Three card poker hole card team vs solo
April 13th, 2018 at 9:42:05 AM
permalink
Hi everyone,
I was reading a post by Jacobson (I'm a new member so not allowed to post links, but you can find the post on Jacobson's "AP Heat" blog, it's called "Hole-Carding Three Card Poker: Random Walks & ROR") where he describes an experience a friend of his had with hole carding three card poker.
Just to summarize, his friend had a 50k bankroll and a betting unit of $200. According to Jacobson's analysis that results in a RoR of 0.32% with perfect play. After three months the whole bankroll was lost.
Now, I am aware that the variance at hole carding three card poker is crazy, not even counting mistakes and cover play, but my question is: could he lower the risk even further by playing with a team betting 7 spots of $50 (same 50k bankroll) for example?
I come from a blackjack background where playing multiple spots critically reduces risk so I'm wondering if it's the same with three card poker.
Thanks
I was reading a post by Jacobson (I'm a new member so not allowed to post links, but you can find the post on Jacobson's "AP Heat" blog, it's called "Hole-Carding Three Card Poker: Random Walks & ROR") where he describes an experience a friend of his had with hole carding three card poker.
Just to summarize, his friend had a 50k bankroll and a betting unit of $200. According to Jacobson's analysis that results in a RoR of 0.32% with perfect play. After three months the whole bankroll was lost.
Now, I am aware that the variance at hole carding three card poker is crazy, not even counting mistakes and cover play, but my question is: could he lower the risk even further by playing with a team betting 7 spots of $50 (same 50k bankroll) for example?
I come from a blackjack background where playing multiple spots critically reduces risk so I'm wondering if it's the same with three card poker.
Thanks
April 13th, 2018 at 9:56:24 AM
permalink
My gut feeling says you are tripling your risk, and are better off looking at other carnival games with sloppy dealers.
That may not be the case, I only say it because I know how crushing 3CP can be, even when you have an advantage.
That may not be the case, I only say it because I know how crushing 3CP can be, even when you have an advantage.
April 13th, 2018 at 10:16:27 AM
permalink
Thanks, just to clarify, I'm not attempting to do this, I'm curious from a purely mathematical point of view. With BJ I can just put the numbers in CVCX but I don't know how to calculate risk with three card poker.
April 13th, 2018 at 10:58:48 AM
permalink
Those numbers are if you play perfectly with no mistakes or discreet play. Making mistakes in this game can be disastrous.
DUHHIIIIIIIII HEARD THAT!
April 13th, 2018 at 11:03:30 AM
permalink
Quote: IbeatyouracesThose numbers are if you play perfectly with no mistakes or discreet play. Making mistakes in this game can be disastrous.
I played with the person Eliot was referring to. I hope he doesn't see this but I highly suspect he was making mistakes when he didn't have a clear view of the hole card. Lots of times I had to assume a random card because I couldn't see it but he played as if he saw it. He would probably argue my vision was not as good, which is a valid point. I'm pretty sure I caught him making errors in his read of the hole card, but this was about 15 years ago so I don't remember exactly.
"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)
April 13th, 2018 at 11:07:39 AM
permalink
My largest monetary loss was over $9000 in one day with a $200 bet and perfect reads. This dealer was so sloppy, you could see the card from the other end of the casino. This game can be brutal.
DUHHIIIIIIIII HEARD THAT!
April 13th, 2018 at 11:26:19 AM
permalink
Thank you Wizard. I'm trying to reverse engineer some of his calculations just to teach myself.
Can anyone tell me how to reach his 0.32% RoR considering his analysis of variance?
Thank you
Can anyone tell me how to reach his 0.32% RoR considering his analysis of variance?
Thank you
April 13th, 2018 at 12:04:03 PM
permalink
Member teliot is pretty good at math. Maybe he can help reverse engineer the calcs:-)
"So as the clock ticked and the day passed, opportunity met preparation, and luck happened." - Maurice Clarett
April 13th, 2018 at 7:42:48 PM
permalink
Quote: IbeatyouracesMy largest monetary loss was over $9000 in one day with a $200 bet and perfect reads. This dealer was so sloppy, you could see the card from the other end of the casino. This game can be brutal.
Last weekend I ran into one of those dealers and had to fold my first 10 hands.
3CP does nasty things like that all the time.
April 13th, 2018 at 7:44:21 PM
permalink
Quote: gamerfreakLast weekend I ran into one of those dealers and had to fold my first 10 hands.
3CP does nasty things like that all the time.
You saw an ace, king or queen 10 times in a row?
DUHHIIIIIIIII HEARD THAT!
April 13th, 2018 at 7:47:50 PM
permalink
Quote: IbeatyouracesYou saw an ace, king or queen 10 times in a row?
I meant I lost the first 10 hands :(
April 13th, 2018 at 7:55:35 PM
permalink
Quote: gamerfreakI meant I lost the first 10 hands :(
It's pretty easy when you see a 4, go blind, dealer reveals Q,4,2 offsuit and you have 2,3,6 offsuit. 🤔
DUHHIIIIIIIII HEARD THAT!
April 16th, 2018 at 11:52:59 AM
permalink
Thanks everyone.
Like I said I'm not really interested in playing the game, I'm just trying to teach myself some combinatorial analysis.
I understand the best way to calculate risk is through simulations of random walks using the parameters in the analysis. I have a good knowledge of Excel and I'm trying to do this but I'm hitting a few bumps. If someone can help me simulate a random walk in this scenario I would really appreciate it and can reward for your time and a job well done.
Thanks
Like I said I'm not really interested in playing the game, I'm just trying to teach myself some combinatorial analysis.
I understand the best way to calculate risk is through simulations of random walks using the parameters in the analysis. I have a good knowledge of Excel and I'm trying to do this but I'm hitting a few bumps. If someone can help me simulate a random walk in this scenario I would really appreciate it and can reward for your time and a job well done.
Thanks
April 16th, 2018 at 1:27:10 PM
permalink
I disagree with your understanding 100%.Quote: gunbjI understand the best way to calculate risk is through simulations of random walks using the parameters in the analysis.
in this case a simple Markov chain can calculate the probability of hitting a target before ruin.
That is what I think holds more weight.
say turn 250 units into at least 500 or ruin (0 or 1 unit left)
easy in Excel as I do
or in R if code is available. (I do both at times now)
I think I will try and post my results.
simulations help to 'see' where one should be about
calculating is
most times
so much easier
Sally
I Heart Vi Hart
April 16th, 2018 at 2:00:00 PM
permalink
I came up with (using the probability distribution in the article)
.0032 as the article mentions also. (for ruin that is)
7144 is the average number of rounds to target or ruin
for just ruin, needs more calculating 1st (well, not that much. need to normalize a vector 1st)
calculation took me about 3 minutes to do in Excel
(I did it twice to check for any errors)
one snap to see
1 in 310 is not that small of a risk of ruin, imho
most think that would never happen to them, only to others!
*****
chock it up to bad luck...
or someone was not following the playing rules 100% of the time
(I caught a team cheating at Baccarat a few years ago as they were NOT hitting their target often enough. seems they were taking profit during play. stupid mistake, I do say so.)
Sally
I will see if I can make this into some R code that can be run in a browser.
that is later
.0032 as the article mentions also. (for ruin that is)
7144 is the average number of rounds to target or ruin
for just ruin, needs more calculating 1st (well, not that much. need to normalize a vector 1st)
calculation took me about 3 minutes to do in Excel
(I did it twice to check for any errors)
one snap to see
1 in 310 is not that small of a risk of ruin, imho
most think that would never happen to them, only to others!
*****
chock it up to bad luck...
or someone was not following the playing rules 100% of the time
(I caught a team cheating at Baccarat a few years ago as they were NOT hitting their target often enough. seems they were taking profit during play. stupid mistake, I do say so.)
Sally
I will see if I can make this into some R code that can be run in a browser.
that is later
I Heart Vi Hart
April 16th, 2018 at 2:31:17 PM
permalink
Having a team use multiple seats at the same table , or an individual playing multiple hands, is less effective than you might first imagine at reducing variance.
All seats at a table where player's hand is compared to dealer's hand share this in common- they are all affected by the quality of the dealer's hand. In 3CP, whenever the dealer has trips or a flush, your team is likely to lose all 7 hands. This is a form of covariance. To reduce variance you would be much better off having 7 team members playing against 7 different dealers.
All seats at a table where player's hand is compared to dealer's hand share this in common- they are all affected by the quality of the dealer's hand. In 3CP, whenever the dealer has trips or a flush, your team is likely to lose all 7 hands. This is a form of covariance. To reduce variance you would be much better off having 7 team members playing against 7 different dealers.
So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.
April 16th, 2018 at 2:34:54 PM
permalink
Thanks very much mustangsally, I will PM you to follow up!
April 16th, 2018 at 3:52:11 PM
permalink
done.Quote: mustangsallySally
I will see if I can make this into some R code that can be run in a browser.
that is later
I have to clean up the code as I took another R code and changed it to 3CP stuffs.
then I can post it online as it works very well in my laptop but I have yet to test it in a browser.
*****
some data.
looks like just having a 300 unit stake would have made for a 1 in 1000 ror.
little things can make a large difference too
example: ror.3cp(500,1000) this calls the function
500=stake and 1000=target
> ror.3cp(250,500)
[1] "3 card poker. Overshoot target possible"
data
stake 250.000000000
p(target) 0.996779348
p(ruin) 0.003220652
avg trials 7144.016356530
[1] 500
[1] 250
> ror.3cp(250,1000)
[1] "3 card poker. Overshoot target possible"
data
stake 250.000000000
p(target) 0.996769199
p(ruin) 0.003230801
avg trials 21453.497953254
[1] 1000
[1] 750
> ror.3cp(300,600)
[1] "3 card poker. Overshoot target possible"
data
stake 300.000000000
p(target) 0.998978076
p(ruin) 0.001021924
avg trials 8608.233020773
[1] 600
[1] 300
> ror.3cp(400,800)
[1] "3 card poker. Overshoot target possible"
data
stake 400.0000000000
p(target) 0.9998974605
p(ruin) 0.0001025395
avg trials 11494.6780778597
[1] 800
[1] 400
> ror.3cp(500,1000)
[1] "3 card poker. Overshoot target possible"
data
stake 500.00000000000
p(target) 0.99998971953
p(ruin) 0.00001028047
avg trials 14367.93929503257
[1] 1000
[1] 500
>
nba tonight!
Sally
Go Angles!! (yes baseball 2)
the R code can be run online here
this was fun!
Last edited by: mustangsally on Apr 16, 2018
I Heart Vi Hart
April 16th, 2018 at 6:38:44 PM
permalink
I don't understand how people are finding these hole card games? Every casino I've been to lately uses an opaque card that covers the bottom. The cards in the machine go onto this opaque card then the cards are lifted up and put in front of the dealer. I never see anything even from sloppy dealers. Am I missing something?
April 16th, 2018 at 6:53:32 PM
permalink
Quote: alphastormI don't understand how people are finding these hole card games? Every casino I've been to lately uses an opaque card that covers the bottom. The cards in the machine go onto this opaque card then the cards are lifted up and put in front of the dealer. I never see anything even from sloppy dealers. Am I missing something?
Leave Vegas.
DUHHIIIIIIIII HEARD THAT!
April 17th, 2018 at 7:22:52 AM
permalink
lol, good luck with that.Quote: IbeatyouracesLeave Vegas.
Playing it correctly means you've already won.
April 17th, 2018 at 7:25:34 AM
permalink
Quote: Romeslol, good luck with that.
With leaving Vegas or finding a game outside Vegas?
DUHHIIIIIIIII HEARD THAT!
April 17th, 2018 at 9:40:01 PM
permalink
Quote: mustangsally
I just have one question for you... What are those!!
100% risk of ruin
April 19th, 2018 at 2:23:58 PM
permalink
Can anyone tell me anything about the differences between the Wizard's and Eliot Jacobson's combinatorial analyses? Again, sorry if I can't post links yet, the Wizard's hole-card analysis is easy to find and called "Three Card Poker Flashing Dealer". The numbers of the events are pretty wildly different, yet the probabilities are the same, how is that possible?
PS: I must be missing something to ask this question but... what's the situation in which you can win 7 units at Three Card Poker? I thought the biggest paying hand was a 5:1 on the ante bonus for a straight flush plus a 1:1 on the play bet, resulting in a 6 unit win?
Thank you
PS: I must be missing something to ask this question but... what's the situation in which you can win 7 units at Three Card Poker? I thought the biggest paying hand was a 5:1 on the ante bonus for a straight flush plus a 1:1 on the play bet, resulting in a 6 unit win?
Thank you
April 19th, 2018 at 2:26:36 PM
permalink
Quote: gunbjPS: I must be missing something to ask this question but... what's the situation in which you can win 7 units at Three Card Poker? I thought the biggest paying hand was a 5:1 on the ante bonus for a straight flush plus a 1:1 on the play bet, resulting in a 6 unit win?
Thank you
Winning with a straight flush and the dealer qualifying on the 1/4/5 ante bonus schedule.
You get 1 to 1 on BOTH the ante and play bet PLUS another 5 to 1 on the ante bonus for a total win of 7 units.
DUHHIIIIIIIII HEARD THAT!
April 19th, 2018 at 2:55:52 PM
permalink
Quote: IbeatyouracesWinning with a straight flush and the dealer qualifying on the 1/4/5 ante bonus schedule.
You get 1 to 1 on BOTH the ante and play bet PLUS another 5 to 1 on the ante bonus for a total win of 7 units.
Makes perfect sense, thank you
April 19th, 2018 at 2:57:13 PM
permalink
Quote: gunbjMakes perfect sense, thank you
You're welcome. You were just forgetting to pay the ante. 😉
DUHHIIIIIIIII HEARD THAT!
April 19th, 2018 at 3:08:43 PM
permalink
Quote: IbeatyouracesYou're welcome. You were just forgetting to pay the ante. 😉
Years ago had many arguments with dealers ,floors,pits and even shift bosses that it was a bonus so with a straight for example you would get paid 2 units on the ante,1 for the pay and 1 for the bonus.
The mountain is tall but grass grows on top of the mountain.
April 19th, 2018 at 5:30:47 PM
permalink
If our straight flush is beaten by a dealer's higher straight flush, do we still get the straight flush ante bonus? (Same question for a straight).
April 19th, 2018 at 5:36:07 PM
permalink
Quote: EdpokernutIf our straight flush is beaten by a dealer's higher straight flush, do we still get the straight flush ante bonus? (Same question for a straight).
Yes. But sometimes you have to fight for them.
