help me get the edge

i just lost 200 playing at Harrah's $10 tables. as a poor college student, that is significant money. the annoying thing is that i played perfect basic strategy, and i mean perfect. i looked up the exact basic strategy chart needed for an 8 deck, S17, DA2, DAS, No surrender game, and even quizzed myself using flash cards. i realize this still leaves the house with an approximate .47% edge.
none the less i played and lost. however i am not angry... on the contrary i'm hungry... for the egde.
i just recently read Fred Renzey's book "Blackjack Bluebook II" and found it well-written and thorough. i am currently attempting to master the Stage II Kiss Count explained in the book. (black 2's (+1), 3-6 (+1), 10-K (+1). since i began practicing the count several DAYS AGO, i was obviously unprepared to use it in the casino, so i just stuck to basic strategy. however, this brings me to my questions:

1. Is the Kiss Count a good level one count, or should i use a different count, or perhaps a balanced count?

2. After i have chosen a good entry level count to practice, what are the practice techniques i should use to master it? Are there other techniques besides simply counting down a single deck in under 30 seconds?

3. as a comparatively poor college kid, i have a limited bankroll. i understand bad swings happen, which leaves me vulnerable. how do i account for this?

I simply refuse to go back to a casino as a favorite to lose in the long run - that is for suckers. i want an edge.

i welcome all comments, suggestions, and criticisms, as well as suggestions for additional readings. please help me.

-losing in new orleans

"I simply refuse to go back to a casino as a favorite to lose in the long run - that is for suckers. i want an edge."

How did you just lose $200 while admitting the house had a .47% advantage? Or is this a "new" resolution?

If you are dying to play with an advantage on a very limited budget you are not in the right town. An 8 deck game at Harrahs with poor penetration is not the way to go. You don't have the bank to survive any 2 Deck game in NO($25 min) and there are very few anyway.

To be honest a poor college student should stay clear of BJ even if he is playing with a 1% advantage. Even with an advantage you are just gambling your rent, tuition, books, gas, or spending money away. You simply cannot survive the swings.

Practice away with a PC. Read books like Professional Blackjack, Blackjack Attack, and Burning the Tables in Vegas. Learn the count system you feel the most comfortable with. Keep it simple. Don't play again until 1) You KNOW you have a long term edge 2) You have money you can afford, and are willing to gladly lose in pursuit of a long term edge. Wait until you have the bank for a comfortable RoR or are able to set aside a monthly budget. Stick to the lowest bets possible.

I hope you realize that even after you do all this you will probably only be making a few bucks an hour.

Not to depress you too much but a job is so much easier.

Krillion the Great

i just lost 200 playing at Harrah's $10 tables. as a poor college student, that is significant money. the annoying thing is that i played perfect basic strategy, and i mean perfect. i looked up the exact basic strategy chart needed for an 8 deck, S17, DA2, DAS, No surrender game, and even quizzed myself using flash cards. i realize this still leaves the house with an approximate .47% edge.
none the less i played and lost. however i am not angry... on the contrary i'm hungry... for the egde.

You've come to the right place.

i just recently read Fred Renzey's book "Blackjack Bluebook II" and found it well-written and thorough. i am currently attempting to master the Stage II Kiss Count explained in the book. (black 2's (+1), 3-6 (+1), 10-K (+1). since i began practicing the count several DAYS AGO, i was obviously unprepared to use it in the casino, so i just stuck to basic strategy. however, this brings me to my questions:

1. Is the Kiss Count a good level one count, or should i use a different count, or perhaps a balanced count?

You are asking a loaded question, which has no correct answer. I'm not familiar with the Kiss count, but it I would guess, based on the tag values, that it's less effective than either HiLo or KO (this is just a guess, so if I might be wrong). HiLo is the industry standard and most literature is based on it. KO is relatively new simple, powerful system with a large following (www.koblackjack.com).

Personally, I use KO for single deck and HiLo for shoe games. This might not be optimal from an EV standpoint, but I do it this way because it's simpler for me. In single deck I don't have to perform a RC to TC conversion with every card using KO. In shoe games, the RC to TC conversion occurs less frequently and the entry and departure points are easier to determine using HiLo.

In general, you will find that in the tradeoff between simplicity and EV, simplicity is usually you're best choice.

2. After i have chosen a good entry level count to practice, what are the practice techniques i should use to master it? Are there other techniques besides simply counting down a single deck in under 30 seconds?

There's a link to BJ School on this site which has good training drills. Otherwise, I would use a blackjack playing program. www.qfit.com sells Casino Verite, which, from what I have heard and read, is awesome. But if you're cheap like me, Sage Blackjack is more than adequate. I've used it for ten years. A shareware version is available at www.s-a-g-e.com.

3. as a comparatively poor college kid, i have a limited bankroll. i understand bad swings happen, which leaves me vulnerable. how do i account for this?

If you plan on counting, you may actually increase your risk because of the bet spread. You will either have to be lucky early or have to accept that you may need to replenish your bankroll at some time. There are techniques that will reduce your risk, but usually at a loss of expected value.

I simply refuse to go back to a casino as a favorite to lose in the long run - that is for suckers. i want an edge.

Just knowing I have a edge is more reward for me than the winings. :o)

i welcome all comments, suggestions, and criticisms, as well as suggestions for additional readings. please help me.

You have come to right place.

-losing in new orleans

Here's to winning. :o)

according to the numbers provided in Fred Renzey's book (they were generated using Stanford Wong's Blackjack Count Analyzer)KO and Hi/Lo both have high betting correlations but lower playing coefficients (97%/55%) and (97%/51%) respectively, whereas the KISS count has (89%/61%)

thanks for the adive on practicing tips. i will use them.

1)I'm not familiar with the count you are using, but let's assume it is not the problem.

2) Practice playing at home, keeping the count and accurately adding your hand totals and making necessary deviations from basic strategy. Practice counting down the deck for speed, but accuracy during play is more important.

3) To reduce your vulnerability to swings, the first step is crucial: Don't play bad games. You can find decent games with fewer decks, played at lower limits just over an hour's drive East of you. It is worth putting in the miles for a quality game.

Check out the online catalog here for some prime reading on the game. One of my favorites is Blackbelt In Blackjack by Arnold Snyder- it is coming back into print this Fall.

Keep at it, and good luck-

Sentry

at Boomtown in Harvey, Bally's, and on the Coast at Casino Magic BSL, and the Copa. Play the single deck games.

3) To reduce your vulnerability to swings, the first step is crucial: Don't play bad games.

I stand by my statement "Don't play bad games", but the truth is that you will get some of your most brutal swings in the good games. Your best playing opportunites come there, but so do your biggest bets that miss. If you are playing with an edge, you can expect to come out favorably in the end.

With single deck games you can use a much smaller spread. Also Boomtown's Single deck game gives you an advantage off the top. Be careful there, the ploppies are pretty vicious.

I think Krillion, Panama, and bigplayer;

all gave good replies. Thinking about these, the following two
questions seem to arise and I think, amazingly, these questions have
not been directly addressed yet. I say "amazingly" because so much
good math work on BJ has been done already.

Questions:
1) Even if playing with a small theoretical advantage via counting and
bet spread, if one has such a small bankroll that one's risk of ruin
is large (say greater than 50%), is it really correct to say that one
is playing with an advantage over the house? I think the answer may
be "no" -- in this case the house really has an advantage over you.

2) [This one I think moreso has not been addressed.] Consider the
perspective of the casino. Suppose a large number (say 1,000) of
players start playing the game with a count and bet-spread that gives
them theoretical advantage (say 0.5%) but they all have small bank
roll giving them 50% risk of ruin. Question: will the casino make or
lose money? [Clearly the answer will involve time -- how long can the
500 players who don't go broke play? and even actuary, how long will
they live on average...]

Hmmm...
replying to my own question. At first glance, from the casino
perspective, if there are 1,000 players playing at an advantage, I am
going to lose money. BUT on the other hand, if I see 1,000 players
showing up with a pot of money and I know in advance that 500 of them
will give me the whole pot, while the other 500 will on avegerage make
a bit, I may think "bring them on!"

The answer is not so obvious.

" Even if playing with a small theoretical advantage via counting and
bet spread, if one has such a small bankroll that one's risk of ruin
is large (say greater than 50%), is it really correct to say that one
is playing with an advantage over the house? I think the answer may
be "no" -- in this case the house really has an advantage over you"

It is a matter of semantics, but it still is correct to say that one is playing with an advantage over the house. The reason is that even though the counter is likely to lose his bankroll (due to overbetting leading to a large risk of ruin) if he bucks the odds and doesn't lose it all he will win much more than just his bankroll (again due to overbetting). So even though he is more likely to lose than win, the rewards of succeeding outweigh the price of failure in net. This can be seen more clearly in your second question.

"Consider the
perspective of the casino. Suppose a large number (say 1,000) of
players start playing the game with a count and bet-spread that gives
them theoretical advantage (say 0.5%) but they all have small bank
roll giving them 50% risk of ruin. Question: will the casino make or
lose money?"

From the casinos perspective this is a losing proposition. From the casinos perspective the counters are playing with a joint bank so they are not really overbetting their bankroll. This means that even though 50% of the counters will lose their individual bankrolls due to risk of ruin, those that do not lose will win more than all of the others lost.

Furthermore, looking at a small bankroll as having a high risk of ruin can be very deceptive if the bankroll can and will be replenished from outside funds. A person who starts out at zero but saves $500 per month to play blackjack with(6D game, betting $10 - 2x$80, 85% penetration) will have a high risk of ruin each playing session. Initially he will probably lose quite a few $500 banks, but eventually he will catch a positive swing which will carry him to win more than he had lost previously.

As long as someone is playing with a positive expectation thay are playing with an advantage, no matter what their risk of ruin.

Umm:
What's with the name? anyhow...

I think you have not understood some of the subtleties of the
arithmetic. The answers to my two questions are still not really
addressed in the aggregate. Consider just one example: An extreme
case -- suppose the thousand players come with a strict bankroll of
10 units; they will play with a spread of 1-7 which will give them
a small "advantage". In this case they will all go broke and stop
playing -- the casino will make 10,000 units, except for a handful
of players that may escape ruin but it will be a very long time before
they win more than 10,000 units. If the casino cares about profits
and not abstract notions, the casino will want these players.

.. is the same thing as one player with a 10,000 unit BR. This assumes that all players are playing with the same skill level, the same edge, and that they aren't all playing at the same 1,000 spot table - which I think you are implying with your post.

This is something that is quite easy to prove mathematically and has been addressed. For example, playing a DD game with S17 DAS spreading 1-8 with roughly 60% penetration requires slightly over 500 units to maximize DI and play full Kelly (ROR=13.5%). At 250 units, the ROR is ~36.9%, at 125 units the ROR is 60.8%, at 10 units the ROR is ~96.1%.

From basic probability, we know that our chances of failure with a bank of 10 units is 96.1%. We also know that our chances of failing twice in a row is 96.1% * 96.1% or .961^2. The general formula for probability of failing n times in a row with a 10 unit BR would be

P = .961^n

It does not take much of a leap to see that a 500 unit BR is made up of 50 smaller 10 unit BRs, so the probability of busting that 500 unit BR should be equal to

P = .961 ^ 50 = ~13.68%

(Note that the errors here come from rounding. The ROR for a 500 unit BR is 13.64%)

For your original question, if the 1000 counters all descended upon the casino at once, there would be some skewing of the math as their results would be somewhat correlated (ie-dealer busting once would result in 7 people winning, not just one, etc). However, I do not believe this was your intent.

This is often why most of us who are able to replenish the BJ bankroll if necessary due to a swing of bad standard deviation "luck" do not worry about lifetime ROR and only consider trip ROR. It does not make sense to keep a large amount of capital dedicated to BJ and bet extremely small, essentially "losing" EV to missed comps and other "money growing" possibilities, simply to keep the lifetime bust possibility low.

"I think you have not understood some of the subtleties of the
arithmetic."

I understand the concepts quite well actually. This topic has been covered extensively in quite of number of threads on these pages through the years. I find it quite odd that you are confused about something and ask a question about it, then when I give you a 100% truthful, accurate answer you tell me you think I don't understand the subtleties of the arithmetic.

Anyway:
"Consider just one example: An extreme
case -- suppose the thousand players come with a strict bankroll of
10 units; they will play with a spread of 1-7 which will give them
a small "advantage". In this case they will all go broke and stop
playing -- the casino will make 10,000 units, except for a handful
of players that may escape ruin but it will be a very long time before they win more than 10,000 units."

Sorry, this is simply incorrect. If they are truly playing with a small advantage, then as a group they are likely to win more money from the casino than they will lose. In the long run most will go broke, but a small few will more than make up for it.

Yes it will take a long time for for the 'winners' to make more than the 10,000 aggegate units, but it will also take a long time for many of the 'losers' to go broke. They will bounce around near breaking even for a long time before finally succumbing to ruin.

You could start you experiment, and then stop it at any given time and tally up all of the counters money. On average they should have on aggregate 10000 + (Expected Value * total amount bet). If you wait long enough for 96% (or whatever their individual risk of ruin is) to go broke, the remaining 4% on average should have more than $10,000.

It is simple mathematics.

Even an underfinanced counter is a long term threat to the casino, because as a group, underfinanced counters still take money from a casino.

The way I see it, you have two choices. Adopt any of the premiere counting systems, become professional at it and then find yourself a stakehorse. Or if that doesn't appeal to you, holdem poker offers good chances. Although you'll still need a decent a bankroll, it won't require nearly as much as Blackjack.
Furthermore, if you want a good solid Blackjack practice program with many useful accountrements, use 6-7-8 Stickysoft by Hal Marcus. I had the S-a-g-e Version 5.3 software but ended up tossing it out.

Umm:
"I find it quite odd that you are confused about something and ask a question about it, then when I give you a 100% truthful, accurate answer you tell me you think I don't understand the subtleties of the arithmetic"
1) I believed, and still believe, that you answered a question but not
the actual question that I was indicating.
2) I found it odd that I posed a polite and interesting question, and
you responded with a name that appeared insulting ("Umm..." as in
"Duh..."). If that wasn't your intent, I apologize.

Regarding the question, your simple math is correct, but it is
oversimplified by smoothing out the e.v. and applying it as if it
were a continuous function. If we get to small enough bankrolls, the
discrete nature of the game becomes relevant. If the game is play all
(not Wonging where one ONLY plays positive ev hands) then the players
play more negative ev hands than positive ev hands. If we limit
each player to a sufficiently small bankroll, I believe there is
some point where the players do not get to play enough positive ev
hands for the result to reach the continuous function limit that you
estimated. I admit I have not done the actual calculations, and I
might be wrong, but I believe there are some limiting cases here that
have not been addressed in this forum or in the primary books.

"I found it odd that I posed a polite and interesting question, and
you responded with a name that appeared insulting ("Umm..." as in
"Duh..."). If that wasn't your intent, I apologize."

"Umm" is my regular posting name (to this and many other different message boards) and has nothing to do to with you or your post. It comes from a habit I have where I pause and think (sometimes excessively) before speaking. Even though I don't regularly say the word aloud, my wife calls the obvious pauses "Umm moments", hench the nickname.

"Regarding the question, your simple math is correct, but it is
oversimplified by smoothing out the e.v. and applying it as if it
were a continuous function."

In the long run with suffciently large number of counters, on average, it is just like a continuous function. The actual individual details will be bumpy and random, but in aggregate, the results will be just like a continuous function.

"If we get to small enough bankrolls, the
discrete nature of the game becomes relevant. If the game is play all
(not Wonging where one ONLY plays positive ev hands) then the players
play more negative ev hands than positive ev hands. If we limit
each player to a sufficiently small bankroll, I believe there is
some point where the players do not get to play enough positive EV
hands for the result to reach the continuous function limit that you
estimated."

If the bankroll is so small (and the counters are forced to play all) that they don't have enough money to last long enough to play enough positive EV hands, then they aren't playing with a positive expectation. You are changing the rules. If they aren't playing with a positive EV then there is no argument. I agree with you, casinos love people who are not playing with a postive EV (and over/under financed bankrolls aren't even applicable). I think you are mixing definitions here.

If the counters are playing with a positive overall EV (like in your original example) then no matter how small the EV is, they are a threat to the casino and will in aggregate win money from the casimo.

I didn't make any estimation. It is by definition. If they are playing to a positive EV, then by definition they would be expected to win money from the casino. A high risk of ruin doesn't change a positive EV into a negative EV. Risk of ruin doesn't affect EV.

You could break it down to the smallest bankroll you wanted. A group of counters with a bankroll of 1 unit each playing a game with a +.001 EV. As long as the EV is truely positive the group would be expected to win money from the casino on average. Now with such a small EV you would have to get an extremely large group of counters to see statisitcal significance, but the expectation would be for them to win on average.

When a player makes a bet with the advantage the casino is a net loser. The amount of money a player has in his pocket, at home, or in the bank makes absolutely no difference. If a thousand players across the country are each making a bet with a 1% edge, as a group they'll almost always win. The fact that some of them are losing their last dime does not change the equation in the least. And the fact that you put some smaller group of them together for the purpose of summing results also does not matter. Whether one little sub group comes out ahead and another behind does not affect the grand scheme of things.

Let's say you could buy one ticket for a drawing. The drawing has only 10 entries. The ticket cost $1. The winner of the drawing gets $1 million. You have a very huge advantage even though there is a 90% chance that you will lose. Let's group four of the guys together and say that this $1 comprises the entire bankroll of each. Both individually and as a group the chances favor their going bust. Does the house have the edge? I don't think so.

DD'
I see your reasoning, and certainly I agree that
"When a player makes a bet with the advantage the casino is a net loser."
I do not question that.

My question is this. (First I note that this is of course only for
hypothetical interest, no one should or would play this way.)
Consder playing heads up with the dealer, especially in a shoe
game. The player plans to bet one unit on all negative expecation
hands and ramp up to 5 units on positive ev hands, as long as the
player has that many units. If he has fewer units than called for
he bets what he has. As long as the player has enough units this
will, on average over many shoes, give him the advantage. The player
plays untill he has zero units.

Now let's consider whether the bakroll that the player starts with
has any effect on the average expected return (or loss) to the
casino.

Start with the most extreme case:
The player starts with one unit. The first round of any shoe is
negative expectation. The odds are greater than half that he will
go bust and stop playing on the first hand. If he doesn't go bust,
he continues to the second round -- depending on the size of the
shoe, this round may also be necessarily negative expectation and is
certainly much more likely to be negative than positive, etc.
Now, mathematically if every time this player goes bust, a new player
also with one unit bakroll and same conditions starts playing on
a new shoe, the long term result for the casino will be different
(better for the casino) than will be the long term result for players
starting with big banks.

Just as you say, and I agree, that the size of a player's bank at
home has no effect on the result of a given hand, also a player's
idea that when he gets to positive expectation hands he will bet more
has no effect on his overall result if he goes bust and stops playing
while he is playing through the negative hands in hopes of getting
to the positive ones.