Join Bridge Winners
All comments by Nicolas Hammond
1 2 3 4 5 6 7 8 9 ... 135 136 137 138
You are ignoring the author of this comment. Click to temporarily show the comment.
I believe it was Colin Jordan of the National Socialist Movement who created the slogan, not the Tory party. Peter Griffiths, the Conservative (Tory) candidate for Smethwick in 1964 did not approve, but also did not condemn the ad. Griffiths won.
17 hours ago
You are ignoring the author of this comment. Click to temporarily show the comment.
Some of those reporting recently are on-line games not F2F.
March 30
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes, but you have to play on-line.
March 28
You are ignoring the author of this comment. Click to temporarily show the comment.
For example: go to The Vugraph Project (TVP) at link above.

Click on 1955. Click on Bermuda Bowl. Click on seg1.

The URL is something like
https://www.sarantakos.com/bridge/vugraph/1955/seg1.lin

To get BBO to display, prepend the following:

https://www.bridgebase.com/tools/handviewer.html?bbo=y&linurl=

So, for this example

https://www.bridgebase.com/tools/handviewer.html?bbo=y&linurl=https://www.sarantakos.com/bridge/vugraph/1955/seg1.lin
March 28
You are ignoring the author of this comment. Click to temporarily show the comment.
You are ignoring the author of this comment. Click to temporarily show the comment.
@Cameron: I think that's the point.
March 24
You are ignoring the author of this comment. Click to temporarily show the comment.
Wikipedia:
https://en.wikipedia.org/wiki/Bridge_Murder_case
Book:
The Devil's Tickets: A Vengeful Wife, a Fatal Hand, and a New American Age
https://www.amazon.com/Devils-Tickets-Vengeful-Fatal-American/dp/1400051630
March 24
You are ignoring the author of this comment. Click to temporarily show the comment.
@MR: MM?
March 19
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm happy to devote a little time to help. I have the data on MZ's lead. But need some volunteers to do the crowd-sourcing of the videos. Specifically to record the time of the tray being passed and the time of the opening lead.

This would help substantiate or disprove the allegation of using BIT to help with the opening lead. Three or four leads doesn't do that.
March 19
You are ignoring the author of this comment. Click to temporarily show the comment.
“Cheating is likely to reduce mistakes, simply because they still exist does not mean cheating is not present.”

Agreed. But their number of mistakes is very high compared to top pairs and nowhere close to any of the known cheating pairs. Their declarer and defensive rankings are similar; another indicator of no cheating. They rank in the third quartile in both. Pairs that are bad at declarer player, but fantastic on defense are red flags. This does not apply to MZ.

The allegation appears to be that MZ have illegal communication before the opening lead, possibly based on tempo.

I'm happy to co-ordinate a crowd-source effort to time-stamp videos to get the data so it can be analyzed. I just need some crowd-source volunteers… PM me if interested.
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
OK. So let's find all videos with MZ, time stamp them, then we can have a proper discussion on the facts. We should do at least one other pair for reference. Are you willing to time stamp a video? Will take about 30-60 minutes. If so, PM me.
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
That's not where you are supposed to stick a meat thermometer.
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
At NT, the expected declarer advantage is that less than 75% of the time will the opponents find a good lead. Arguably picking a suit at random (and then the right card!) is better than what world class players do. The top experts only find a safe lead against a NT contract 73% of the time.

Knowing this, then 1 in 4 boards, the opening lead will give up a trick.

So… using the ACBL Clubs data for 9 HCP - 13,386 - 54.5% (62.6%)

Before the opening lead 45% of contracts will not make. Assume that 1 in 4 leads are bad. So declarer should make an addition 11% of contracts. Subtract out the contracts that were going to take 7 tricks DD and giving up a single trick on opening lead doesn't affect if the final contract makes and you get close to the SD value.

My point with all of this is that the OP value of 54% of 3NT making with 16 opposite 9 is invalid in a SD world. The SD world shows approximately 65%. My guesstimate of 64% (see above) was quite close.
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
This anonymous 4 minute video shows Zimmermann drinking coffee, has three (it might be four) of Zimmermann's leads from various tournaments, a Multon director call, and androgenic alopecia (early onset of baldness) in male Bridge players.

It proves/shows absolutely nothing (apart from the coffee and alopecia).

I could put together a video of any top pair showing the same (coffee, leads, baldness etc.)

I have over 400 leads by Zimmermann on BBO (and same for Multon).

If anyone does an analysis, then you need to include ALL videos, not select a very small set and imply something nefarious. This is not fair to MZ.

If someone wants to put together a counter-factual video to dispel this rumor, let me know.

There is some data from MZ from the 2014 EBTC. See http://www.bridgescoreplus.com/data/ebtc_2014/index.html. Click on “Pairs”. Click on “Multon-Zimmermann”. Not all of their videos from EBTC were crowdsourced. I haven't watched the video links of when Z is on lead.

If someone has some time to watch some Bridge video and time-stamp leads, I can send you a spreadsheet to fill out. I can then publish the same data in the same way as EBTC 2014 data. From the crowd sourcing effort, it takes about 30-60 minutes to time-stamp a two hour video. If you are interested, PM me with your e-mail.

If anyone has links to videos of MZ playing (saves me trying to find them on Youtube), let me know. I can look up the BBO record and then get them time-stamped by volunteers, if we have any.

“A lie will go round the world while truth is pulling its boots on.”. Spurgeon from “Falsehood will fly from Maine to Georgia, while truth is pulling her boots on” (1820).
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
To add a time stamp to a Youtube videos add “&” (additional argument) “t=” time value “HHhMMmSSs” where HH is the number of hours (not needed if 0), MM is the number of minutes and SS is the number of seconds.

In this case add “&t=1m59s”

The URL becomes

https://m.youtube.com/watch?v=Kf0MHPiX8HM&t=1m59s
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
I was curious so I looked at the real world data.

In general double dummy analysis is fine, but when applied to a specific case, it may not be.

First: OP's definition was 16 HCP, either 8, 9 or 10 HCP in dummy. Balanced hand. Contract of 3NT. This assumes no opponents bidding/interfering to the 4 level. What is weird is that dummy has the 16 HCP and declarer the 9 HCP. Dummy is announcing “aces and spaces”. This is unusual,

My definition was 16 HCP in declarer (any shape), 8, 9 or 10 HCP in dummy (any shape). Final contract of 3NT.

This is a different base set than OP because players have decided to bid 3NT on their combined hands.

But… it should be reasonably close.

I have data from ACBL clubs, tournaments (regionals/sectionals), NABCs, Norwegian Bridge, BBO top tournaments.

Thus…. about 15 different data sets ( x )

These have a wide variety of skill level and MPs/IMPs.

The ACBL data is nearly all MP. NBF is mixed (IMPs/MPs), Top tournaments are IMPs.

“When the 16-pointer has random aces, 3NT makes 54.1% of the time.” (dummy has 9 HCPs)

The point about my previous post is that double dummy (DD) and single dummy (SD) differ widely when marginal to game. Here's the real word data for 9 HCP:

Data set - number of boards - DD % (actual percentage)


8 HCP

ACBL tournaments - 8,252 - 45% (55%)
ACBL Clubs - 4,103 - 40% (54%)
NABCs - 4,278 - 35% (55%)
Top tournaments - 549 - 39% (51%)
NBF - 2,085 - 44% (57%)

9 HCP

ACBL tournaments - 11,302 - 63.0% (66%)
ACBL Clubs - 13,386 - 54.5% (62.6%)
NABCs - 7,729 - 55.5% (63.4%)
Top tournaments - 856 - 65.6% (72.0%)
NBF - 2841 - 64.5% (68.6%)

10 HCP

ACBL tournaments - 39,271 - 77.4% (76%)
ACBL Clubs - 22,585 - 78% (75%)
NABCs - 8,137 - 72% (73%)
Top tournaments - 865 - 74% (78%)
NBF - 2841 - 64.5% (68.6%)

The ACBL Clubs and NABC data for 9 HCP hands for DD % are close to the OP. The others are not. This suggests players making choices on the value of their hand.

In all 9 HCP hands, SD makes more contracts than DD.

“Specifically when the 16-pointer has zero aces, 3NT makes 42.8%. Holding one ace, 3NT makes 47.7%. With two aces 54.6%, with three aces 61.9%, and with four aces and spaces — drumroll — 3NT makes 64.9%. So the sims suggest aces and spaces are a highly desirable holding, at least for this narrow configuration.”

To me “Aces and spaces” might be A432 A32 A32 A32 but “Aces” may be A1098 A109 A109 A109. The latter to me is much stronger than the former. I did not program the logic of “Aces and spaces” just “Aces”.

Same data, but this time I look at the number of aces held.
This is the % of contracts that actually make, not what DD states.
n/a means total number of hands is < 50 making the percentage too variable to be meaningful
8 HCP: 0 - 1 - 2 - 3 - 4 - Aces

ACBL tournaments - 62% - 54% - 55% - 54% - 53%
ACBL Clubs - n/a - 53% - 56% - 48% - n/a
NABCs - n/a - 53% - 57% - 51% - n/a
Top tournaments - n/a - 56% - 50% - 45% - n/a
NBF - n/a - 60% - 60% - 55% - n/a

9 HCP: 0 - 1 - 2 - 3 - 4 - Aces

ACBL tournaments - 75% - 65% - 67% - 63% - 63%
ACBL Clubs - 71% - 60% - 63% - 63% - 67%
NABCs - 69% - 73% - 56% - 71% - n/a
Top tournaments - n/a - 66.8% - 73% - 74% - n/a
NBF - n/a - 66% - 67% - 78% - n/a

10 HCP: 0 - 1 - 2 - 3 - 4 - Aces

ACBL tournaments - 70% - 74% - 77% - 75% - 77%
ACBL Clubs - 72% - 73% - 75% - 76% - 89 %
NABCs - 49% - 69% - 74% - 79% - n/a
Top tournaments - n/a - 80% - 79% - 70% - n/a
NBF - n/a - 78% - 76% - 75% - 75%

Conclusion:

If your partner is an American, swap him/her for a Norwegian.
March 18
You are ignoring the author of this comment. Click to temporarily show the comment.
“If they wanted to cheat, they would have a subtle code.”

I've probably spent more time than most looking for cheating pairs. You can detect cheating through a pairs' statistics. Based on MZ's statistics it is highly unlikely they are cheating during the play of the hand. They simply make far too many mistakes.
March 17
You are ignoring the author of this comment. Click to temporarily show the comment.
Richard refers to the general case, not a specific case (# aces in 16 HCP hand).

RP: “Whatever the case, if both sides are equally imperfect, the errors should tend to cancel out in the long run, approaching the double-dummy state.” Problem is that both sides are not equally imperfect. The opening lead errors rate is 1 in 4 for NT contracts. After the opening lead, defenders play better than declarer. Data from top events, IMP scoring. However the opening lead factor is still the largest determinant.

I agree with RS (above).

But using the value of 54.1% to make the determination of bidding 3NT is incorrect. Assuming that the other 46% of hands fail by one trick (invalid assumption ,but go with it), defenders will give up a trick on 1 in 4 hands on the opening lead. So about another 10% of hands will make, given an approximate total of around 64%. That's a big difference.
March 17
You are ignoring the author of this comment. Click to temporarily show the comment.
www.okbridge.com
March 17
You are ignoring the author of this comment. Click to temporarily show the comment.
@Avon; “Forgive my whining, but what is your explanation for that?”

It might be simple statistical expectation.

Zimmermann's accuracy rate is 64%, top pairs are just under 81%. We would probably expect a bell curve around the 81% mark. It could be that the pro tries to find the killer leads, the sponsor the boring leads.

I have just over 400 leads for Zimmermann. We are talking about 12 boards in 400. Change those to a bad lead and Zimmermann is average.

Greco is #60 (just above average), Hampson is #105 (below 79%). Their difference is 2%. MZ difference is 3%.

The accuracy of opening leads is not a good test to detect cheating, there are better ones. The perfect example of the flawed nature of using this metric is Ron Schwartz.

Let me use Bob Hamman as an example to show statistical variations:

With Soloway, 83%, #22.
With Zia, 78%, #116
With Wolff, 80.6%, #69.

Why such a discrepancy with his three partners?

I also have the same opening lead data for no trump contracts. Zimmermann drops to #15. There are other sponsors better than Zimmermann.
March 17
1 2 3 4 5 6 7 8 9 ... 135 136 137 138
.

Bottom Home Top