Myooligan: Value of Precision - Preliminary Results

[Myooligan]

A week or so back there was a discussion about using index numbers to the first decimal place. Don has been reviewing/reworking a spreadsheet I made that deals with that subject, and it should eventually make it to these pages in one form or another. But in the meanwhile, I thought it'd be interesting to show what happens when you go the other direction - become less precise than accuracy to the "ones" place.

The tables below compare the SCORES of precise, risk-averse index numbers (roughly 150 indices) to another set in which the index numbers are averaged between single deck and multi-deck, and then rounded to the nearest "5,"(In other words, index numbers are 0, +-5, +-10, +-15, etc). Four exceptions: Insurance, 12v3, and A,8v3 = +3, and 13v3 = -3. I kept the precise numbers intact on these 4 decisions because the plays are "volatile" and the index happened to be "in the middle," thus not well-suited to rounding.

When I had to make decisions about rounding up or down, I favored the single deck numbers, because that's the game I play most often. But it's noteworthy, because even with these low "loss" percentages, these are still "compromise" sets of index numbers, landing somewhere in between Reno, Vegas, and AC figures. So, if you were merely rounding to the nearest five, and not also "compromising," you might be able to do even better.

 

H17 NDAS 1D 

	 Benchmark	 Rounded        Loss  

1-5	 $189.30 	 $187.87        1% 

1-4	 $160.27 	 $158.66        1% 

1-3	 $120.65 	 $118.99        1% 

1-2	 $66.71 	 $65.28         2% 

1-1	 $7.83 	         $7.30          7% 

 

H17 NDAS 6D - play only @ TC >= 0 

(Only 1 variation from 1D matrix: Insurance = +5) 

	 Benchmark	 Rounded        Loss	 

1-16	 $167.84 	 $163.38        3% 

1-12	 $162.91 	 $157.91 	3% 

1-10	 $161.86 	 $157.18 	3% 

1-8	 $157.04 	 $152.20 	3% 

1-6	 $148.25 	 $143.34 	3% 

 

S17 DAS 6D play only @ TC >= 0 

(30 play variations from 1D matrix) 

	 Benchmark	 Rounded        Loss 

1-16	 $210.43 	 $209.12 	1% 

1-12	 $208.95 	 $207.46 	1% 

1-10	 $207.61 	 $206.05 	1% 

1-8	 $204.21 	 $202.55 	1% 

1-6	 $197.49 	 $195.87 	1%
						

[Don Schlesinger]

Now that we know how costly being off by 5 is (not!), I can't wait to learn how much tremendous extra edge Francis is getting for precise, decimal-point indices. Has to be hundreds of ... pennies in a ... lifetime. :-)

Don

[Designated Driver]

> Now that we know how costly being off by 5 is (not!),
> I can't wait to learn how much tremendous extra edge
> Francis is getting for precise, decimal-point indices.
> Has to be hundreds of ... pennies in a ... lifetime.
> :-)

> Don

I hope it is okay to chime in here and contribute my own two cents(and sixth sense: intuition), but it seemed to fit in here the best. Also I think I was in the original discussion about index precision and "volatility", and the use of decimals.

While I agree that Myooligan has produced a very fine and "lovely" piece of work I do have some serious problems with his findings, not so much with the results themselves, but with the consequences they entail.

It seems to me that if this were the case(the results from his "Value of precision" study which he had posted above), then it would almost undermine the entire theory of card counting itself. I mean if the systems were/are so imprecise that they could be rounded and averaged(to the nearest FIVE, 5!) from so many different games and rules variations, then it was be so crude of a method that it would hardly have any accuracy and therefore validity of use, IMHO.

How would you refute this? Is card counting really only a crude estimation of the ratio of high cards to low? And if it is, how accurate, precise(quantitatively), meaningful, significant and reliable is this information? Is it really enough to to be able to beat the dealer consistently, winning a lot of money in the process?

Desi. D.

[Don Schlesinger]

> It seems to me that if this were the case (the results
> from his "Value of precision" study which he
> had posted above), then it would almost undermine the
> entire theory of card counting itself. I mean if the
> systems were/are so imprecise that they could be
> rounded and averaged(to the nearest FIVE, 5!) from so
> many different games and rules variations, then it was
> be so crude of a method that it would hardly have any
> accuracy and therefore validity of use, IMHO.

Well, quite simply, the evidence shows that your intuition is wrong. Rounding indices to the nearest 5 means that, at worst, you are off by 2.5. While that might represent a decent error for a few plays, it doesn't matter very much for most others. The bulk of the gain in card counting (especially in multi-deck with large spreading) comes from bet variation. While there is additional gain to be had by using indices, we have stated over and over agin (some people listen better than others!) that being very precise with those indices simply is not very important.

Francis is always quite vociferous in his protests, but he has never produced any simulations or numbers to back up his claims, and the reason is quite clear: He can't, because no such evidence exists. In fact, the evidence to the contrary is quite clear. That said, I see no reason to round to 5, because I am quite capable of doing "better." But, I don't delude myself into thinking that "better" is worth very much. It isn't.

> How would you refute this? Is card counting really
> only a crude estimation of the ratio of high cards to
> low? And if it is, how accurate,
> precise (quantitatively), meaningful, significant and
> reliable is this information?

Don't attack card counting. See above.

> Is it really enough to
> to be able to beat the dealer consistently, winning a
> lot of money in the process?

Yes.

Don

[Francis Salmon]

Myooligan said he would round every index to the nearest 5.
This would mean that any index of up to +2 would be rounded down to 0 thus altering BS-play.
For example: A,8 v 5 which has an index of +1.5 would be doubled already at a neutral count.Now we know that at TC 0 standing here has an EV of 44% versus 41% for doubling.
This makes an error of 3% of your bet for being off by only 1.5 TC.
The worst case error would therefore be 5% for being off by 2.5 TC. With a $200bet this represents a loss of $10 for one single case.
The worst case error for rounding to whole numbers is 0.5 TC or 1%. With a $200bet this represents $2 for one single case.
So you see that we're not just talking pennies here,and what's more important you can get this extra money for free, just memorizing the right indexes!

Francis Salmon

[Don Schlesinger]

> For example: A,8 v 5 which has an index of +1.5 would
> be doubled already at a neutral count.Now we know that
> at TC 0 standing here has an EV of 44% versus 41% for
> doubling.
> This makes an error of 3% of your bet for being off by
> only 1.5 TC.
> The worst case error would therefore be 5% for being
> off by 2.5 TC. With a $200bet this represents a loss
> of $10 for one single case.
> The worst case error for rounding to whole numbers is
> 0.5 TC or 1%. With a $200bet this represents $2 for
> one single case.
> So you see that we're not just talking pennies
> here,and what's more important you can get this extra
> money for free, just memorizing the right indexes!

You're right, it's not pennies; it's hundredths of a penny! Finish the analysis. We get a holding of A,8 v. 5 once every 1,111 hands. The count comes into play, roughly, 5% of the time (a guess). So, once every 22,000 hands, we lose $2 (if we've bet $200!!), if we round to the nearest integer.

Using a rounded index for this particular play, therefore, costs us, on average (using the standard 100 hands per hour), $2/2,200 = 0.09 CENTS per hour!!!! (Translation: Play 11 hours, and it's worth a penny to you personally. For the player who plays for 1/10 of your stakes, it's worth a penny every 110 hours, or maybe a penny a year.)

Now, just how stupid do you feel to use a decimal index for this play?

Don

[Don Schlesinger]

When I wrote: "Now, just how stupid do you feel to use a decimal index for this play?," I should have written: "Now just how stupid do you feel claiming that it matters to use a decimal index for this play?"

There is a difference. You needn't feel stupid at all for using your decimals. They make you feel good, and they cause little harm, so what's the problem? There's only a problem when you pompously (and, I'm afraid, somewhat ignorantly) claim that using such indices actually earns you much more money for all your alleged preciseness. Clearly, that is abject nonsense, and I know it hurts you somewhat to learn that these values are of theoretic import only.

Live with it.

Don

[Don Schlesinger]

I wrote, above:

> So, once every
> 22,000 hands, we lose $2 (if we've bet $200!!), if we
> round to the nearest integer.

> Using a rounded index for this particular play,
> therefore, costs us, on average (using the standard
> 100 hands per hour), $2/2,200 = 0.09 CENTS per
> hour!!!! (Translation: Play 11 hours, and it's worth a
> penny to you personally. For the player who plays for
> 1/10 of your stakes, it's worth a penny every 110
> hours, or maybe a penny a year.)

The above should have read "$2/220 = 0.9 CENTS per hour."

So, I've been overly harsh on Francis. His knowledge for this play is actually worth almost an entire penny per hour. :-)

Don

[Francis Salmon]

So according to you, next time I encounter the situation A,8 v 5 at a TC of exactly +1, I should double my $200 bet knowing full well that this is a 1%-error. After all, it costs me only 1 ct./hour in the long run.
As Arnold Snyder used to say. This is like handing $2 to the dealer and I would really feel stupid doing that.

Francis Salmon

[Don Schlesinger]

> So according to you, next time I encounter the
> situation A,8 v 5 at a TC of exactly +1, I should
> double my $200 bet knowing full well that this is a
> 1%-error. After all, it costs me only 1 ct./hour in
> the long run.
> As Arnold Snyder used to say. This is like handing $2
> to the dealer and I would really feel stupid doing
> that.

I quantified the value, and the message is clear: The play is worth one cent an hour to you. Obfuscating by saying "next time it comes up" is just foolish. Next time I win the Mega- lottery, I'll have 300 million dollars!

Don

[Francis Salmon]

I picked one example among many others to show that index precision can be rewarding.In the situation I described the right play yields two extra dollars and this is true regardless of the frequency of its occurrence. You may continue to lose your two dollars in that situation - after all it's your own money - and I'wont even call you a fool for it. But allow me to keep mine.

Francis Salmon

[Don Schlesinger]

> I picked one example among many others to show that
> index precision can be rewarding.In the situation I
> described the right play yields two extra dollars and
> this is true regardless of the frequency of its
> occurrence. You may continue to lose your two dollars
> in that situation - after all it's your own money -
> and I'wont even call you a fool for it. But allow me
> to keep mine.

After a while, you really begin to grate on people's nerves, because you are, in the purest etymological sense of the word, incorrigible -- that is, incapable of being corrected.

If it isn't already clear to you, let me hit you over the head with it: I'm not writing these posts for you; I'm writing them for everyone else on this board, so that no one takes you seriously and no one attempts to utterly waste his or her time memorizing 100-150 indices to a tenth of a decimal place. You may delude yourself from now to kingdom come thinking that your mental masturbation matters, or that you have even -- with your crude, infinite-deck calculations -- come up with indices that are "more correct" than those used by everyone else. Knock yourself out, but leave the rest of us alone with your gibberish.

Personally, I couldn't care less what you do with your $2, every 200 hours, which, for most of our readers, who don't play for your stakes, would be about 20 cents every 200 hours.

I have known players on every major blackjack team that has operated in the past 20 years or so. And, I have known what sytems they use and how they play. No one has ever once even mentioned using indices to one decimal place. Are you so arrogant as to think that the whole world is ignorant and that you're the only smart one out there? Don't you think that if trying to calculate a decimal index and then using it were remotely important, everyone else would have done it before you? Get over yourself, and give us a break.

This will be my last post on this topic, because, frankly, I'm bored by the same melodrama played out over and over again with you. Being incorrigible is one thing, but when you start to become ineducable as well, it grows tiresome.

Don

[Cyrus]

"Don't you think that if [XYZ] were remotely important, everyone else would have done it before you?"

This strikes me as a recipe for nipping in the bud all counter-intuitive, paradigm-breaking, anti-consensus, potentially innovative thinking !

"After a while, you really begin to grate on people's nerves ... you are ... incorrigible ... let me hit you over the head with it ... your mental masturbation ... your crude, infinite-deck calculations ... your gibberish ... Get over yourself ... you start to become ineducable."

Here's a good suggestion for a counting system that might help people when posting on internet forums : When you're feeling angry, count to 20 before putting up a response. When you're feeling really angry, count to 100. And when you want to choke the other guy, shut the PC down and count your blessings.

--Cyrus