One thing about being a smart gambler is that you quickly find out
you are rara avis.
Innoculated against the ravages of gambling
addiction by my early instructive losing streak, I was mystified when
Y pumped a nice $500 win into a $25 slot machine because "my luck has
been so good recently." This is someone who has an advanced degree
in physics! And X, who had studied the math (not really complicated
stuff, it's all basic algebra and statistics) until he could bore
you for hours with the effects of a rule change or strategy variation
on your Expected Value (EV), couldn't seem to avoid getting wiped out
in tragic negative swings. Back to work, at least for a couple of
months, back to the tables, oopsie, back to work...
I prosyletized. Many of my friends and coworkers gambled, and they
found my success interesting. I made no secret of how to get the
best edge, work the comp system, and parlay the tournaments into a
The most common response was: "Well, I really like to play the slots."
I can readily tell you the number of Normals who saw what my friends
and I accomplished and took even a tentative step toward duplicating
it. Not succeeded, but tried. It was zero, out of
And then there were the folks we played against in the tournaments.
Nice people, generally intelligent people, but prone to do the most
amazingly ill-thought-out things in situations that were worth
hundreds or thousands of dollars. We laughed and joked and of course
much of the thousands of dollars a year I was taking home were coming out of
their pockets. At times I got a bit weirded out by that.
Every casino game of skill has an optimal strategy which can be determined
mathematically or by computer analysis. You always lose the least in
the long run by playing this "Basic Strategy," no matter what might happen
to happen on an individual hand. Thus, you always hit 16 against a
dealer 10 at Blackjack. Yes, you'll probably lose; in a statistical
sense you've probably lost the hand already. But computer analysis
shows you will lose less if you take the hit and risk busting.
If you're playing correctly there really are no decisions to make, but
almost nobody plays correctly. The theoretical house edge at most
casino Blackjack games ranges from 0.1 to 1.5 percent or so depending
on the rules. But the actual take is typically 2 to 5 percent.
This reliable gap is entirely the result of incorrect play. (This EV is in
terms of "action," the sum total of all bets made. The "hold," loss
figured against only the "drop" gamblers buy-in for, is another matter entirely.)
Occasionally this preponderance of poor play encourages a casino to do something really stupid.
One promotional idea which emerges infrequently is the "2 to 1 Blackjack"
game where natural 21 pays 2:1 instead of 3:2. Most players still
face a loss, but the perfect Basic Strategist sees the edge the
casino usually expects to see, about one percent -- without counting cards.
The Boomtown Belle, one of the "me, too" riverboat casinos legalized
in Louisiana, was desperate to get people on their boat during the
legally required 1.5 hour cruises. In September of 1995 they decided the
solution was 2:1 Blackjack, up to two hands of $25. X and Y literally
scraped together all the money they could lay hands on and showed up.
(I was now working full-time, and was unable to go with them.)
At first they nearly lost their stake of $1,200 because they were really
betting too much and had early bad luck. But then they won $14,000 over
the next 9 days in about 30 hours of play.
At which time the casino became tired of their action and invited them
Card Counting Finally Succeeds
X was pretty much an expert on the mechanics of card counting. It
doesn't work like the scene in Rain Man.
It's all in the faq.
X played smart. I watched him play and I knew exactly what he was
doing wrong. Hell, he even knew what he was doing wrong on
an intellectual level but he didn't really believe it.
When you are playing with an advantage, there is a calculation called
"risk-of-ruin". It tells you that if you have such bankroll
and bet at so level, you will have hmmmm chance of losing all your
money before you double it. Card counting is high-variance; you are
literally waiting for opportunities to bet (high counts). You can
still lose those high-count bets. As a very vague rule of thumb you
need about 100 big bets in your bankroll to count cards. X was playing
with more like 20 big bets, if that. His advantage couldn't overcome
his risk of ruin.
X lost his share of the Boomtown Belle $14K and went back to another
work, quit, lose-stake cycle. In May of 1997 the Jubilee (now relocated in
Greenville, MS) decided a 2:1 promotion was the way to lure
monied gamblers to a dumpy burg located at the most inconvenient
possible point between Jackson, MS and Memphis, TN. Most of the
monied gamblers who showed up knew how to beat the promotion and it
only lasted a few days, but X and Y made off with almost $10,000.
Not long afterward the Lighthouse Point, also in Greenville, got the
same bonehead idea and yielded several thousand dollars more.
Time to quit the job again, time to...
...call me, as it turned out.
I spent the next few weeks fielding calls from X. He was actually
going over the risk-of-ruin calculations to convince
himself they weren't bullshit. After some algebra refreshers and a
quick course on QBasic, he built some simulations and did some figuring
and convinced himself. He dutifully
went to the soon-to-close Flamingo Hilton and started playing a
modest $5-$25 bet spread. It was a fortunate choice, because a
bureaucatic screw-up of titanic proportion had lost them their
gaming license, and the lackluster staff at this doomed
boat were in no mood to protect Hilton's already-lost investment
against X's faint probing.
When counting cards, you make the money by betting small when they have
the edge and high when you do. The count tells you this; each card has
a point value, negative or positive. When the count is positive the
deck is rich in tens and the odds favor you; you bet more. When it's
negative the deck is rich in fives and sixes. The dealer is less likely
to bust, you're less likely to get dealt good natural hands. You
reduce your bet or go to the bathroom.
The gap between your smallest and largest bets is your "spread."
It's also how you get caught. While the count does tell you to play some
hands contrary to Basic Strategy (you stand on 16 against 10 in high
counts, for example) this strategy variation isn't enough to beat the
game. You have to spread your bets, and the more you spread the more
you win. And the more noticeable your play is.
In February 1994, when he had been playing with red chips and
losing anyway, X had been "backroomed" by [censored].
Security guards literally dragged X into a back room,
took his picture, and read him the Trespass Act. But the lackluster
staff at the Flamingo gave him an excellent game and once he'd
built up his bankroll he found himself still able to play at some
As 1997 rolled over into 1998 he
parlayed his modest stake into $80,000. But in that course he was
kicked out of every casino on the Gulf Coast. His big bets were now
in the $100 black chip range, and everybody knew he was a counter from
his losing days. Once he began to win, he was shown the door, although
usually with more politeness than [censored] had shown.
"Ah, Mr. X. We must contratulate you on your really excellent
play. Yes, we have noticed that you are very, very good. In fact,
you're really too good for us. You're welcome to play any of our
other games, but we really can't offer you a Blackjack game any more."
X now had a bankroll but couldn't play. We could play, but had no
bankroll. X knew we could be trusted to play by the agreed strategy
(as we had in so many tournaments) and to account properly for the
money (as we had in so many tournaments and as Y had in the 2:1 promos.)
He called and offered a deal: If we played for him we could keep half
the profit. I didn't have the time, but Y weighed half the win against
none of the risk and decided it was a no-brainer.
Y was no card-counter; we knew the theory but with no bankroll had had
no reason to drill the ability to actually maintain the count.
There is a large gap between knowing that tens are -1 and sixes are +1
and being able to count up and down as cards are dealt in a live game.
Y drilled with the computer for about
a month, an hour or two a day, then after favorable testing by X
entered a casino with several thousand dollars of his money. Y was
nervous but X pronounced the results acceptable and Y continued to
play. Known as the cheapskates that we were, we covered the story of
our newfound "wealth" with a white lie about an inheritance. Since
I didn't have the time to do the drills or play enough to make it
worthwhile, when I accompanied Y we covered my own continued cheapness
with vague hints that I was worried about Y's incipient
It worked like a charm, and before long we were being wined and dined
like royalty. At least, until our hosts figured out what we were
Continues in Part 3: The American Dream?