Thread Rating:

Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 17th, 2010 at 9:34:14 AM permalink
Hello all,

I have a newbie question which I hope the more experienced of you can help me with (I've forgotten most of my high school math!)

Some baccarat tables offer a player and/or banker pair bet. The payouts vary. (ranging from 1 to 6, to 1 to 12?)

My question is as such: How do I calculate the probability that the very first hand of an 8-deck baccarat game will consist of a pair? And subsequently, as more and more cards are used, how do we calculate the changing probabilities?

My thoughts will be to simply assume the first card is, lets say a nine. Then the probability that the next card will be a nine will be (32-1)/(416-1)= 0.74699. Since one nine and one card out of the 416 cards have come out. This is not what the wizard says. Where am I going wrong? And how do I calculate the probabilities that a pair will appear once a number of cards (with their values recorded) have been used?

Thank you very very much in advance to anyone who replies.
Doc
Doc
  • Threads: 46
  • Posts: 7287
Joined: Feb 27, 2010
July 17th, 2010 at 9:45:16 AM permalink
Quote: Magical

I've forgotten most of my high school math!
....
(32-1)/(416-1)= 0.74699



Proved your point. ;-)



Sorry. Couldn't pass up the opportunity. I'll let someone else do the math correctly.
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
July 17th, 2010 at 11:01:44 AM permalink
Quote: Magical

Hello all,

I have a newbie question which I hope the more experienced of you can help me with (I've forgotten most of my high school math!)

Some baccarat tables offer a player and/or banker pair bet. The payouts vary. (ranging from 1 to 6, to 1 to 12?)

My question is as such: How do I calculate the probability that the very first hand of an 8-deck baccarat game will consist of a pair? And subsequently, as more and more cards are used, how do we calculate the changing probabilities?

My thoughts will be to simply assume the first card is, lets say a nine. Then the probability that the next card will be a nine will be (32-1)/(416-1)= 0.74699. Since one nine and one card out of the 416 cards have come out. This is not what the wizard (https://wizardofodds.com/baccarat/baccaratapx5.html) says. Where am I going wrong? And how do I calculate the probabilities that a pair will appear once a number of cards (with their values recorded) have been used?

Thank you very very much in advance to anyone who replies.


Your decimal is in the wrong spot. 31/415 = 0.074699 . That's the the probability that the players first 2 cards are a pair. Sometimes the player gets a third card, which may result in three of a kind, or may pair up one of the first 2 cards. (I think that the side bet here only counts pairs in the first 2 cards, and three of a kind, but not pairs made by the third card.)
“Man Babes” #AxelFabulous
Wizard
Administrator
Wizard
  • Threads: 1518
  • Posts: 27033
Joined: Oct 14, 2009
July 17th, 2010 at 11:03:29 AM permalink
Quote: Doc

I'll let someone else do the math correctly.



(32-1)/(416-1) = 31/415 = 0.074699
"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)
Doc
Doc
  • Threads: 46
  • Posts: 7287
Joined: Feb 27, 2010
July 17th, 2010 at 12:26:12 PM permalink
Yes, Wizard, I had done that part of the math, which was why I made my snide remark. It was such things as the twists and turns that miplet mentioned that I didn't want to get into myself. I don't even know the game itself.
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 17th, 2010 at 3:19:30 PM permalink
Yes, that was a mistake on my part, i put the decimal at the wrong place. My bad.

However the question is still the same, the answer i come up with is 0.074699. However the numbers on the WOO site shows 0.071663 and 0.071864 for player and banker respectively. And yes, I am talking about the first 2 cards of the player and banker, the third card is not involved in determining if the hand has a Pair. I realize my calculations ignore the fact that the cards are dealt player, banker, player, banker. How do I include the situation where the cards are dealt this way? And how are the numbers 0.071663 and 0.071864 calculated?

Thank you to those who replied so far :)
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
July 17th, 2010 at 3:36:09 PM permalink
Quote: Magical

Yes, that was a mistake on my part, i put the decimal at the wrong place. My bad.

However the question is still the same, the answer i come up with is 0.074699. However the numbers on the WOO site shows 0.071663 and 0.071864 for player and banker respectively. And yes, I am talking about the first 2 cards of the player and banker, the third card is not involved in determining if the hand has a Pair. I realize my calculations ignore the fact that the cards are dealt player, banker, player, banker. How do I include the situation where the cards are dealt this way? And how are the numbers 0.071663 and 0.071864 calculated?

Thank you to those who replied so far :)


You did the calculations just fine. The 0.071663 and 0.071864 aren't counting pairs the become 3 of a kind when a third card is delt.
“Man Babes” #AxelFabulous
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 18th, 2010 at 11:25:46 AM permalink
Hmm... I'm not counting the pairs that become 3 of a kind either. I'm not including the third card into the calculations. What am I missing? And how do i calculate the probability of a pair after lets say all the aces and 8 kings are used up already?
Doc
Doc
  • Threads: 46
  • Posts: 7287
Joined: Feb 27, 2010
July 18th, 2010 at 11:43:22 AM permalink
Partial answer: I think he is saying that to get the right figure for pairs, you have to consider the third card. If you don't, some of the pairs that you find will end up not being pairs after the third card is dealt -- they will be part of a three of a kind and should be excluded. Excluding those gives the slightly lower probability of a plain pair. Or did I misunderstand your post?

Edit for clarification (hopefully): When the first two cards are a pair, you can't know whether they will remain a "pair" until you determine whether there will be a third card and what the chance is that it will change the "pair" into trips.
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
Thanked by
spinswizard
July 18th, 2010 at 1:11:26 PM permalink
Quote: Magical

Hmm... I'm not counting the pairs that become 3 of a kind either. I'm not including the third card into the calculations. What am I missing? And how do i calculate the probability of a pair after lets say all the aces and 8 kings are used up already?


If you are only calculating if the 1st 2 cards make a pair use (d*4-1)/(d*52-1) where d is the number of decks. This is what you already did. The side bet in the baccarat appendex pays more when the pair becomes a three of a kind. I think that is why you are confused. Here are the numbers for player pair:
pairs that become trips (0.003036) + pairs that don't become trips (0.071663) = original pairs (.074699)
The main baccarat page also lists pairs just on the 1st 2 cards.
I'll answere your other question when I get my brain into thinking mode. I know how, its putting it into words.
“Man Babes” #AxelFabulous
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
July 18th, 2010 at 2:50:27 PM permalink
Number of ways to have a pair of kings: k= # of kings. k*(k-1)/2 or combin(k,2) in excel. repeat for all the values.
Add them up. that sum = p .
t = total number of cards. = t*(t-1)/2 or combin(t,2) in excel. call this x.
p/x is the number you want.

Arrrg that looks confusing. I'll make a google docs spread sheet later tonight.
“Man Babes” #AxelFabulous
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 19th, 2010 at 12:56:07 AM permalink
Thank you! I now understand why my numbers are different. Basically coz i was calculating the probability for all the instances that pairs would appear (both trips and non-trips pairs). Stupid me haha.

Your second post is so over my head tho. It might be simple math, but don't understand ANY of it. Thank you very much for the replies and I look forward to the google doc spread sheet miplet!
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
Thanked by
spinswizard
July 19th, 2010 at 9:45:54 AM permalink
google docs pairs
You can ignore the colunms for colored,suited,non colored as I will be adding more paytables often used in blackjack side bets.
“Man Babes” #AxelFabulous
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 20th, 2010 at 12:04:40 AM permalink
Wow! Thank you very much miplet. Hugely appreciated.
I'll play around with the spreadsheet and see how it works exactly.
Do you think it is possible to consistently have an edge over the house (in terms of pair bets)?
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
July 20th, 2010 at 12:58:05 AM permalink
Quote: Magical


Do you think it is possible to consistently have an edge over the house (in terms of pair bets)?


No.
“Man Babes” #AxelFabulous
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 20th, 2010 at 9:05:09 AM permalink
Why do you say that? As the cards are used during play, the probability of a pair appearing will change. Is it not possible that the chance of a pair appearing will be be higher than 1/11? Or is it a very rare occurrence; rendering it useless? (like card counting in baccarat)
miplet
miplet
  • Threads: 5
  • Posts: 2142
Joined: Dec 1, 2009
July 20th, 2010 at 9:20:37 AM permalink
Quote: Magical

Why do you say that? As the cards are used during play, the probability of a pair appearing will change. Is it not possible that the chance of a pair appearing will be be higher than 1/11? Or is it a very rare occurrence; rendering it useless? (like card counting in baccarat)


There is no simple +- tags that you can use, as there is in blackjack. My guess is it is a very rare occurrence even with perfect play.
“Man Babes” #AxelFabulous
Magical
Magical
  • Threads: 1
  • Posts: 7
Joined: Jul 17, 2010
July 24th, 2010 at 11:06:23 AM permalink
Yep I agree, but IF a computer program can be used to track the cards played (ie online) will it be a good idea to try it out?
BrockWindsor
BrockWindsor
  • Threads: 1
  • Posts: 14
Joined: Jun 10, 2010
October 14th, 2010 at 12:55:15 AM permalink
If it's on a Midi Bac game with the cut card 14 cards from the back (95-98% pen). Are we still not going to see enough profitable remaining subsets to make it worth the time? Each rank would be counted on the scorecard. Some thoughts or analysis public for this?
-BW
  • Jump to: