Spanish 21 book

FLASH1296

Well-Known Member

An excellent count particularly for shoe games.

Kudos to the inestimable redoubtable Automatic Monkey

(Dead link: http://www.mediafire.com/?4yyznoa2w2n)
 
FLASH1296 said:

An excellent count particularly for shoe games.

Kudos to the inestimable redoubtable Automatic Monkey

(Dead link: http://www.mediafire.com/?4yyznoa2w2n)
I dig it, I dig it... But I'm still partial to the AOII. Hi-OptII coming in a close second. I find side counting the aces very easy, since I treat it like a neutral in my main count, and simply use a letter tag. I also combine this with a word, and every 10 aces shift my big toes: Left foot big toe up on top of my next toe = +10, Right foot big T doing the same = +20, and none at all is either 0 or +30. Then the letter tags are like this: Ask Bob Call Dave Eat Food Grow Hair In Jail. So, three aces out, with a running count of +2, I say "2, Call" in my head. I remember the word better than the letter, and the next word to finish the combination makes it more readily available should a large gap of aces occur. I remember that I have to "Call Dave." Then once I say "Jail," my toe position shifts, and I start back again at "Ask". I find this much better than using 30 letters and much better than shifting both feet in ridiculous ways (although shifting chewing gum is also a good idea). I realize that my gain is marginal in a shoe, but if the pen is good, it's no longer marginal but significant. Also I don't want to learn a seperate count for hand held than shoes. But even more than EV is the reduction in Standard deviation (as pointed out by Wong in PBJ) for the higher level count that attracts me most. So, for SP21 I'm liking the revised AOII. All I have to do shift the 7 and 8 values, and adjust aces slightly differently. Sweet.
 
Pyrrhonian said:
I dig it, I dig it... But I'm still partial to the AOII. Hi-OptII coming in a close second. I find side counting the aces very easy, since I treat it like a neutral in my main count, and simply use a letter tag. I also combine this with a word, and every 10 aces shift my big toes: Left foot big toe up on top of my next toe = +10, Right foot big T doing the same = +20, and none at all is either 0 or +30. Then the letter tags are like this: Ask Bob Call Dave Eat Food Grow Hair In Jail. So, three aces out, with a running count of +2, I say "2, Call" in my head. I remember the word better than the letter, and the next word to finish the combination makes it more readily available should a large gap of aces occur. I remember that I have to "Call Dave." Then once I say "Jail," my toe position shifts, and I start back again at "Ask". I find this much better than using 30 letters and much better than shifting both feet in ridiculous ways (although shifting chewing gum is also a good idea). I realize that my gain is marginal in a shoe, but if the pen is good, it's no longer marginal but significant. Also I don't want to learn a seperate count for hand held than shoes. But even more than EV is the reduction in Standard deviation (as pointed out by Wong in PBJ) for the higher level count that attracts me most. So, for SP21 I'm liking the revised AOII. All I have to do shift the 7 and 8 values, and adjust aces slightly differently. Sweet.
Back when I used HO2 with the ace sidecount my mnemonic was "DHL Package Transit Xpress" D=4, H=8, L=12, P=16, T=20, X=24. That represents the expected number of aces per deck, and it made it easier to determine how many extra/short aces were there to be divided by the remaining number of decks.

SP21 has an intrinsically lower standard deviation than BJ, because of the double down rescue rule and the fact that we are hitting a lot more stiffs, resulting in more pushes.
 
Automatic Monkey said:
Back when I used HO2 with the ace sidecount my mnemonic was "DHL Package Transit Xpress" D=4, H=8, L=12, P=16, T=20, X=24. That represents the expected number of aces per deck, and it made it easier to determine how many extra/short aces were there to be divided by the remaining number of decks.

SP21 has an intrinsically lower standard deviation than BJ, because of the double down rescue rule and the fact that we are hitting a lot more stiffs, resulting in more pushes.
I like it. Succinct. The idea of 4's makes sense. Good to know about the lower SD. What made you stop using the count, and which do you use now?
 

Kasi

Well-Known Member
Automatic Monkey said:
SP21 has an intrinsically lower standard deviation than BJ, because of the double down rescue rule and the fact that we are hitting a lot more stiffs, resulting in more pushes.
FWIW, on page 48 she says the SD for BJ is slightly less than the SD for North American SP21 due to doubling on any number of cards and splitting more often.

And, when re-doubling allowed, higher than that lol.

Maybe her numbers do not include some of the multi-card bonus pay-outs but yours do or something?

No big deal, just wondering just because SP21 is my favorite game, in AC and Canada anyway, and has been for many years - long before her book came out lol.

Scratch Canada though. Like I'm gonna pay $100 for a passport to cross the border. Plus the bridge toll and still have to deal with the difference between buy and sell currency exchange rates while being required to actually buy a drink at the table and dealing with $25 mins at Fallsview. Not to mention why they always ask if I have any tobacco products when I'm I'm smoking a cig and have a pack in my pocket lol. Then they ask if I have alcohol like they did once when I went on a bus trip and the guy came on the bus. I showed him my flask of booze I had on me and he said that didn't matter which made me wonder why he asked in the first place lol. It was a little embarrassing since, apparently, I was apparently the only one on the bus bringing alcohol across.

That's alot of -EV for a mostly flat-betting, semi-progressive voodoo-betting, dependent on the cost of alcohol in his EV calcs when determining "break-even" point (goal for the trip), to overcome :grin:
 
Kasi said:
FWIW, on page 48 she says the SD for BJ is slightly less than the SD for North American SP21 due to doubling on any number of cards and splitting more often.

And, when re-doubling allowed, higher than that lol.

Maybe her numbers do not include some of the multi-card bonus pay-outs but yours do or something?

No big deal, just wondering just because SP21 is my favorite game, in AC and Canada anyway, and has been for many years - long before her book came out lol...
For the S17 game I calculated it to be quite a bit less. You have the doubling on any number of cards, but you actually do less doubling because most of the soft doubles and a few of the hard doubles go away in SP21. Soft doubles are high risk plays. And certain tough plays like DD 11 vs. A you don't do on more than 2 cards (looking for the 5-card bonus.) Splitting you do a bit more of, but being able to draw to your split aces more than makes up for it!

I haven't done the redoubling sim but I suspect the SD is much higher.

The multi-card bonuses in themselves add to the SD, just as any non 1:1 payout in any game does. But your return is 25% less dependent on 3:2 naturals, which reduces the SD.
 

N&B

Well-Known Member
Well, I stayed outta this thread long enuff. I like sp21, but NOT the way practiced locally. Years back I could hit split Aces, and now I can't. This is WAY worse than BJ21, and so I'm not a regular sp21 player.

My intuitive side did get almost all plays correct w/o a strat card. The ones I got wrong were DD 9 vs. 6 (i didn't DD on 9) no DD A4vs 6, and A5 vs. 5. Those last 2 i DD'd. And I used to chase 678 suited at all times, and unsuited against a 4 if I had 6-8. IIRC 6-cards unbusted wins, so 5 cards soft always hits (Now I see its 7-cards so no more of that either!). I figure what was once a -0.4 game is like -0.85. Absolutely skankypottamus compared to 6D 70%P S17 DAS LS regular 21 at -0.35.

But I digress, Kat has done a wonderful job in the book. For those with real sp 21 with HSA good luck. Its certainly a viable game.
 

FLASH1296

Well-Known Member
Forewarned is forearmed.

The "cost" of being unable to [resplit/hit/double on] split Aces is (unintuitively) high.

Seven miles from Mohegan Sun, is Foxwoods, where this BAD situation does not exist.

Forewarned is forearmed.
 

N&B

Well-Known Member
Thanx FLASH, I gave up on BOTH Houses of Gaming figuring if Mohegan did it so did FW. I really didn't want to plunk down Benjamins to find out.

I believe No HSA for S17 sp21 costs -0.32%
 
Illustrious Indices for Secret Monkey count?

Automatic Monkey said:
Interesting thoughts here. Of the counts you list, the best will be D) because it doesn't short the ace.

Ace-neutral is a whole different kind of count and your initial approach is a good one. Using -4 for the ace will be best, as the EoR of the ace is 2.2-2.7 times that of the 10. Remember the value of the ace is unstable and non-linear and its EoR will go down in a shoe that is depleted of 10's.

You need to have CVData 4.0 and do a bit of experimentation to get your sim and indices. The multiple bonuses screw up CVIndex so set up a configuration that just has the bonus for an automatic win on 21 and use the composition dependent index generator to get your numbers.

The Secret Monkey Count is 1,1,1,1,1,0,0,0,-1,-2 and it works just fine with Kat's index numbers if you add 4 to them.
In what order is the most important indices with the Secret Monkey Count?
If you were to play with just the Spanish illustrious 18 indices (or actually 20)...Does it change cause you add 4 to each of Kat's indices? Thanks in advance for all your input in this forum AM)))
 
Matador AP said:
In what order is the most important indices with the Secret Monkey Count?
If you were to play with just the Spanish illustrious 18 indices (or actually 20)...Does it change cause you add 4 to each of Kat's indices? Thanks in advance for all your input in this forum AM)))
Yes, use whatever Kat would use, just add 4 and you will be close enough. The most important indices are surrendering 16 vs. 10 and A, 13 vs. 5, 14 vs. 3.
 
Hi again! I just need to clarify two issues with any takers:

a) On page 52 of her book, Katarina states that a 7 is 20% more powerful in Pontoon/Spanish 21 compared to BJ. The next page of her book is a chart showing the data of EOR (effect of removal) and we see that the EOR% of 7 in Pontoon is 0.04 whereas the EOR% of BJ is 0.24%. I thought that the higher EOR makes the number more powerful. On the same page and chart we see the EOR% of 2-6 in Pontoon is a bit lower compared to the EOR% of 2-6 for BJ and she states the small values of 2-6 are not as harmful to the Pontoon player compared to BJ - this makes sense because we can draw to 5 or more card 21 and get pay bonuses. However the statement about 7 is a bit conflicting to me - can anyone clarify?

b) In Basic Strategy, there are differences between how you play the following hands in Spanish 21 and Pontoon:

A,A vs A and A,A vs 10.

In Pontoon, the ace is more threatening compared to the ace in H17 Spanish 21 US games. Even though the majority of Australian casinos dont take ENHC when the dealer gets a natural, most will take BB+1 and OBBO. Pontoon is played with the NHC whereas Spanish 21 is played with the hole card.

In Pontoon, you should hit A,A vs A whereas with Spanish 21 because the dealer finds out that he doesnt have a natural at the start of the round, you should split and this move is considered offensive in both BJ and Pontoon even if you cannot draw to split aces. Also in Pontoon, you should hit your A,A vs 10 if BB+1 applies.

Is that a correct statment?

I have read more about Pontoon and can now make a comparision between Spanish 21 and Pontoon. I can now state that that Pontoon/Sp21 is better than BJ but a larger bankroll is required. A more larger bankroll is required if you play the double-double games as offered in the state of Washington. With Pontoon and Spanish 21, variance will be much larger than BJ because there is double/splitting, the option to forfeit after a double and surrendering on weak hands at high counts. Because S.D = squared root of variance, since variance in Pontoon is larger than BJ, standard deviation will be ONLY slightly higher compared to BJ, thus a slightly higher bankroll is required. Instead of having at least 2,000 units to minimize RoR, you probably need 2,300 units. I wish I was a mathematician to give you the exact number of units. In the double-double game, you will need more than the usual bankroll.

Pontoon is better than Spanish 21 because its played with a NHC. One in every twelve hand, you will draw 21 and win automatically whereas with Pontoon, if the dealer gets a natural at the start of the round, then unless you have a natural, you will lose. In Pontoon, the dealer's hole card will be given after all the people on table has made a playing decision. Even though Pontoon doesnt allow soft doubling (well it does but if you double A is counted as 1 and not as 1 or 11) - but soft doubling is not worth that much in the long run. Not doubling on 10 or 11 will be costly to the BS players and counters alike, but if you dont soft double even on 5 and 6, it won't cost you that much in the long run. Newb99 had a chart which illustrated the penalty for not doubling at various true counts. Also with soft doubling, you are still an underdog to win and you would never soft double against the dealer upcard of 7 or above. Therefore, you wouldnt forfeit after soft doubling generally speaking.

http://www.blackjackinfo.com/bb/showthread.php?t=15719

The only difference is that with Pontoon, you can only surrender on ten and ace whereas with Spanish 21, you can surrender on any dealer's upcard. However, aces are far more powerful and stronger than any other number (incl. 10) in Pontoon compared to Spanish 21. Both games offer late surrender NOT unconditional early surrender. At high counts, Spanish 21 is superior compared to Pontoon because as the count increases, you should surrender more often but I think aces and to a lesser extent ten are quite powerful.

Therefore I conclude at the present time, that the double-double game is better than the AU version of Pontoon which is better than the North American S17/H17 games.
 

FLASH1296

Well-Known Member
Blackjackstudent,

You said: "A more larger bankroll is required if you play the double-double games as offered in the state of Washington."

I may be mistaken, but to the best of my knowledge the North American Double-Double Span21 game, often offers the variant in question, but there is an additional factor that comes into play. It is significant.

There are quite a few little casinos in Washington State (and "Wildhorse" in Oregon) that offer the variant. BUT - the casinos are mostly very small and offer low stakes and have a CAP on the bet - the Table Maximum wager. This means that if your bet is over ½ of the Table Maximum, after an initial double you CANNOT redouble because the bet has already been capped. To get the full "redouble" your wager would need to be equal to or less than ¼ of the Table maximum Bet.

This would not be a problem if the Table Maximum was high enough. The casinos I am referring to have table maximums of $100, $200, $300 and $500.
 
Even with the redouble cap, you'll still want a bigger bankroll to play Spanish. When I consider most of my Spanish 21 sessions, the result is often hinges on the one round where I started with two hands at my max, but after all the doubles and splits I ended up with six or more max bets on the table which will win or lose based on whether or not the dealer busts their hand.

In blackjack, there are situations where you'll get six or more bets on the table (starting with two hands), but it doesn't happen nearly as often as it does in Spanish. Redoubling is only one factor. Also consider that you can double-down on 3+ cards, you can resplit aces up to four hands, and you can hit (and double-down) after splitting aces. The 5+ card 21, 6-7-8, and 7-7-7 bonus hands also contribute to increased variance. These bonus payouts are figured into your edge and you play a more aggressive hitting strategy in an effort to win a few of them, but their frequency is somewhat low. These are all favorable rules, but they add to your variance.

My suggestion is to size the Spanish 21 bets to be roughly 3/4 the size of your BJ bet with the same edge. For example, if you're betting $100 with a 1% edge in BJ, only bet $75 with a 1% edge in Spanish.
 
EmeraldCityBJ said:
Even with the redouble cap, you'll still want a bigger bankroll to play Spanish. When I consider most of my Spanish 21 sessions, the result is often hinges on the one round where I started with two hands at my max, but after all the doubles and splits I ended up with six or more max bets on the table which will win or lose based on whether or not the dealer busts their hand.

In blackjack, there are situations where you'll get six or more bets on the table (starting with two hands), but it doesn't happen nearly as often as it does in Spanish. Redoubling is only one factor. Also consider that you can double-down on 3+ cards, you can resplit aces up to four hands, and you can hit (and double-down) after splitting aces. The 5+ card 21, 6-7-8, and 7-7-7 bonus hands also contribute to increased variance. These bonus payouts are figured into your edge and you play a more aggressive hitting strategy in an effort to win a few of them, but their frequency is somewhat low. These are all favorable rules, but they add to your variance.

My suggestion is to size the Spanish 21 bets to be roughly 3/4 the size of your BJ bet with the same edge. For example, if you're betting $100 with a 1% edge in BJ, only bet $75 with a 1% edge in Spanish.
Got to differ with you a little bit on this. My sims show that your standard deviation is quite a bit lower in SP21 (the S17 version, not the redouble version) the biggest contributor being the double down rescue. You also do a lot more hitting so there are more pushes. The ability to hit split aces lowers SD a bit, and almost all of your soft doubles are of the low variance variety.

Not to discount the need for a large bankroll when playing SP21! But my SP21 bets (again, S17) are a bit higher than the equivalent for BJ.
 
Top