 14:51:34 -0400

Subject: Re: Match Length Proposal

From: danny...@gmail.com
To: Ulti...@aol.com
CC: usbf...@googlegroups.com

[Uday Ivatury]

FWIW, from the peanut gallery:  Three things.  

One:

It doesn't matter so much if the Diamond-Fleisher match is slightly less true to form.

It does matter if the Uday-Fleisher match is less true to form.   

We should much prefer to risk the "wrong" team advancing in a later round than in earlier rounds. We don't want to lose a real contender bec. of early randomness, but later, they're all contenders anyway.

So given a choice of whether to make earlier rounds longer vs later rounds longer, I think it is smarter to make the earlier rounds longer.  Of course, once a match is "long enough" (whatever that is) it doesn't matter.

Two: 

I hear this over and over again:  

"sending the team that plays the best in the USBC"

vs

"sending the team that plays the best through the year"

Our actual practice of handicaps via seeding grants an advantage to teams who have played best through the year. 

Is the actual goal easily expressed by anyone in the know? 

Three: 

We seem to be of two minds about whether we actually want a large field vs a "large-enough" field.  The current conditions ( expense, duration, schedule-uncertainty) make it unlikely we'll ever have a large field.

The recent excellent Chicago trial was "large enough" but if we want a much larger field, we might have to accept an online component for some of the earlier, less-meaningful rounds.  One way to do this might be to invite 15 teams (or limit to 15 teams via USBF seeding points) and hold a series of scrimmages online or at clubs for the 16th position.  This would allow the full duration of the physical trial to focus on the KOs, saving time which can be repurposed towards either longer KO matches or a shorter event.   

[Alan Frank]

Response from the peanut gallery:

I'm not sure I agree with the premise nor do I believe the conclusion
follows from the premise. Ignoring repechage for the moment, if the top
two teams are in the finals and the inferior team wins, we send the
second-best team to the BB. If the inferior team gets luck and beats the
#1 team in the R16, then the #2 team may still win the event. And if any
other top team loses early, the #1 seed may still win. So not sure it
makes much difference in what round an inferior team wins.

Secondly, the skill difference between the #1 seed and the #16 seed, even
if not the best and sixteenth-best teams, is usually considerable. With
high likelihood, a one-day match should be enough to overcome the luck
factor. (Not necessarily with #8 vs #9, but it is not a big deal [except
to the players!] who wins that, both because they are likely closer in
ability and the winner is likely to lose in the R8 or R4 anyhow.) OTOH,
#1 and #2 may be fairly close in ability and will need a longer match to
give an adequate chance of form being true.

--Alan

P.S. Trivia: without looking it up, what's the origin of the term "peanut
gallery?"