Brick: The Salmon Saga.

[Brick]

Reminds me of the SSR Saga,now we have what seems to have developed into the Salmon Saga. Francis has been around the internet for a decade(I do not know him personally) and it is quite evident he's a BJ veteran,the real thing. He's my friend and so is everybody else(except those darn casino trolls) who has interest of AP blackjack. I'm not sure how the whole ordeal started and dont care to re-read all the rumble,but both parties are guilty in my opinion(which has little value.

No BJ pro likes to be accused of wrong-doing and no BJ expert enjoys being contested. Francis should have restrained his comments,it was obvious he became very upset. On the other hand, others should not have accused him of being flat out wrong,when he's not necessarily wrong. I'm not sure which came first and dont care,but the truth is using exact index numbers has little gain in respect to ev,however,...there is gain, so nobody is wrong. If FS has no problem using his system,then hats off to him.

I would think many of us would like to have Francis around for a long time and hope he does not get barred or leave over this bloated up SAGA. He's probably tired of getting barred

@

[Don Schlesinger]

I created the I18 because, for 99% of players, the need to use more indices simply isn't there. I never said that using 150 indices wouldn't be "better," I simply said it wouldn't be so much better as to be worth the effort for the average player.

The same is true in Francis's case. No one has said that using indices to one decimal place is "wrong." What is uncontestably true is that, for 99% of all players, using whole-number indices is more than adequate. And, that is all that any of us have tried to explain to Francis.

The discussion became a problem when Francis tried to "explain" to us that we were underestimating the "power" and "value" of using one-decimal-place indices. The simple fact is that there is no power and there is virtually no value. And no matter how loud Francis shouts the contrary, it won't make it true.

This isn't about being "friends." It's about disseminating accurate, reliable information to the readers of this site. And, I don't want our readers to begin to think that they can somehow significantly increase their e.v. by learning decimal indices. Since that isn't true, I have a responsibility to the board to say so.

And, that's all there has ever been to the conversation.

Don

[Random Poster]

Although I enjoy a good debate, it seems my former inquiry sparked a downright ugly battle, so I'll try to tread a little lighter this time.

Don, I'm assuming when you published the SCOREs for the various systems, you used their most recently published index numbers, and calculated the TC with flooring (when the indices were meant to be rounded). If this is the case, aren't you introducing a new variable that may or may not make a difference in the comparison?

Thx

[Francis Salmon]

As I know from other occasions that any attempt to challenge the prevailing opinion here will provoke a furious reaction, I never got upset and I actually restrained my comments.It's true, Norm did say I was plain wrong but it turned out that he was better in dealing out punches than in taking them.Don is just trying to ridicule me in his usual patronizing manner (of,course he has a responsability to the reader!) but let me put something straight.
I didn't need to hire a whole team of researchers for my indices and I got them in no time. At the table no additional effort is needed. When we have an RC of +24 with 5 decks remaining, I see immediately this is TC+4.8 (24 doubled with a decimal point!)and I know I should split 10/10v6 and stand on 16v9. Don's reader will have to establish first is it 4 or is it 5 and for this he has to calculate as well.He will floor it to 4 and as a result he will go wrong in the two cases I have mentioned.
Remembering decimal indices is no more difficult than memorizing whole numbers. Once you use them regularly you will never forget them again.
I see from your numerous questions that for using floored indices properly a lot of know how is required. With mine it's very simple ,when your TC (decimal or integer!) is smaller than the index then hit otherwise stand or split or double or whatever.

Francis Salmon

[a BJ player]

> As I know from other occasions that any
> attempt to challenge the prevailing opinion
> here will provoke a furious reaction, I
> never got upset and I actually restrained my
> comments.It's true, Norm did say I was plain
> wrong but it turned out that he was better
> in dealing out punches than in taking
> them.Don is just trying to ridicule me in
> his usual patronizing manner (of,course he
> has a responsability to the reader!) but let
> me put something straight.
> I didn't need to hire a whole team of
> researchers for my indices and I got them in
> no time. At the table no additional effort
> is needed. When we have an RC of +24 with 5
> decks remaining, I see immediately this is
> TC+4.8 (24 doubled with a decimal point!)and
> I know I should split 10/10v6 and stand on
> 16v9. Don's reader will have to establish
> first is it 4 or is it 5 and for this he has
> to calculate as well.He will floor it to 4
> and as a result he will go wrong in the two
> cases I have mentioned.
> Remembering decimal indices is no more
> difficult than memorizing whole numbers.
> Once you use them regularly you will never
> forget them again.
> I see from your numerous questions that for
> using floored indices properly a lot of know
> how is required. With mine it's very simple
> ,when your TC (decimal or integer!) is
> smaller than the index then hit otherwise
> stand or split or double or whatever.

> Francis Salmon

I would "wait" until the TC is really 5 and not 4.8, when it is about splitting the tens vs. 6.
Good luck to the european fraction.

PS. Salmon, are you english, netherland or belgium citizen?

[Norm Wattenberger]

The facts are that Salmon is using an invalid technique to generate indexes. He claims to be using combinatorial analysis; but is skipping most of the combinations. Doesn't work that way. His numbers are not more accurate. Mathematically speaking, increasing the number of decimals with an invalid technique decreases accuracy. Even if they were accurate, it wouldn't make enough difference to cover a tip to the waitress that brings your drink. And his statements that truncation/flooring/rounding cause unacceptable errors are just plain incorrect. In fact, he does this himself. Just at a different level.

I haven't heard anyone talking about 'barring' him. But, it is our responsibility to respond to bad advice. And, frankly speaking, I personally have better things to do than continue correspondence with someone that uses insults rather than logic.

www.Blackjack-Scams.com

[Don Schlesinger]

> Don, I'm assuming when you published the
> SCOREs for the various systems, you used
> their most recently published index numbers,
> and calculated the TC with flooring (when
> the indices were meant to be rounded).

No, I didn't. Norm and I differ on this point, and each of our points of view has merit. Norm believes that when people buy a book with a system in it, they are inclined to use the indices that are published in the book. I believe that, with today's software and index-generating capabilities, players ought to be motivated to use the most accurate set of indices available for their system, which is rarely, if ever, the set in the original book.

This is not to cast aspersions on the various authors; rather it is simply what progress is all about. I know of no blackjack book, published before BJA3, that has a complete set of perfectly accurate indices for the count system included in the book. Of course, some do a better job than others, but all have errors, some of which are just egregious typos (see "Million Dollar Blackjack," for example, for some horrendous ones).

As a result, when John Auston did the SCORE sims, he used SBA to generate our own set of indices so that there would be as much consistency and apples-to-apples nature to the comparisons as was possible at the time. As the saying goes, if you use the published indices in a book, your mileage may vary. This is why, among other reasons, Norm's canned sims for CVCX will not usually agree perfectly with the SCORE values in BJA3.

> If this is the case, aren't you introducing a
> new variable that may or may not make a
> difference in the comparison?

See above.

Don

[Don Schlesinger]

When your indices are generated (not used, but created), what method (degree of accuracy) is employed to reckon the true count?

Don

[Francis Salmon]

You just praised a post of Cacarulo in which he said that the method I use is in "most cases very accurate" and now you are calling this method "invalid".Your statements are utterly incoherent and you're obviously trying to discredit me.
For the rest I let the reader decide who is insulting who and who is operating with logical arguments.They will also decide on their own whether my advice is good or bad.

Francis Salmon

[Francis Salmon]

> I would "wait" until the TC is
> really 5 and not 4.8, when it is about
> splitting the tens vs. 6.

If you don't mind giving up some EV, why not? See my post "+5 is even worse" further down.

> Good luck to the european fraction.
Thanks

> PS. Salmon, are you english, netherland or
> belgium citizen?

Neither of the three.As a principle, I don't answer questions that could lead to my identification.

Francis Salmon

[Occasional 10 Splitter]

> This is not to cast aspersions on the
> various authors; rather it is simply what
> progress is all about. I know of no
> blackjack book, published before BJA3, that
> has a complete set of perfectly accurate
> indices for the count system included in the
> book. Of course, some do a better job than
> others, but all have errors, some of which
> are just egregious typos (see "Million
> Dollar Blackjack," for example, for
> some horrendous ones).

How good are the indices in Wong's Professional Blackjack, 1981 ed. (rounded) and 1994 ed. truncated? Is there much to be gained by generating new floored indices?

[Francis Salmon]

May be I shouldn't reveal this professional secret as I have a horrid feeling you're gonne use this information against me.But blow it!
In fact, it's very simple: I run my program through a loop with TC-increments of 0.1.The program prints out the TC and the EV's for standing and splitting in a particular situation.
The point where the EV for standing becomes higher than the one for hitting is the index.That's all.

Francis Salmon

[G Man]

You wrote:
> When we have an RC of +24 with 5
> decks remaining, I see immediately this is
> TC+4.8 (24 doubled with a decimal point!)and
> I know I should split 10/10v6 and stand on
> 16v9. Don's reader will have to establish
> first is it 4 or is it 5 and for this he has
> to calculate as well.He will floor it to 4
> and as a result he will go wrong in the two
> cases I have mentioned.

I wanted to stay out of this but...
When I see 24 running over 5 I immediately calculate it as +5 (rounding 4.8 to 5 decks).At this TC I will automatically split 10s against 5-6 and stand 16 vs 9. So what`s so hard in this method ? What` s so different than yours ? Tu devrais prendre un peu de recul et cogiter sur les conseils qui te sont gracieusement offerts sur ce site. Tu deviens franchement aga?ant avec tes arguments futiles.