[helene_t]

Good choice, Csaba.

It's not really relevant to this thread, though.

[Fluffy]

We shouldn't try it becuase the morons won't understand it. You would be good politics I guess, but politics and logic aren't very close.

[helene_t]

No Gonzalo, the problem with your suggestion is that the morons will understand that if they have had good hands all night they are bound to get some bad ones towards the end.

Of course morons will think nonsense like that regardless of whether the cards have been cooked, but what we want is that at least we can tell them that the nonsense is wrong

[hotShot]

khaggblo, on Jan 21 2010, 11:48 AM, said:

helene_t, on Jan 21 2010, 05:10 AM, said:

Great comment, Bluecalm. Nuff said.

I agree completely. On the other hand,

hotShot said:

Minimum: 15,958333333333
Maximum: 24,458333333333

how come the min and max are not integers?

Because it's the average of the 24 boards tourney.

khaggblo, on Jan 21 2010, 11:48 AM, said:

Furthermore, they should be symmetrical with respect to 20.

In theory you are right,if I would simulate a infinite number of deals the result should be symmetrical.
Now take a coin and throw it. Both sides are equally likely, but only one side can be up.
So to get symmetry you have to throw it a 2nd time, and there is a 50% chance you will get the same side up. If the same side comes up again you will have to throw another 2 times to have a chance for symmetry again. So whenever you set a limit lower than infinite for the the number of tries/simulations, you will find deviations from the symmetry.

khaggblo, on Jan 21 2010, 11:48 AM, said:

If the NS max is 20+n, the EW min must be 20-n (why consider only one side?).

Because the other side is not independent, because the sum has to be 40.

khaggblo, on Jan 21 2010, 11:48 AM, said:

The same principle applies to the simulation of hcp distribution (see HotShot in a previous post). If there are N deals where NS have, say, 10 hcp in a given number of deals, there must be exactly N deals where EW have 30 hcp. In theory, 20-n hcp occurs exactly as often as 20+n hcp. One could thus improve the accuracy of simulations somewhat by using the mean value of the number of deals having 20-n and 20+n hcp for both 20-n and 20+n hcp deals.

[khaggblo]

hotShot, on Jan 21 2010, 07:31 AM, said:

khaggblo, on Jan 21 2010, 11:48 AM, said:

helene_t, on Jan 21 2010, 05:10 AM, said:

Great comment, Bluecalm. Nuff said.

I agree completely. On the other hand,

hotShot said:

Minimum: 15,958333333333
Maximum: 24,458333333333

how come the min and max are not integers?

Because it's the average of the 24 boards tourney.

You are right.

hotShot, on Jan 21 2010, 07:31 AM, said:

khaggblo, on Jan 21 2010, 11:48 AM, said:

Furthermore, they should be symmetrical with respect to 20.

In theory you are right,if I would simulate a infinite number of deals the result should be symmetrical.
Now take a coin and throw it. Both sides are equally likely, but only one side can be up.
So to get symmetry you have to throw it a 2nd time, and there is a 50% chance you will get the same side up. If the same side comes up again you will have to throw another 2 times to have a chance for symmetry again. So whenever you set a limit lower than infinite for the the number of tries/simulations, you will find deviations from the symmetry.

Not necessarily. Just add up the NS statistics and the EW statistics and you have a completely symmetrical result. Why don't you think that would be more accurate? On average I think you are sitting EW as often as NS...

[hrothgar]

[quote name='Fluffy' date='Jan 21 2010, 02:31 PM'] We shouldn't try it becuase the morons won't understand it. You would be good politics I guess, but politics and logic aren't very close. [/QUOTE]
[QUOTE]The reasons that we shouldn't try this are

1. This violates the Laws of the game of bridge. We shouldn't break the laws because we feel like it.

2. This is entirely unnecessary. A number of people have posted statistical information that shows that - for all intents and purposes - the ridiculous cases that you postulate do not occur. Think about the last set of numbers that HotShot posted.

On average, NS will get dealt 20 HCP with a standard deviation of about 1 HCP
~66% of the time, NS will get deal between 19 - 21 HCPs
~95% of the time, NS will get dealt between 18-22 HCP
~99.7% of the time, NS will get dealt between 17-23 HCP

Wikipedia has a nice little chart that shows

Range
% Population in range
Expected frequency outside range
Approx. frequency for daily event

μ ± 1σ 0.682689492137 || 1 in 3 (Twice a week)
μ ± 2σ 0.954499736104 || 1 in 22 (Every three weeks)
μ ± 3σ 0.997300203937 || 1 in 370 (Yearly)
μ ± 4σ 0.999936657516 || 1 in 15,787 (Every 43 years - twice in a lifetime)
μ ± 5σ 0.999999426697 || 1 in 1,744,278 (Every 5,000 years - once in history)
μ ± 6σ 0.999999998027 || 1 in 506,842,372 (Every 1.5 million years)

BTW, this type of information is why is was blatantly obvious that the example that you posted in the the very first thread

[QUOTE]I averaged around 14 HCP, with my partner averaging 12 HCP, wich doesn't leave much for E-W.[/QUOTE]

was either a fabrication or, alternatively, a sign that something is very wrong with the dealing machines in the state of Spain (BTW, I'll note in passing that simplicity is really a virtue with these types of systems. The more options that you provide, the more complex the machine becomes. The more complex the machine the greater the chance of some kind of cock up)

3. Implementing this type of system would create an enormous number of hassles. The "morons" ask stupid questions and make stupid suggestions all the time. Its in their nature to do so. At the moment, there is a simple, logical answer as to why things work the way they do. Once we decide that we want to start playing god and choosing probability distributions that we like, we open the door to all sorts of subjective arguments about what constitues a "good" PDF as opposed to what constitutes a bad PDF, how, this should all be implemented, and all sorts of other inane blatter. I, for one, would prefer to sidestep this entire process.

With this said and done...

Where would you draw the line?
What type of spread is too great? (One sigma, two sigmas, three??? four???)

You must have some kind of idea about what constitutes too much?

[hotShot]

khaggblo, on Jan 21 2010, 02:01 PM, said:

Not necessarily. Just add up the NS statistics and the EW statistics and you have a completely symmetrical result. Why don't you think that would be more accurate? On average I think you are sitting EW as often as NS...

While I assumed that the curve should be symmetrical, because distributional functions usually are, my laziness suggested to get that as result of the simulation instead of thinking it trough.

[Fluffy]

hrothgar, on Jan 21 2010, 01:30 PM, said:

BTW, this type of information is why is was blatantly obvious that the example that you posted in the the very first thread

Only a fool would use this argument to discredit me when I made it OBVIOUS (for anyone with half a brain) that it was just my estimation.

But hey, it proves something, if I can be that wrong on my estimations given that I am much better in mathematics than the average player, your assumptions that people will be able to react based on the hands already played is obviously wrong.

LOL, sorry to go harsh, I don't have anything agaisnt you, but I like the way you go mad in this subject. Don't take me too seriously.

[Fluffy]

About rotating some hands vs redealing the full set, I am quite sure redealing is better in practice, althou theoreticahally I think it will not affect the HCP distribution.

However redealing a set for the HCP might affect the distribution on a subtle unaccountable way, the difference is so very very slim that I doubt anyone would ever be able to notice but it is there for sure, so in theory, rotating something at random is better. In practice it makes no difference at all.

[hrothgar]

Fluffy, on Jan 21 2010, 05:16 PM, said:

hrothgar, on Jan 21 2010, 01:30 PM, said:

BTW, this type of information is why is was blatantly obvious that the example that you posted in the the very first thread

Only a fool would use this argument to discredit me when I made it OBVIOUS (for anyone with half a brain) that it was just my estimation.

But hey, it proves something, if I can be that wrong on my estimations given that I am much better in mathematics than the average player, your assumptions that people will be able to react based on the hands already played is obviously wrong.

LOL, sorry to go harsh, I don't have anything agaisnt you, but I like the way you go mad in this subject. Don't take me too seriously.

Fine. From now on, when you make comments like

Quote

I averaged around 14 HCP, with my partner averaging 12 HCP, wich doesn't leave much for E-W. ...  I am sure that what happenedyesterday was really unfair.

we'll understand that you're just guessing about stuff and we shouldn't assume that there is any relation between your statements and what really happened.

I'm going to repeat my request that you provide a link to the boards played in that session. I'd love to know just how how large the discrepancy is between your perception and reality...

As to your comment about being much better at mathematics than the average player... Your original quote reminded me of the sort of fanciful tales told by an enthusiastic eight year old (I remember my sister telling us how she had rescued a deer from a frozen lake and it let her ride it). These two examples - averaging 26+ HCP across a regulation tournament and riding a wild deer - are about just about as likely. I wouldn't want to be lauding my maths skills right after advancing this sort of example.

(For the record, the issue at hand is not your perception regarding the strength of the hands. We all know that perceptions can be way off... What is important is that you tried to advance an argument buttressed by this sort of factoid. If you really were that good at math, you wouldn't expect folks to swallow an anecdotal discreption of a six sigma event)

[Fluffy]

sorry richard, my mistake, it should had been obvious to me that what you wanted is a fight and to get over me, maybe in a look of reaching higher in the social hierarchy or something.

I can try and give you some fight back, but I will be bored soon though, I hope it will be enough for you to feel smarter than me.

Quote

Comment 1:  We already have to deal with inane discussion with idiots who complain that statistically unbiased random number generators aren't fair enough.  The only thing that makes these discussions tolerable is that its relatively easy to demonstrate whether or not a random number generator is biased.  I, for one, don't want to have to deal with entirely new inane discussions with idiots who complain that statistically biased random numbers generators are unfair in the wrong way.

Fluffy, on Jan 21 2010, 02:31 PM, said:

We shouldn't try it becuase the morons won't understand it. You would be good politics I guess, but politics and logic aren't very close.

Quote

The reasons that we shouldn't try this are

1.  This violates the Laws of the game of bridge.  We shouldn't break the laws because we feel like it.

2.  This is entirely unnecessary.  A number of people have posted statistical information that shows th...

I get that I am wrong whatever I write, even if its the same you wrote before. Maybe you switched your mind in the proccess of the thread?

[billw55]

hotShot, on Jan 21 2010, 05:06 AM, said:

To finalize the topic
I simulated about 114 years of 7 24 board tourneys a week.

Number of tourneys: 41666
The mean HCP NS: 20,00165502648
The variance: 0,95194261701304
The standard deviation: 0,97567546705503
Minimum: 15,958333333333
Maximum: 24,458333333333

Personally, I am a little surprised that the sigma is so low, for sets of only 24 boards. But a sim is a sim. I guess it's another example of human perception/expectation often not matching real probabilities.

hrothgar said:

3. Implementing this type of system would create an enormous number of hassles. The "morons" ask stupid questions and make stupid suggestions all the time. Its in their nature to do so. At the moment, there is a simple, logical answer as to why things work the way they do. Once we decide that we want to start playing god and choosing probability distributions that we like, we open the door to all sorts of subjective arguments about what constitues a "good" PDF as opposed to what constitutes a bad PDF, however, this should allbe implemented, and all sorts of other inane blatter. I, for one, would prefer to sidestep this entire process.

Hell yes. I get tired enough of hearing dealing conspiracy theories as it is. If they had any actual facts to back it up, I don't think I could take it.

[hrothgar]

Fluffy, on Jan 21 2010, 07:35 PM, said:

Quote

The reasons that we shouldn't try this are

1.  This violates the Laws of the game of bridge.  We shouldn't break the laws because we feel like it.

2.  This is entirely unnecessary.  A number of people have posted statistical information that shows th...

I get that I am wrong whatever I write, even if its the same you wrote before. Maybe you switched your mind in the proccess of the thread?

I'm going to repeat one of my original posts:

Quote

Simply put:

1. If you build enough "state" into the system such that people can notice, they're going to start adapting their behaviour [modifying would have been a much better word choice]

2. If you don't build enough state into the system for people to notice, what's the point?

I don't think that its inconsistent to advance this line of reasoning...

[bluecalm]

Quote

We shouldn't try it becuase the morons won't understand it. You would be good politics I guess, but politics and logic aren't very close.

Man, you are really out of your mind. You are trying to mastermind a solution to conform expectations of small group of players (it's safe to say that if you can use term "moron" to any group probably that one is the best candidate) breaking law of bridge and sense of justice of majority of others.

Your idea is based on this : "some people probably would like to have some cards in every tournament". What about majority of players want to:
-safe feeling that hands are random and nobody manipulated them
-a chance to get all the cards (or majority) of hand in the tournament to finally have chance to win (they are very good at bidding and declaring, weak at defence)
-chance to get all the cards in the tournament to have the best bridge evening in their lifetime
-be sure that previous (or following) hands don't affect the one played

What about :
-people who are good at "going to the next hand"; I guess Hamman and Meckwell would need to reconsider their approach Thinking all the time about previous hand would be crucial
-people who think randomness itself is valuable part of the game
-people who are developing their judgement and methods in bridge assuming that all hands are independent events

Most players want at least one of the above. Some small group of usually uneducated players wants "justice" (in the very false sense) in EVERY tournament played. Why would you want to conform to this unreasonable demands sacrificing all the above ?

[kayin801]

billw55, on Jan 21 2010, 09:39 AM, said:

hotShot, on Jan 21 2010, 05:06 AM, said:

To finalize the topic
I simulated about 114 years of 7 24 board tourneys a week.

Number of tourneys: 41666
The mean HCP NS: 20,00165502648
The variance: 0,95194261701304
The standard deviation: 0,97567546705503
Minimum: 15,958333333333
Maximum: 24,458333333333

Personally, I am a little surprised that the sigma is so low, for sets of only 24 boards. But a sim is a sim. I guess it's another example of human perception/expectation often not matching real probabilities.

Does anyone else find it eerie that one standard deviation is almost exactly 1 HCP? (Granted this is only for 24 board tourneys) Or am I just being a geek? BTW, thanks for this sim Hotshot, it's exactly what I wanted to know.

[hrothgar]

kayin801, on Jan 21 2010, 10:38 PM, said:

billw55, on Jan 21 2010, 09:39 AM, said:

hotShot, on Jan 21 2010, 05:06 AM, said:

To finalize the topic
I simulated about 114 years of 7 24 board tourneys a week.

Number of tourneys: 41666
The mean HCP NS: 20,00165502648
The variance: 0,95194261701304
The standard deviation: 0,97567546705503
Minimum: 15,958333333333
Maximum: 24,458333333333

Personally, I am a little surprised that the sigma is so low, for sets of only 24 boards. But a sim is a sim. I guess it's another example of human perception/expectation often not matching real probabilities.

Does anyone else find it eerie that one standard deviation is almost exactly 1 HCP?

I would have used the word "convenient"...

This is a factoid that is going to remain embedded in yee old noggin...
I really hope its accurate ;-)

[billw55]

kayin801, on Jan 21 2010, 02:38 PM, said:

Does anyone else find it eerie that one standard deviation is almost exactly 1 HCP? (Granted this is only for 24 board tourneys) Or am I just being a geek? BTW, thanks for this sim Hotshot, it's exactly what I wanted to know.

Well there must be some number of boards for which sigma is 1 hcp, or very close to it. As it happens, it's pretty close for 24 boards. Maybe Hotshot ran sims for a few different set sizes, and presented the one with sigma closest to 1. Or maybe he just got lucky his first try

[hotShot]

I took 24 because hrothgar used that number.
Seems I have been lucky.

30 board tourneys:
Number of tourneys: 33333
The mean HCP NS: 20.00167601676
The variance: 0.76401016443091
The standard deviation: 0.87407674973706
Minimum: 16.333333333333
Maximum: 23.5

18 Board tourneys
Number of tourneys: 55555
The mean HCP NS: 20.00167601676
The variance: 1.2653701780311
The standard deviation: 1.1248867400903
Minimum: 15.388888888889
Maximum: 24.5

128 Boards
Number of tourneys: 7812
The mean HCP NS: 20.001689108103
The variance: 0.1810962450686
The standard deviation: 0.42555404482697
Minimum: 18.4609375
Maximum: 21.515625

[Fluffy]

I have though this evening that maybe it is best not to rotate hands to even the number of HCP, but rather to rotate hands so there is at most 60-65% of hands in a line, where the par contract belongs to them (even if the par contract is a save)

[jdonn]

Fluffy, why? No one would ever notice a difference. If they did then they would find a way to take advantage of it. So either you are wasting your time or changing the game, and in either case breaking the laws. I think you just have to accept there is very little support for something like this.