Single Deck Ace Adv

[Pro21]

I think some of the best betting oppertunities appear when a few cards have been played after the key.

It depends on how many times they riffle the cards.

 

[Automatic Monkey]

Mis-communication from my part, we are talking about two different techniques, sequencing vs cut card

There are sequencing techniques that call for some steering too, but for "pure" sequencing all you are doing is predicting an ace and raising your bet accordingly.

Also to clarify something else- sometimes in a pure sequencing game you want other people at the table, even if they are civilians. Ace prediction is inexact, and on the times I bet big and miss the ace, I sure don't want the dealer to get it, so having a buffer of civilians between your first base spot and the dealer's first card is very desirable. All depending on the exact nature of the game, of course.

 

[stophon]

Also to clarify something else- sometimes in a pure sequencing game you want other people at the table, even if they are civilians. Ace prediction is inexact, and on the times I bet big and miss the ace, I sure don't want the dealer to get it, so having a buffer of civilians between your first base spot and the dealer's first card is very desirable. All depending on the exact nature of the game, of course.

Are you really ever gonna know that the ace will be in the next 3 cards? In most shuffles? Even with a bare minimum of two riffles no strips no cuts the ace is likely to fall in six or seven spots.

I think you would lose more money based on the fact that you will lose betting oppertunities.

So why is everyone so secretive about ace tracking? I've had to look back at forum archives to confirm what I've discovered myself as no one likes to answer questions. Is the topic really that sensitive??

 

[Pro21]

In 2 riffles there should be 3 cards between the key card and ace.

Stophon, I suggest you get some decks and start dealing yourself some hands. Also, the shuffle data in Wong's book is nothing like the data we gathered. Do your own shuffling and deal yourself some hands.

 

[stophon]

In 2 riffles there should be 3 cards between the key card and ace.

Stophon, I suggest you get some decks and start dealing yourself some hands. Also, the shuffle data in Wong's book is nothing like the data we gathered. Do your own shuffling and deal yourself some hands.

In 2 perfect riffles! Most dealers drop lots of 2 card clumps and a couple 3's, this really skews that data. However if you are dealing with a fine riffle I agree that 3 cards should be the mean. Once you get to a third riffle this mean tends to flatten out over a range of cards I find.

What data did you gather? How does it differ? PM if you don't want to say.

I have shuffled decks myself and seen how many cards I am likely to interleave. My own riffle isn't far off from most studies I have seen.

Wong's data is slightly different than the data generated by curtis's study but not too far off. Epstein conducted a larger study which points towards a finer riffle. I believe his results are a result of dealers being extra careful when shuffling b/c they are being tested but if you believe you are dealing with a very fine riffler you can use his results. I believe most riffles drop one card around 3/4 of the time, 2 cards around 1/5 and 3 cards most of the rest of the time. I computed that wong's dealer probably dropped 1 card around 71% of the time, 2 cards 16%, 3 cards 6%, 4 cards 4% and 5 cards 3%. That data generates numbers very close to wong's. Why is that shuffle unusual?

same sim details as before except
using epstein shuffle data (drop 1 card - 80%, 2 cards - 18% and 3 cards - 2%)
riffle-riffle

cards separating ace and key

0	3.34%
1	12.98%
2	19.87%
3	32.10%
4	16.86%
5	8.60%
6	3.90%
7	1.48%
8	0.53%
9	0.18%
10	0.04%
11	0.01%
12	0.00%

There is 2 fine riffles and while the mean is 3 cards it is far from what will occur the majority of the time!

 

[stophon]

I've never seen any data on how the count effects the advantage the player gets when he is dealt an ace so I thought I would post it.

100,000,000 hands

count 	adv
-4	49.56%
-3	49.70%
-2	50.41%
-1	51.60%
0	52.66%
1	53.66%
2	54.72%
3	55.37%
4	56.04%
5	56.77%
6	57.86%
7	58.82%
8	59.30%
9	60.54%
10	61.95%
11	61.38%
Overall	52.72%

So when the player is dealt an ace, each ascending count gives him about a .8% adv increase.

Dealer upcard ace
Only 10,000,000 hands here so the numbers have a little standard error to them but the trend is evident

adv            count
-36.80%        -4
-37.19%
-37.20%
-37.21%
-37.47%         0
-37.51%
-37.91%
-37.73%
-38.11%
-38.36%
-38.28%
-38.63%
-38.15%
-39.05%
-39.31%
-38.69%      11

when dealer dealt a hole card ace
also only 10,000,000 hands here but the trend is both very surprising and evident

adv              count
-41.60%          -4
-41.00%
-40.32%
-39.59%
-39.20%          0
-38.43%
-38.51%
-37.69%
-37.09%
-36.93%
-36.94%
-36.21%
-36.16%
-36.94%
-35.98%
-36.76%        11

Higher counts actually hurt the dealer when he is dealt an ace to the hole card. I think this is because in low counts the dealer may have an ace in the hole but a 5 or a 6 showing so the player doubles and splits against it and then gets slaughtered...

With the dealer upcard and hole card adv averaged each higher count hurts the dealer by about .1%

So no surprise, higher counts always favor the player. The overall change is slight compared to the overall advantages though therefore I doubt this information will effect anyone's betting.