Single Deck Ace Adv
- Thread starter stophon
- Start date
iCountNTrack
Well-Known Member
I am not sure about being in the hole card, i have to work out the math. But for the Ace being the dealer's upcard it is about -38%, and if it the players first card it is about +52%.
Dealer up card = ace EV=-38.91% (don't know hole card)stophon said:
Dealer up card = ace EV=-33.91% (know hole card)
Dealer up card = ace, player first card = ace EV=+10.29% (don't know hole card)
Dealer up card = ace, player first card = ace EV=+13.28% (know hole card)
Player first card = ace EV=+52.64% (won't know hole card after up card is dealt)
Player first card = ace EV+62.78% (will know hole card after up card is dealt)
Calculated using total dependent basic strategy, no resplit, 1 card to split aces, H17, NDAS, double any 2 cards, no surrender, bj pays 3:2, full peek
Pro21
Well-Known Member
CAA doesn't have a chapter on ace tracking. It does have charts for the value of all dealer cards for 1,2,& 6 decks with various rules, both as up card or hole card. It also has the value of player bet cards. These charts are extremely valuable to the players that use them but are just one of many examples of pages that many readers won't even see. CAA is a very valuable resource and well worth the money to any professional ready to move "beyond counting."
If dealer has an ace in the hole and you don't know his up card it is the same as knowing the dealer's up card is an ace and not knowing his hole card.stophon said:
But it doesn't make any sense that you would know the dealer's hole card and not his up card, so if his hole card is an ace you would also know his up card.
Code:
[u]Up Hole[/u]
T A -96.33%
9 A -45.93%
8 A -13.44%
7 A +10.17%
6 A +2.50%
5 A +11.07%
4 A +4.68%
3 A -1.03%
2 A -5.43%
A A -12.78%No, because you'll double down a lot more and make riskier plays when you don't know his hole card and it is ace. And wouldn't you always know the dealer's upcard? I'm trying to use these numbers to calculate E.V. for ace tracking.k_c said:
Now if your 12-15% sure the dealer's hole card is an ace, does it affect basic strategy? Does the effect make enough of a difference in EV to bother learning about?
My point is that if you only know 1 of the 2 dealer cards it doesn't matter which one it is unless insurance or surrender is being considered.stophon said:
If you don't know dealer's hole card then it's simply a normal situation: you know the up card and normally a player knows his first 2 cards as well, so that info is taken into consideration when computing EV and determining strategy.stophon said:
If player has more or less info than this, best strategy can possibly be altered. For example if player has 6-5 v 9, normal strategy is to double. Normally dealer's hole card will be an (ace,2,3,4,7,8) 4/49 of the time, (5,6,9) 3/49 of the time, and (10) 16/49 of the time assuming deal is from a full single deck.
If player thinks he knows an ace will be in the hole 8/49 of the time rather than 4/49 the probabilities are altered to:
(ace) 8/49, (2,3,4,7,8) 4/49*41/45, (5,6,8) 3/49*41/45, (10) 16/49*41/45
These probabilities are different than would normally be expected due to the assumption that an ace in the hole is more probable than normal and could possibly yield a different strategy.
EV for ace in the hole is -54.17% (doubling) & -9.52% (hitting) and its weight=8/49
EV for 2 in the hole is -8.20% (doubling) & -3.89% (hitting) and its weight=4/49*41/45
EV for each of the other cards in the hole is whatever. I'm not going to list them.
Once all EVs are known simply sum EV*weight for each possible up card and play according to whatever strategy that the calculation dictates.
All that's being done is to use the professed additional knowledge to make a more informed decision. Whether it's worth it or not is up to you. Knowing the hole card 100% of the time EV=~+10%. Knowing only the up card EV=~-.5%. Having knowledge that the probability of a given card being in the hole is different than what would normally be expected is somewhere in between.
My opinion is that if it's not simple enough, it's not worth it.
Last edited:
Pro21
Well-Known Member
I am trying to imagine a scenario where you are tracking aces and you reach a hand where you think the dealer has a 12-15% chance of having an ace in the hole.stophon said:
But to answer your question - yes that would effect your playing decision.
BUT don't worry about this. You are getting caught up in the minutiae. You make big money from betting when you know you are catching an ace. The rest of the hands you are betting small and these small deviations mean almost nothing to your overall win rate.
BTW - the title of your thread is "single deck ace advantage." Are you planning to play this on 6-5 games? There are very few single decks that still pay 3-2 and I can't imagine any of them that I know taking much action. You do realize the ace is not worth as much in a 6-5 game?
Last edited:
Results from sim:Pro21 said:
Code:
Riffle performed using data obtained by Anthony Curtis
dealer randomly strips 6-10 cards
single deck
riffle-riffle-strip-riffle-cut
data for when key card is one of first 39 cards out of shuffled deck (key card can be card 1-39 before shuffle)
[U]cards separating key card and ace
[/U]
0 3.63%
1 5.96%
2 6.34%
3 6.08%
4 5.63%
5 4.84%
6 4.19%
7 3.23%
8 2.53%
9 1.88%
10 1.43%
11 1.03%
12 0.76%
Sum 47.54%
Key before Ace 24.91%100-24.91-3.63-6.34=65.12%
5.96/65.12=9.16%
There are 51 other cards that the hole card could be. 100-9.16=90.84%. 90.84/51 = 1.78%. 3 of those are aces. 1.78*3=5.34%
9.16+5.34=14.5% chance dealer hole card is ace
I don't plan to play 6:5. I compute the ace being worth 43.6% in 6:5, barely enough to gain an advantage if the ace could land on the dealer or the player. Though if they shuffled less in 6:5 it could be quite a money making opportunity...
Thanks for the post k_c, was interesting to read.
I'm focusing on advantages in the 1-5% range. I figure you can count cards and then if you see the key card you can add +3 to your count if you compute that the ace coming gives you a 1.5% advantage. The extra EV would be huge.
Last edited:
You're welcome, but it's even more complicated......stophon said:
After a little more thought, more would still have to be done in order to compute EVs based on an assumed probability of a given card appearing, in this case an ace. You would have to continously update all probabilites based on differing conditions.
For the example I gave (6-5 v 9), dealer could have anything in the hole but we are assuming the probability of an ace is greater than it would normally be. Whenever dealer's hole card is not an ace the next card the player draws would be even more likely to be an ace than originally assumed. In other words all probabilites would have to be continuously updated based on each specific possibility based upon the original assumption. It could be done but it would be a royal pain in the neck.
This is the reason I abandoned the idea of doing a combinatorial analyzer based upon a count. A counter is basically ignorant of anything except the imbalance of high and low card groups and penetration. It is possible to compute the probability of each rank based on this limited information. However to compute what happens after that would require that each specific possibility be addressed, requiring continuous updates of probabilities based upon the original information. I went ahead and did this to compute insurance EV for Hi-Lo and KO. Theoretically it could be done to compute other EV, but it's not practical so I just settle for computing actual shoe compositions.
iCountNTrack
Well-Known Member
You mean total strangers, your teammates could be definitely helpfulPro21 said:
iCountNTrack
Well-Known Member
I guess we were talking about different things, the idea i had in mind is that you would work with your teammates to steer the ace to a spot where you have the big bets.Pro21 said:
iCountNTrack
Well-Known Member
Mis-communication from my part, we are talking about two different techniques, sequencing vs cut cardPro21 said:
Because you will not get to make as many bets as the key and the ace could come out in the same round more often. Very good point. Was this the reason why snyder decided it is a bad idea to spread to multiple hands to catch an ace? (mess up your other keys if in single deck)Pro21 said:
For the shuffle data I described above, if the key were to come out as the last card of the round before, I would play one hand with an increased bet then if the ace still did not show I would spread to 2 or 3 hands (depending on how many cards used in first round). I think some of the best betting oppertunities appear when a few cards have been played after the key.
Last edited: