which system to use?

Hi all,

just wanted to get a few opinions on which "system" to use, for the game that I play. ie count type, hi-low? ko?, betting spread/bankroll etc...

my current game that i play is

8 decks
75 percent pen
dealer stands soft 17
double any 2 cards
double after splitting
no resplit aces, aces get 1 card only
and the bet minx and max are $5-100 on the one table
and another table is somehting like $25 - 500 min and max

any help would be great guys
thanks in advance
matt

I'd start off with a 1-level count unless you are a very regular player. It's easy to keep on top of a 1-level count (Hi-Lo is what I use, but you mentioned KO which is also just fine). Going to a 2-level (or more) count requires more practice, and more play to keep on top of it. The problem with Hi-Lo is the TC conversion, where the unbalanced counts eliminate this at the cost of a bit of accuracy.

About the only bankroll advice anyone can give you is to only bankroll money you can afford to lose with no pain. The bigger the better. You might make 2% of your bankroll your absolute top bet. For a 6 deck game, you need to spread 1-12. At the $5 table that turns into $5-$60. If you can get away with more, I'd go for it as 1-12 is very marginal and 1-20 will increase your winning speed. It will also increase attention if you are not careful. Good rule of thumb is to have your bankroll sized to 50 top bets, at least, and more is better to reduce risk of ruin.

Another suggestion. Buy something like Casino Verite that will give you lots of training drills, with cards turned every which way, in different combinations, etc. Plus you can play real BJ on the thing without risking money, and have it verify you are playing BS correctly, keeping the RC correctly, converting to TC correctly, sizing your bet correctly, etc. All without throwing money away...

You said 8d. I read "6d". 1-12 is not enough spread for 8 decks. You might win with 1-12, but it is going to be _slow_...

so lets say i used a spread of 1 - 12 or 1 - 20
what would the betting spread look like
1-2 units - count of +1
3-4 units - count of +2

or am i going in the wrong direction?

ive been practicing hi-lo and KO, im pretty comfortable with both, converting to true counts is pretty easy for me. however im just undecided on my spread and betting "system"

You are going in the right direction. A common idea comes from the "Kelly Criterion".

The simple idea is to take your advantage, which is derived from the True count, and multiply that times .76 (not 1.0 because on a signficant part of your hands you will double or split increasing your bet and risk) and then multiply that by your "betting unit".

IE with a TC of 1, the game is essentially even if you have good rules. With a TC of 2, you have an advantage of .5%. Multiply .5 x .76 and you get .38%. Bet .38% of your total bankroll. IE if you start out with $1000, 1000 x .0038 = $3.80. At a TC of 4, your advantage is +1.5% roughly, so 1.5 x .76 is 1.14%, or $11.40.

You can double those numbers, but you greatly increase your risk of ruin, ie the probability that you will double your bankroll as opposed to losing it all first goes down.

www.blackjack-school.com has a series of online "lessons" that covers both the Hi-Lo counting system as well as betting as described above.

A note is that you obviously won't be betting $11.40. I round down which makes it more conservative, most of the time. If you follow the above "progression" at a TC of 7 you are betting 2.28% of your bankroll. That won't happen that often, but it will certainly cause significant fluctuations in your BR.

okay, i read the article on the blackjack school and i think i understand it a bit better
when you say a betting spread of 1 - 12 it could look like this

TC 0 or lower $5
TC 1 $5
TC 2 $10
TC 3 $20
TC 4 $40
TC 5 $50
TC 6 $60
in that spread the desired level is reached 5 X 12 = $60

so would another example of a 1-12 spread look like this
TC 0 or lower $10
TC 1 $10
TC 2 $25
TC 3 $50
TC 4 $100
TC 5 or more $120 x 2 hands (or maybe just $100 X 1 or 2 hands?)
so again 10 X 12 = $120

what are the advantages and disadvantages to each, and im guessing a bankroll of about $6000 is needed for the 2nd spread?

thanks again

that looks OK. Just remember that if you look too robotic, someone will notice and if they see your bet exactly tracking the count, it can cause trouble although at $5-$10 tables I haven't seen it as much of a problem...

I usually double my bet at TC=2, and as the count climbs, I try to double (by parlaying a win if possible) when appropriate to get to the right bet point. I think it is a perfect plan to have a betting schedule before you hit the table, so that you know what you should bet depending on the TC. How you get there is another issue, but any time you can use a parlay or progressive looking bet, you will be better off. IE if you have the BR to support it, you could double at 2, double again at 4, double again at 6 and double again at 8, if you get there. The "doubling" works well. Also when the TC is way high, you will still lose. You can spread even further there if you want by doubling yet again, (again remembering your BR so you don't ruin yourself) while using the expression doubling when losing or something similar.

For two hands, you have to be careful of risk of ruin. IE if you play one hand of 100, going to 2 you should play both at about $75. Because you are going to split or double from time to time, which will make your average bet on a 75$ hand closer to $100, which is your target.

Spreading to multiple hands is a good way to get more money out, and you can utter some superstitious nonsense when someone gets up and add a second hand to "keep the flow of the cards right". Going from 100 to 2 X 100 is perfectly OK, but it increases your risk of ruin because your average bet is going to be something like $130 X 2, which requires a bigger BR to lower the odds of going bust before doubling your BR.

One other note... You can increase your bet by any amount you want when the TC is +2 or better, as you have the advantage. But big bets with small advantages leads to greater fluctuation. But you do have some flexibility as to when you increase your bet and by how much. You just want to have big bets out when you have the advantage and small bets out when you do not. Probably the best betting schedule is small bet at TC <= 1, big bet at TC > 1. Only problem is the big jump will attract unnecessary attention and can get you backed off... Of course, if you simply win at certain Vegas casinos, you can get barred... even if you aren't counting or betting big.

that makes complete sense, in regards to risk of ruin etc..
i understand it now completley 100 percent.
especially were you said 75 dollars X 2 hands, because of the doubling, splitting, etc..

p.s. is there a website or anything to calculate risk of ruins, suggested bankrolls etc... just to get a ballpark figure and to make sure im doing the correct math?

Now that ive gotten a nice spread down pat prior to entering a table
the only thing that i doubt, or give consideration into, is which count to go with

i like both of them the high low and the KO, do you think one is better than the other? or whats the advantage to each, im completley comfortable with each of them, know them both just the same.
keep in mind my game situation

8 deck
75 percent pen
dealer stands soft 17
double any 2 cards
double after 2 cards
split aces get one card
no surrender

any ideas?
ill make a seperate post at the top of the page
cheers
matt

You ought to visit www.qfit.com. I don't work for 'em, but I use their software. IE CVBJ is really nice for honing counting skills, basic strategy, count indices for your chosen counting approach, etc. They have a simulator that lets you define the way you want to play, and it will play millions of hands and tell you whether you would win or lose that way. And they have calculators built in that can help you with BR size, determine your ROR for a BR/max bet size, etc.

Well worth the money.

There are probably other packages as well from other vendors, but qfit I can vouch for...