Page 1 of 1
Something wrong with BBO dealer?
#1
Posted 2020-August-08, 06:13
In the past month or two, I've been seeing, in "competitive" tables and private team matches, what seemed an abnormally large number of 5-0 splits and similar rarities. A friend who competes frequently in robot events just e-mailed me TWO deals in which somebody had a nine-card suit (both spades). Have other people had similar experiences?
#2
Posted 2020-August-08, 08:07
ceblair, on 2020-August-08, 06:13, said:
In the past month or two, I've been seeing, in "competitive" tables and private team matches, what seemed an abnormally large number of 5-0 splits and similar rarities. A friend who competes frequently in robot events just e-mailed me TWO deals in which somebody had a nine-card suit (both spades). Have other people had similar experiences?
Please look at any of the last 50 threads debunking the idea that the BBO dealer is borked.
Alderaan delenda est
#3
Posted 2020-August-09, 01:33
I have also heard that on BBO, suits sometimes split 5-0. Unbelievable!
The easiest way to count losers is to line up the people who talk about loser count, and count them. -Kieran Dyke
#4
Posted 2020-August-09, 03:54
As I was under the same impression about the frequency of uneven hands,I decided to check the distributions
over about 20 daylong tournaments that I played lately. I was mostly interested in suit contracts played with 8-card fit.
Interestingly I counted three 5-0, fourteen 4-1 and thirty-two 3-2 distributions which, within the margin of error
inherent to the limited size of the sample, pretty much follows the expected probability of each distribution.
I also checked for 9-card fit and it turned out a little bit off, but the 2-2 distributions was higher than expected.
This goes to show that our brain likes to remember the oddities but quickly forget what is just "normal".
I do not play team but I would expect the BBO dealer to follow the same rules as for the daylong tourneys.
Now, two nine-cards suit in the same tournament that does not sound right.
over about 20 daylong tournaments that I played lately. I was mostly interested in suit contracts played with 8-card fit.
Interestingly I counted three 5-0, fourteen 4-1 and thirty-two 3-2 distributions which, within the margin of error
inherent to the limited size of the sample, pretty much follows the expected probability of each distribution.
I also checked for 9-card fit and it turned out a little bit off, but the 2-2 distributions was higher than expected.
This goes to show that our brain likes to remember the oddities but quickly forget what is just "normal".
I do not play team but I would expect the BBO dealer to follow the same rules as for the daylong tourneys.
Now, two nine-cards suit in the same tournament that does not sound right.
#5
Posted 2020-August-09, 05:31
Players who complain about (extreme) patterns and or the amount HCP they don't know or get statistics of 52!. But that is ok just trust BBO it is not in their interest to mess with the deals. And if they do (with for example goulash boards) they will tell you.
#6
Posted 2020-August-09, 13:03
cherdano, on 2020-August-09, 01:33, said:
I have also heard that on BBO, suits sometimes split 5-0. Unbelievable!
I haven't got a 5-0 trump split against me since I joined the Platinum Double Prime Club. I did get a 4-1 trump split the other day that kind of rattled me, but it turns out my membership was almost expired and BBO was sending me a renewal "notice". Once I renewed, 3-2 splits are the order of the day. It's great not having to worry about bad splits wrecking your contracts.
#7
Posted 2020-August-09, 22:24
The chance of any one unusual thing happening is very small. The chance of something unusual happening is actually fairly high, and that's what we remember.
As usual, xkcd covers it nicely: https://xkcd.com/882/
As usual, xkcd covers it nicely: https://xkcd.com/882/
#8
Posted 2020-August-10, 06:50
No set of humanly acceptable artificially generated sequences of random numbers is really random at all because so many sequences are excluded. For example,
Even more amusing: Suppose the sequence of ten million digits looks random and passes human randomness tests. Belatedly, however, somebody notices that it comprises the first ten million digits of pi = 314 ...
- Suppose you ask the random number generator for ten million random decimal digits.
- It produces a sequence of ten million sevens,
- This would almost certainly be rejected
- But that sequence is just as likely as any other specific sequence of ten million digits
Even more amusing: Suppose the sequence of ten million digits looks random and passes human randomness tests. Belatedly, however, somebody notices that it comprises the first ten million digits of pi = 314 ...
#9
Posted 2020-August-12, 09:35
nige1, on 2020-August-10, 06:50, said:
Even more amusing: Suppose the sequence of ten million digits looks random and passes human randomness tests. Belatedly, however, somebody notices that it comprises the first ten million digits of pi = 314 ...
That wouldn't pass, about 18% of the digits of pi are 3
Ok, just looking at the first 50 digits...
Page 1 of 1