Mow21
Joined: Feb 9, 2015
• Posts: 28
February 18th, 2016 at 4:04:35 PM permalink
It is indeed a max of \$25. If only it were more it could offer a decent return.
teliot
Joined: Oct 19, 2009
• Posts: 2210
February 18th, 2016 at 4:14:57 PM permalink
Quote: Mow21

It is indeed a max of \$25. If only it were more it could offer a decent return.

Okay ... time to do the AP analysis ... okay, it looks like a max bet of \$25 at DW has about the same earning power for the counter as a max bet of \$500 as an ordinary hi-lo card counter. Details soon.
Last edited by: teliot on Feb 18, 2016
Poetry website: www.totallydisconnected.com
teliot
Joined: Oct 19, 2009
• Posts: 2210
February 18th, 2016 at 4:57:21 PM permalink
Quote: Mow21

It is indeed a max of \$25. If only it were more it could offer a decent return.

Here are the details. Using the count system (0,-10,1,1,1,1,1,1,1,0,1,1,1) - in the order A, 2, 3, 4, ... , J, Q, K - in a six-deck shoe game with the cut card at 260 cards (1 deck from the end):

Trigger true count = + 3
EV per bet = 14.39%
Bet frequency = 30.23%
Units won per 100 hands = 4.35.
With a \$25 unit, the counter earns about \$108 per 100 hands.

To put this in perspective, blackjack hi-lo card counting using the illustrious and a 1-8 spread wins about 0.22 max bets per 100 hands. So, the ordinary card counter would need to have a maximum bet of roughly \$500 to equal the earning power from betting \$25 on Deuces Wild. In other words, DW is about 20 times as good as ordinary blackjack card counting.

I'll double check this, then I'm posting it to APHeat. In the mean time, someone's going to get a spanking.
Poetry website: www.totallydisconnected.com
beachbumbabs
Joined: May 21, 2013
• Posts: 14232
February 18th, 2016 at 5:08:21 PM permalink
Quote: teliot

Here are the details. Using the count system (0,-10,1,1,1,1,1,1,1,0,1,1,1) - in the order A, 2, 3, 4, ... , J, Q, K - in a six-deck shoe game with the cut card at 260 cards (1 deck from the end):

Trigger true count = + 3
EV per bet = 14.39%
Bet frequency = 30.23%
Units won per 100 hands = 4.35.
With a \$25 unit, the counter earns about \$108 per 100 hands.

To put this in perspective, blackjack hi-lo card counting using the illustrious and a 1-8 spread wins about 0.22 max bets per 100 hands. So, the ordinary card counter would need to have a maximum bet of roughly \$500 to equal the earning power from betting \$25 on Deuces Wild. In other words, DW is about 20 times as good as ordinary blackjack card counting.

I'll double check this, then I'm posting it to APHeat. In the mean time, someone's going to get a spanking.

Ooh! Is there an invited audience? (kidding). Appreciate the work and the info, teliot.
If the House lost every hand, they wouldn't deal the game.
Mow21
Joined: Feb 9, 2015
• Posts: 28
February 18th, 2016 at 5:27:00 PM permalink
Quote: teliot

In other words, DW is about 20 times as good as ordinary blackjack card counting.

That's incredible. Thank you for your work!
teliot
Joined: Oct 19, 2009
• Posts: 2210
February 18th, 2016 at 5:39:47 PM permalink
Quote: Mow21

That's incredible. Thank you for your work!

The desirability index is 30.2. If you know what that number means, then you know just how good this is. Enjoy. Then tell us the story (or write it up for my blog).

In this day, with all of the information out there, and all of the efforts I have personally made to help educate the industry, there are still those who believe the moon landing was a hoax.
Poetry website: www.totallydisconnected.com
Mow21
Joined: Feb 9, 2015
• Posts: 28
February 18th, 2016 at 6:19:07 PM permalink
Thanks again for the math on the DW side bet! Before I start memorizing the values for each card I wanted to ask a couple of quick questions.

I noticed that you had A as 0 even though the payout for 2-A is 1-1. There was some confusion when I started the thread that 2-A was 11-1 but after seeing the table today (and posting it to the thread earlier & down below) it is the same as K,Q, and J.

2-2 ............. 20-1
2-10............ 10-1
2-9.............. 9-1
2-8.............. 8-1
2-7.............. 7-1
2-6.............. 6-1
2-5.............. 5-1
2-4.............. 4-1
2-3.............. 3-1
2-A,K,Q,J... 1-1

Would A still be counted as 0 with the above table?

Also, I noticed you said true trigger value +3. Does that mean that immediately when the count for DW goes to +3 I bet the max(\$25)? There is no bet ramp like Hi-Lo?

Thank you for being kind and taking the time to analyze this!
teliot
Joined: Oct 19, 2009
• Posts: 2210
February 18th, 2016 at 6:54:11 PM permalink
Quote: Mow21

I noticed that you had A as 0 even though the payout for 2-A is 1-1. There was some confusion when I started the thread that 2-A was 11-1 but after seeing the table today (and posting it to the thread earlier & down below) it is the same as K,Q, and J.

2-2 ............. 20-1
2-10............ 10-1
2-9.............. 9-1
2-8.............. 8-1
2-7.............. 7-1
2-6.............. 6-1
2-5.............. 5-1
2-4.............. 4-1
2-3.............. 3-1
2-A,K,Q,J... 1-1

Would A still be counted as 0 with the above table?

This is a new pay table, hence a new count system. First of all, the house edge for this new pay table is slightly higher at 7.3213%. As for the count system, the following is quite strong:

3, 4, 5, 6, J, Q, K, A = +1
2 = -8
7, 8, 9, 10 = 0

Trigger true count = +3
Average edge = 16.81%
Bet frequency = 26.96%
Units won per 100 hands = 4.53

The win rate is better -- mainly because keeping track of 2's now correlates to the premium hand 2/2 that pays 20-to-1. (The game developer should have designed it to pay 22-to-1, by the way, house edge = 6.18%, don't you think?).

Quote:

Also, I noticed you said true trigger value +3. Does that mean that immediately when the count for DW goes to +3 I bet the max(\$25)? There is no bet ramp like Hi-Lo?

Either bet \$25 or don't bet. No ramp.
Poetry website: www.totallydisconnected.com
Mow21
Joined: Feb 9, 2015
• Posts: 28
February 18th, 2016 at 7:26:17 PM permalink
Awesome, thanks!
teliot
Joined: Oct 19, 2009