xxi: Effect of Mistakes

[xxi]

I've attached a link to an article about the MIT BJ team from this month's Wired magazine. Following is a quote:

"The irony is that a bad counter often will play a more negative game than a solid player who is simply using basic strategy. One mistake per hour obliterates a counter's advantage, and two an hour is more costly than not counting at all. According to Andrew Tay, casinos know this and so rather than automatically ejecting a known counter, they'll 'watch his play, track his wins and losses, and if he's identified as a bad counter, they'll comp him a room, make him feel like a king, and laugh as his 'positive' game slowly bleeds him dry.'"

I've heard figures like this before, and they're scary. As a recreational player, I certainly can't swear on a deck of cards I don't make one mistake an hour. Are they, correct, nonetheless? That'd be discouraging, but would certainly set my sights a lot higher.

Hacking Las Vegas

[Don Schlesinger]

The quote is patent nonsense. A BS player plays 99% of all games at a disadvantage. It's inconceivable to me to imagine what kind(s) of egregious error(s) a counter could make that would complete eliminate his/her entire edge (1-1.5%) and replace it with a disadvantage.

Obviously, some mistakes are more costly than others, but short of reversing a sign and playing a large negative count as if it were a positive one, it is ludicrous to assume that a single error or two will wipe out all of a counter's gain. It simply isn't true.

Don

[Norm Wattenberger]

1. What is an error? First, there are technical errors. These include incorrect play, incorrect bet, incorrect running count, incorrect deck estimation and incorrect TC calculation. We need to know what we are talking about.

2. Making two of any of the above errors an hour will nowhere near wipe out the advantage of a counter paying a reasonable game.

3. However, I would bet that the average person calling himself a counter makes far more errors than two an hour and has no idea that he is making these errors.

4. Then there are the non-technical errors: over-betting your bankroll and purposely betting incorrectly due to boredom, frustration and lack of discipline. These are far more dangerous than technical errors.

So, while the statement that a technical error an hour will wipe out your advantage is utter nonsense, IMO a casino is foolish to bar a counter before evaluating his ability to win consistently.

Near the bottom of the Software & Simulation page you'll find some additional discussion and data on the effect of errors.

[Parker]

I believe that Ken Uston started this "one mistake an hour will cost you your advantage" nonsense, although I have not researched this. At any rate, it has been repeated so often that it has effectively become an advantage play Urban Legend.

If you make a mistake, you lose whatever the difference in EV between the right play and the wrong play. If, for example, you stand on a 16 vs 10 in a negative count with a minimum bet out, the mistake cost you a few cents - hardly an hours worth of the 1% or so average advantage that a good counter in a good game enjoys.

Some mistakes are more costly, but as Don mentioned, it would take a real brain hemorrage, such as somehow switching -6 to +6 true count and shoving out max bets and using +6 strategy deviations, to give up an hours worth of positive EV.

I would agree that a lot (perhaps even a majority) of players who consider themselves counters are playing a losing game, but they are making a combination of mistakes as Norm discussed, and they are making a whole lot more than two mistakes an hour. If they are grossly overbetting their bankroll, it might even be said that they are misplaying every single hand!

[Karel]

Making a standard mistake may cost you a few percent of your current bet size. If you play 100 hands per hour, your edge is 1.5%, your total return is 150% times average bet size. Making a serious error with a maximum bet out may be noticable and you probably should not make more than few per hour. This would hardly happen as there are few maximum bets per hour anyway. With a minimum bet, where error happens more likely since the player may not be that careful, even few crazy error per hour, like standing hard 7 vs. 2, do not wipe out your edge.

Regards,

Karel

> Hi, everyone. I'm a long-time lurker and
> infrequent poster. But now that I've
> re-signed for the redesigned site, I hope to
> contribute more often.

> I've attached a link to an article about the
> MIT BJ team from this month's Wired
> magazine. Following is a quote:

> "The irony is that a bad counter often
> will play a more negative game than a solid
> player who is simply using basic strategy.
> One mistake per hour obliterates a counter's
> advantage, and two an hour is more costly
> than not counting at all. According to
> Andrew Tay, casinos know this and so rather
> than automatically ejecting a known counter,
> they'll 'watch his play, track his wins and
> losses, and if he's identified as a bad
> counter, they'll comp him a room, make him
> feel like a king, and laugh as his
> 'positive' game slowly bleeds him
> dry.'"

> I've heard figures like this before, and
> they're scary. As a recreational player, I
> certainly can't swear on a deck of cards I
> don't make one mistake an hour. Are they,
> correct, nonetheless? That'd be
> discouraging, but would certainly set my
> sights a lot higher.

[ET Fan]

... agree with the first sentence: "The irony is that a bad counter often will play a more negative game than a solid player who is simply using basic strategy."

The second sentence: "One mistake per hour obliterates a counter's advantage, and two an hour is more costly than not counting at all" is an overstatement, but many counters would be better off if they believed it -- at least during the training stage.

It should certainly possible to reduce errors to less than one an hour. I would not brag about an error rate of 1%.

ETF

[Mr.X]

The "one mistake per hour obliterates a counter's advantage" is one of many, many, many factual errors and misrepresentations in that article. IMO, this article averaged one mistake per sentence. As usual for this site, the 5 responses countering this false claim were excellent. I'll just reiterate Parker's classic summation that hits the nail squarely on the head - "If you make a mistake, you lose whatever the difference in EV between the right play and the wrong play."
I ranted about this horrid article on the free pages. I want to point out that I am NOT denying what the MIT team accomplished. I AM denying that they accomplished it in the manner implied by this error-ridden amateurish author. I know little about the real MIT team. My educated guess is they got most of their EV from shuffle tracking and sequencing. But this article treats those 2 techniques as little more than afterthoughts, kind of as a minor supplement to basic team play card counting. A civilian reader could easily get the impression MIT invented card counting and team play, which is inexcusable. The pictures used were so bad, they'd lose the article any credence, if the article had credence to begin with. The transportation of money, the careless, security-less handling of massive amounts of cash, the utter lack of security checks for doling out money that would tempt Mother Teresa, all are so idiotic that anyone who would use the procedures in the article would not only not be of MIT intellectual caliber, they probably couldn't even SPELL "MIT".
This article really irritated me. Buying in for $20,000 cash? Three words- See. Tee. Are.
There are many, many more errors and misrepresentations,but I'll stop my rant. Let me end by saying that Don equated the article to the way the movies "Hurricane" and "A Beautiful Mind" were (rightfully) criticized for factual errors about the real lives of Ruben Carter and John Nash. I'd go a few steps further. This article depicts the real MIT team as accurately as Gilligan's Island depicts real scientific wilderness survival techniques.