Strategy B12

[Evo]

I really like this idea but you lost me

Update on last 5 hand Patterns rich in Player decisions:

Pattern Bet Units

PPPPP bet B 1
PPPPPP bet B 2

BPPPP bet B 1
BPPPPP bet B 2

so I would not use the fist pattern I would need to wait for it to be ended with a B and then make my B bets?


If that is the case with BPPPP would I not need to wiat for a B in this case also or would I bet right away in some posts you seem to say that we will only be betting after a B has ended the run and in other posts it seems that you look over the last 5 hands and if if it is a 4 player to one Banker we will bet so that i the count needs to go 7 player one banker over the last 8 hands for use to be down-3

After reading again and again it seems that you have posted two ways to play what for a run of 4 P or more ended with a B then bet B for at most the next two bets

then you listed patterns based on the last 5 hands
Pattern Bet Units

PPPPP bet B 1
PPPPPP bet B 2

BPPPP bet B 1
BPPPPP bet B 2

PBPPP bet B 1
PBPPPP bet B 2

PPBPP bet B 1
PPBPPP bet B 2

PPPBP bet B 1
PPPBPP bet B 2

PPPPB bet B 1
PPPPBP bet B 2

Total Net Gain = +409.74 units in 27,815 bets. 51.60% Hit Rate



for those patterns you don't need to wait for a B to end the run to start betting you bet right away when you see one of the patterns listed. If I have this right then my question is with the fist pattern in your test that got +409 did you wait for a B to end PPPPP or did you just bet so that a run of 7 P beats you

After reading again I think we are talking about 3 ways of playing
1 is after BPPPP bet B
2 is with the patterns listed above with out waiting for a B to cap runs of long P's
3 waiting for runs of three or more ended with a B then bet B for at most the next two bets

I want to know if your fist test you did was without a cap on the P runs of B and then bet did you test with the stop loss of +1-3 per shoe or was that somehting you added later?
with the pattens you listed was the wait for B and then bet B trigger used for the long runs of P
with the 3rd idea of a run of 3 or more p ended with B then bet B was this tested I tested the first 30 shoes of zumma 600 and it finished +6 units at one point it was up 22 this was with no stop loss per shoe

[Steve]

I really like this idea but you lost me

Update on last 5 hand Patterns rich in Player decisions:

Pattern Bet Units

PPPPP bet B 1
PPPPPP bet B 2

BPPPP bet B 1
BPPPPP bet B 2

so I would not use the fist pattern I would need to wait for it to be ended with a B and then make my B bets?


If that is the case with BPPPP would I not need to wiat for a B in this case also or would I bet right away in some posts you seem to say that we will only be betting after a B has ended the run and in other posts it seems that you look over the last 5 hands and if if it is a 4 player to one Banker we will bet so that i the count needs to go 7 player one banker over the last 8 hands for use to be down-3

After reading again and again it seems that you have posted two ways to play what for a run of 4 P or more ended with a B then bet B for at most the next two bets

then you listed patterns based on the last 5 hands
Pattern Bet Units

PPPPP bet B 1
PPPPPP bet B 2

BPPPP bet B 1
BPPPPP bet B 2

PBPPP bet B 1
PBPPPP bet B 2

PPBPP bet B 1
PPBPPP bet B 2

PPPBP bet B 1
PPPBPP bet B 2

PPPPB bet B 1
PPPPBP bet B 2

Total Net Gain = +409.74 units in 27,815 bets. 51.60% Hit Rate



for those patterns you don't need to wait for a B to end the run to start betting you bet right away when you see one of the patterns listed. If I have this right then my question is with the fist pattern in your test that got +409 did you wait for a B to end PPPPP or did you just bet so that a run of 7 P beats you

After reading again I think we are talking about 3 ways of playing
1 is after BPPPP bet B
2 is with the patterns listed above with out waiting for a B to cap runs of long P's
3 waiting for runs of three or more ended with a B then bet B for at most the next two bets

I want to know if your fist test you did was without a cap on the P runs of B and then bet did you test with the stop loss of +1-3 per shoe or was that somehting you added later?
with the pattens you listed was the wait for B and then bet B trigger used for the long runs of P
with the 3rd idea of a run of 3 or more p ended with B then bet B was this tested I tested the first 30 shoes of zumma 600 and it finished +6 units at one point it was up 22 this was with no stop loss per shoe

Evo


I hope this helps clarify. You do not need to play all patterns listed.

The way I suggest to play (only a selection of patterns) is a Hit & Run style as follows:


1) Wait for BPPPP then bet B (1 unit) and if lose that bet bet B (2 units) and stop.

2) Some players, however, prefer to wait for any potential streak of e.g. PPPPPPPPP... to end first with a B: in which case you could play:

...PPPPB bet B (1 unit) and if lose bet B (2 units).



In addition, If 1) is a loss you still have 2) as an opportunity in the shoe to regain 1 unit... it depends on your style and risk tolerance.



nb. All the other patterns listed need not be played.... they were listed in thread as confirmation on basic approach of waiting for streak rich in ...PPPPs (four or more) before betting Bank in testing. The tests were without the stop loss 3 unit and target 1 unit, and only patterns fully formed inside a shoe were taken.

[fauzy]

Evo


I hope this helps clarify. You do not need to play all patterns listed.

The way I suggest to play (only a selection of patterns) is a Hit & Run style as follows:


1) Wait for BPPPP then bet B (1 unit) and if lose that bet bet B (2 units) and stop.

2) Some players, however, prefer to wait for any potential streak of e.g. PPPPPPPPP... to end first with a B: in which case you could play:

...PPPPB bet B (1 unit) and if lose bet B (2 units).



In addition, If 1) is a loss you still have 2) as an opportunity in the shoe to regain 1 unit... it depends on your style and risk tolerance.



nb. All the other patterns listed need not be played.... they were listed in thread as confirmation on basic approach of waiting for streak rich in ...PPPPs (four or more) before betting Bank in testing. The tests were without the stop loss 3 unit and target 1 unit, and only patterns fully formed inside a shoe were taken.

27,815bets How many shoe ?

[Egalite]

Strategy B12


As example consider all 6 parity outcomes over four bets (i.e. 2W = 2L):

W = Win, L = Loss (bold denotes completed 1,2 bet sequence)

1) LL W W (-1-2+1+1) = -1
2) LW LW (-1+2-1+2) = +2
3) LW W L (-1+2+1-1) = +1
4) W LW L (+1-1+2-1) = +1
5) W W LL (+1+1-1-2) = -1
6) W LL W (+1-1-2+1) = -1

You left out;

7)LLLL (-1-2-1-2) = -6

Maybe you think "it can't or won't happen"

[Steve]

27,815bets How many shoe ?

1600 shoes of both Zumma

[Steve]

You left out;

7)LLLL (-1-2-1-2) = -6

Maybe you think "it can't or won't happen"

Egalite

I did not include sets where WWWW or LLLL etc. intentionally.

Since the method shows that hit rate >50%, my post only listed outcomes for parity Wins = Losses to illustrate, by way of example over 4 hands, how betting 1,2 can improve gain from just flat betting (where the outcome would be zero in the latter case).

Of course LLLL can happen, but that was not part of the example of parity.

[Egalite]

Egalite

I did not include sets where WWWW or LLLL etc. intentionally.

Since the method shows that hit rate >50%, my post only listed outcomes for parity Wins = Losses to illustrate, by way of example over 4 hands, how betting 1,2 can improve gain from just flat betting (where the outcome would be zero in the latter case).

Of course LLLL can happen, but that was not part of the example of parity.

Ok fair enough, let's ignore what can happen at the tables, let's pretend it can't.

The problem with a simple 1-2 progression, is, because you do not extend the progression futher (to protect your bankroll, no doubt), after any LL sequence, you need a 75% hit rate, just to recoup, god forbid should you hit LL twice, which is probably the reason you left it out.

BTW - Betting B only, (no matter how you dice & spice it), has been done to death yonks ago on other older sites. Also wittnessed a player slowly go broke over three days to the tune fo $10k (flat betting B only).

[Steve]

Ok fair enough, let's ignore what can happen at the tables, let's pretend it can't.

The problem with a simple 1-2 progression, is, because you do not extend the progression futher (to protect your bankroll, no doubt), after any LL sequence, you need a 75% hit rate, just to recoup, god forbid should you hit LL twice, which is probably the reason you left it out.

BTW - Betting B only, (no matter how you dice & spice it), has been done to death yonks ago on other older sites. Also wittnessed a player slowly go broke over three days to the tune fo $10k (flat betting B only).

Egalite

Of course if you bet Bank only without an edge (using some random outcome strategy) you will not overcome the house edge and go broke.

That is why you should test the strategy first.

The point here is that we have a positive expectation strategy based on the tests so far with 51.6% hit rate.

If you do not have an edge no progression will save you ( I have said this before), moreover, if the cards are random no strategy (flat betting or otherwise) will overcome the house edge.

An interesting point about the 1,2 progression. I do not consider it a true progression because the amount bet on the specific pattern remains constant.

For example,

BPPPP bet 1 unit on B
BPPPPP bet 2 unit on B

So it is fundamentally flat betting.

If you disagree with the figures, test it for yourself, since ultimately whatever is said the data has the final word.

[Egalite]

Egalite

The point here is that we have a positive expectation strategy based on the tests so far with 51.6% hit rate.

So you think a 1.6% edge will swing things for you? If I bet Banker only against my own test data, I have an "on-paper" edge of 3%, because my test results indicate an overall hit rate of 53% over 10,000 decisions. Yet doing this doesn't work at the tables, maybe it has something to do with the 5% casino tax, which results in zero profit on a .3% edge, or maybe it's simply those punto dominated shoes which can produce themselves any any time.

What you appear to be doing is betting "anti-streak" (a'la system 40 / sniper) applicable to one side only, progression capped at 2u.

[Steve]

So you think a 1.6% edge will swing things for you? If I bet Banker only against my own test data, I have an "on-paper" edge of 3%, because my test results indicate an overall hit rate of 53% over 10,000 decisions. Yet doing this doesn't work at the tables, maybe it has something to do with the 5% casino tax, which results in zero profit on a .3% edge, or maybe it's simply those punto dominated shoes which can produce themselves any any time.

What you appear to be doing is betting "anti-streak" (a'la system 40 / sniper) applicable to one side only, progression capped at 2u.

I agree the edge alone will not provide much of an advantage at the table. If you look at the random walk scenario for 50%v50% outcomes you will see that the fluctuations in profit to loss grows with time (the ratio of win/bets tends to 50%), and this is one way the casino wins over a player who has a small edge 1+% or so because that player will eventually be stopped out before the casino is touched... the casino having the greater bank can ride their wave of losses longer. A gambler even with no house edge (a fair game) will go broke with certainty if he does not know when to quit.... I can give several published papers on this data.

The edge will work in your favor, giving a upward drift in your fortunes, if you use a stopping strategy that includes a profit target, and stop loss with discipline, i.e. money management re: Kelly

Have you posted your strategy?

If not, why not post your strategy it will be of interest to test, I can test it against Zumma and other data... was it 10,000 decisions or 10,000 actual bets?

The former will not be sufficient to establish a true edge in my opinion. Even my tests on strategy are not conclusive, the basic rule is the more testing the better.



Steve


To calculate you Edge or Expectation you must allow for 5% commission.

Expectation per unit on Bank Bet = (Number Win Units - Number Loss Units - 5% commission on Win Units)/Total Units Bet

[fauzy]

As example consider all 6 parity outcomes over four bets (i.e. 2W = 2L):

W = Win, L = Loss (bold denotes completed 1,2 bet sequence)

1) LL W W (-1-2+1+1) = -1
2) LW LW (-1+2-1+2) = +2 ppppbpb Lw ppppb pb Lw
3) LW W L (-1+2+1-1) = +1
4) W LW L (+1-1+2-1) = +1
5) W W LL (+1+1-1-2) = -1
6) W LL W (+1-1-2+1) = -1
__________
+1 unit (- vig.)
__________om limited data set.

__________________________________________________ __________________________________________________ ________________________________________

where can I find it ?is it like above condition?

[fauzy]

Hi
Do you test betting 3p 2b ?

pppbb, bet b 1,2
pbbpp bet b,1,2

[Steve]

As example consider all 6 parity outcomes over four bets (i.e. 2W = 2L):

W = Win, L = Loss (bold denotes completed 1,2 bet sequence)

1) LL W W (-1-2+1+1) = -1
2) LW LW (-1+2-1+2) = +2 ppppbpb Lw ppppb pb Lw
3) LW W L (-1+2+1-1) = +1
4) W LW L (+1-1+2-1) = +1
5) W W LL (+1+1-1-2) = -1
6) W LL W (+1-1-2+1) = -1
__________
+1 unit (- vig.)
__________om limited data set.

__________________________________________________ __________________________________________________ ________________________________________

where can I find it ?is it like above condition?

Yes, that is correct.

[Steve]

Hi
Do you test betting 3p 2b ?

pppbb, bet b 1,2
pbbpp bet b,1,2

Tested all 1s, 2s, 3s, both against the trend such as pppb bet b, and bbbp bet p; and
with the trend pppp bet p, and bbbb bet b.


All gave negative results.

It seems that any strategy based only on the last 4 or fewer hands is insufficient to overcome the house edge. You need to take a longer view of the past decisions.

Something that I am looking into for this strategy to improve hit rate is filtering by looking at shoe history, maybe only take trigger when Bank is showing 2s+

[fauzy]

easy to choose
pppbb set
than ppppb 4p/1b
did you teste ?

[gr8player]

Steve: >Something that I am looking into for this strategy to improve hit rate is filtering by looking at shoe history, maybe only take trigger when Bank is showing 2s+<

Good job, Steve.

Just by reading your thread here, one can easily see that you take this game very seriously, and aren't afraid of doing the necessary "leg work" in order to improve your results.

You're most definitely on the right track when you mentioned "filtering" (read: trending) and "shoe history".

If I may add:

Add "variance" to that mix, combined with a slight shoe "progression", and you're going to hit the "Grail" with this, IMHO.

Edited to add: By variance, I mean: Calculate your strike rates, your loss streaks and your win streaks. You'll be able to "Grail" this play by playing to your variance. It should be very, very tight for this "one play" method.

Keep me posted.

[Steve]

easy to choose
pppbb set
than ppppb 4p/1b
did you teste ?

Fauzy, yes I tested pppb bet B (1 unit) and if lose bet B (2 units) : all negative results on that.

[Steve]

Steve: >Something that I am looking into for this strategy to improve hit rate is filtering by looking at shoe history, maybe only take trigger when Bank is showing 2s+<

Good job, Steve.

Just by reading your thread here, one can easily see that you take this game very seriously, and aren't afraid of doing the necessary "leg work" in order to improve your results.

You're most definitely on the right track when you mentioned "filtering" (read: trending) and "shoe history".

If I may add:

Add "variance" to that mix, combined with a slight shoe "progression", and you're going to hit the "Grail" with this, IMHO.

Edited to add: By variance, I mean: Calculate your strike rates, your loss streaks and your win streaks. You'll be able to "Grail" this play by playing to your variance. It should be very, very tight for this "one play" method.

Keep me posted.

Thank you for the positive feedback gr8player.

[fauzy]

Fauzy, yes I tested pppb bet B (1 unit) and if lose bet B (2 units) : all negative results on that.

pppbb

bet
b1,2

[Newkidintown]

Joker has it correct...

combined patterns listed in post #1 gave positive results.

From point of view of avoiding long streaks I bet after pattern ...

PPPPB bet B (1 unit) if lose bet B (2 units).

So if you see PPPPPPPPPPPPB we have trigger to bet B.

Most of the time, when we see PPPPPPPP...B, at least one or two P will coming out again, I will wait, PPPPPP...BPP then bet on B (1 unit), if I lose, go to the other table, looking for the same pattern, then bet (2 units), if i lose again, keep looking for the same pattern, at the other table then bet (4 units)...

[punvit]

Most of the time, when we see PPPPPPPP...B, at least one or two P will coming out again, I will wait, PPPPPP...BPP then bet on B (1 unit), if I lose, go to the other table, looking for the same pattern, then bet (2 units), if i lose again, keep looking for the same pattern, at the other table then bet (4 units)...

Agreed, I have seen a lot of those.

[Egalite]

Steve had another read of this, IMO you need a complimentary bet selection that you can use in a worst case situation.

A worst case situation would be, you hit 2 Player 6 streaks "off the bat", and you don't see another Player 4 streak in the next 2 shoes. When I looked at these sort of stats from the Zumma Books, betting the opposite of a Banker 6 streak could be a complimentary option. But I really think you need "a solid something else" to fall back on, just in case...

[Steve]

Yes, the thing about waiting for long strings is that you get few bites of the cherry. Something that may be of interest are patterns:

BPBP bet P
BPP bet P
PPB bet P

Doesn't beat Zumma unless have 1 unit profit target per shoe (then wins high % shoes). This may work in deep progression style betting, and with Star, Mongoose, because of no commission.

At best blind mechanical testing gives fractional unit gain per shoe, so it is interesting to think of ways we can break away using 'read of shoe', experience etc. With frequent bet opportunities this gives us room to select placement of when to bet on pattern.

[Egalite]

If you want to base your bet selection around the Zumma stats (complimentary or otherwise), I covered these in the thread "Zumma busters" or something, I'll repeat my crude finding here;

After PPBB bet the two's won't continue, so you first bet B (52%), if lost bet B again (53%)

After two chops bet the third chop, BP bet B (56%), after PB bet P (53%)

Anti-streak options;

Wait for a P 5 streak (not 4) then bet B (53%)

If the P streak goes to 7, bet B (54%)

Trend Options;

After the second line & third line have been consecutively populated in prior columns BBPPP or PPBBB, bet the 3 streak will go to 4, better still wait for the three or more streak to end and then bet 2nd line for the next hand (50~59%).

There are enough options above to construct a few complimentary bet options, or TRIGGERS.

However I suggest these are taken from the Zumma books and will have no bearing whatsoever with what can happen for any individual shoe or session.

(dismiss what I said above about betting against any Banker streak, the results were less than 50% in all cases)

[ensonchong]

Interesting! thanks

[philipph]

These triggers work well individually or as a combination:
BPPPP bet B (1 unit), If lose bet (2 units) next B

PPPBP bet B (1 unit), If lose bet (2 units) next B

The family of other Patterns are listed below.

(ignore Ties)

___________________________________

Against Zumma 600 and Zumma 1000

Hit Rate 52%

___________________________________

Can you prove the hit rate is 52%?

[Steve]

This is the hit rate against Zumma 600 & 1000, (result pdf is listed). Could well change over other data sets. Which is why testing on own data was recommended.


Steve

.................................................. .................................................. .................................................. ...........................................

In Baccarat the House Edge is not sufficient to establish that the game is unbeatable.