Join Bridge Winners
All comments by David Jackson
You are ignoring the author of this comment. Click to temporarily show the comment.
Addendum: Myrtle Bennett continued to play Bridge in local clubs and, of course, the murder was never mentioned in her presence. Many years later a young partner of hers who was unaware of the history overbid the dummy and as he put down his hand he announced. ‘Partner, when you see what I have bid game on you are going to shoot me’. Myrtle had the good sense to faint.
Sept. 11, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Bernard is an excellent ‘after-dinner’ speaker and a highlight of those Senior Camrose (Teltscher Trophys) that Bernard has played in or attended is his speech at the meal which concludes the 3-day event. He usually then introduced ‘His Serene Highness’ who himself was no slouch when he got the microphone in his hand.
Aug. 21, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Its Jan KAMRAS
June 9, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
For some examples, to say that the errors relative to reality from a DD analysis of thousands of hands either in favour of or against the declarer roughly cancel each other MAY be approximately true. So in those cases, fortuitously, the results may be similar to reality. For example, how does a random flat 15 count do together with a random flat 10 count. I can possibly believe that a DD analysis may give a similar answer to what we think is the reality.

However ask the question how does a particular type of flat 15 count (say top heavy in Aces and/or Kings) do against a similar flat 10 count and I don't expect the DD analysis to be close to the reality. Similarly for two flat hands that are top heavy with the lower honour cards, DD and reality will be even further out.

So the real question is what are the assumptions needed to even think that a DD analysis of tens of thousands hands in any situation will come close to approximating the reality? Can somebody list them? I certainly don't know what they are.
June 4, 2018
David Jackson edited this comment June 5, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Averages are tricky to work with. To say as others in this thread have said that a DD simulation sometimes favours declarer and sometimes favours the defenders BUT on average the errors balance each other out and the results reflect the reality is an untrue and childlike statement to make in many cases.

Take two 4333 hands with all the aces and 3 of the kings in any combination but no other honours including Tens. A 25 count in old money. 3nt will be beaten DD nearly always but in reality even against good defenders will make 30-40% of the time and over 50% against me. At the other extreme the Michael Rosenberg example that needs 3-two way finesses to make is 100% DD no matter how the other 26 cards are distributed but nearer 13% in reality.

The only value of the DD result which is never wrong that we see on our hand records is that it occasionally provides an interesting exercise to figure out how such a fantastic result could be achieved
June 3, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
I met Eric at the 1980 Olympiad in Valkenburg and we had a couple of meals together. We walked to a nearby stadium where Ice Hockey was played and although there was no game scheduled we went inside and looked it over. He told me had played Ice Hockey in his youth and loved that game too.
May 25, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Such a rule or judgement call by the SC that a temporary sub would not be likely to improve the team seems common sense to me but that is not a commodity which International Selection committees seem to posses in large quantities. I assume that there have been examples in the past when a sub was required in International Trials. What were the histories like from those past cases.?
March 29, 2018
David Jackson edited this comment March 29, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
A temporary substitute should not be of a standard better than the player being replaced. This judgement is made by those authorised to allow the temporary substitution and they have every right to err on the side of caution. However, having made that judgement, it should not be reassessed given the results obtained. For the organisers to suggest after the event that the substitution improved, rather than lessened, a team's chances is an admission that the organisers themselves were incompetent
March 29, 2018
You are ignoring the author of this comment. Click to temporarily show the comment.
Averages can be misleading quantities to work with. Assuming that because both Declarer and Defenders play DD that these two unrealistic assumptions will always roughly cancel each other out in the long run is wrong.

Consider two 4/3/3/3 hands with no 4/4 fit which contain all the honour cards including Tens BUT without any Ace. So just 24 points. A poor 3NT contract and no doubt one we would like to avoid. But just how bad a 3NT is it? Clearly there are enough solid tricks to be set up if declarer only had enough time. But DD the defence will never give declarer time and simulating a 1000 such hands, the DD analysis will tell us that very few of the 1000 hands will be making 3NT. However, in reality, if the defenders first lead is in one of declarers 4/3 fits then a tempo has been lost and 3NT is much more likely to make. The difference between the simulation (maybe less than 10%) and the reality (around 30% perhaps) for these 1000 hands is huge. The simulation just doesn’t approximate to the reality in this case because it is DD defence that is the key, not DD declarer play. They don’t cancel each other out in this example.
Dec. 31, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Rainer Hermann says ‘However, over many random deals the average double dummy result will come very close to the average actual table result for most contracts.’
And where did Rainer find that piece of information? With generalisations like that ‘Fake news’ seems almost trustworthy.
DD results mean exactly what they say and until we have a reliable automated method of estimating the real world likelihood that an individual contract will succeed we just cannot assume DD result over a large number of hands will average itself out in any particular situation to anywhere near the reality.
Dec. 30, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
xx
Dec. 30, 2017
David Jackson edited this comment Dec. 30, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I am in the camp that is very dubious of this type of DD simulation. What does it mean that say 92% of 1000 hands make when declarer is always assumed to finesse a queen the right way and drop singleton honours and defenders always lead the right suit and card against NT? I want to see an expert consensus on how good a particular contract is in the real world not a DD analysis. As far as I know this consensus does not yet exist for even a relatively simple hand. If a 3NT is only always down on a particular lead, then in reality it may be a big favourite to make rather than zero percent
Dec. 30, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Not correct Nedju. ‘Historic’ cheating can be detected now in much the same way as an old murder case can be solved, because the DNA from way back can only now be properly analysed. What might have appeared inconclusive evidence 6-10 years ago - getting singleton/doubleton ruffing situations right - hitting partners strong suit so frequently etc., can now be explained in the light of the recent video evidence. The evidence of this historic cheating is now overwhelming and grows stronger as more and more ‘suspicious’ hands from the past are being presented. The DNA of the culprits known methods is just all over the tournaments that they murdered.
Dec. 31, 2016
David Jackson edited this comment Dec. 31, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
That clarification has been on the EBL site for a couple of weeks. The issue is ‘Will there be any historic disqualifications if the powers that be insist on video evidence when there is none?’ When the current appeals etc are over (if ever?) can we expect a deluge of disqualifications from prior to 2014? I hope so.
Dec. 29, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
I hope the phrase ‘all EBL events’ means what is says.So if evidence emerges of historic cheating, from however far back, and/or a pair/team is disqualified for any other reason then this rule is not limited to ‘as of, and including, Opatija 2014’. Is that so Jan?
Dec. 19, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Of course there is evidence from way back even if it is not video evidence. Statistical evidence is abundant. When players who have been found to be cheating NOW are shown to have as good if not even better results (Titles,Results, Butlers) in previous years, then either you assume their cheating NOW didn't improve their results or else they were cheating THEN. It is the knowledge that they were cheating NOW coupled with the same or better level of performance in previous years that tells us they were cheating THEN
July 21, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Nicely rearranged Joshua to make 3NT a favourite. Certainly with two balanced hands if you always bid 3NT with 4 Aces 2 Kings and Two Queens and nothing else that could be considered as even 1/2 a point then you will hope for this layout but the other, I don't know how many possible layouts, of these cards will make you a big loser in the long run
Jan. 22, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Hi Patrick.

Yes with x x x x, K x, A Q x, A x x x opposite A x x x, A Q x, K x, x x x x 3NT would certainly be odds on to make with the two 4/4 fits. Unfortunately I would probably be in the near zero percent 4 spades with the 4/4 fit.
Jan. 22, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
The majority of the Bridge World thinks the 90 year old 4321 valuation method of Milton Work undervalues Aces and Kings even in Balanced hands so don't get me going on the KS evaluator. I have just reviewed a book that basically suggests 41/2, 3, 2, 1 for the high honours. So that would be 14 points for each of these hands and 28 points total in that method. Milton Work only counts it as 26

Banzai 54321 has each hand as 17 points/60 (= a poorish 12 in Milton) so borderline to even open the bidding. If you open one of these hands in Banzai you will get to 2NT with approx 56% of the high card points (a little high perhaps) but even a passed out board is likely to be better than 3NT
Jan. 22, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
'As every book CORRECTLY says, 25-26 points only guaranties a reasonable play for a game, not a sure thing. The value of point count is that it simplifies all auctions and the curse of point count is that we still miss good games all too often by undervaluing the features missing in these'

Sure point count is the way to value these hands. Unfortunately these 26 points don't guarantee a reasonable play for game never mind a sure thing and they really should give a decent play for 3NT given the other positive features of the two hands if these cards were really worth 26 points out of 40 or 65% of the Total available. The Truth is that 4321 and maybe a 1/2 for Tens grossly overvalues Aces and Kings in Balanced hands. Counting 54321 for the five honour cards, you will see that these cards represent far far less than 65% of the total points available. The reason these two hands have so little play for nine tricks is the point count used is hopeless for these extreme hands (hands stuffed with Aces and Kings and there are much worse examples one could construct than this) There are just not enough High Card points to give these cards a reasonable play for 9 tricks.
Jan. 21, 2016
.

Bottom Home Top