pm: SCORE

[pm]

What is a SCORE (as it applies to evaluating a game)? Reading BJA3, I thought it was just the hourly win that a particular game will get you (playing all or back-counting), but I'm guessing that I missed the whole point of the SCORE.

[pm]

> What is a SCORE (as it applies to evaluating
> a game)? Reading BJA3, I thought it was just
> the hourly win that a particular game will
> get you (playing all or back-counting), but
> I'm guessing that I missed the whole point
> of the SCORE.

Actually, let me see if I have this right. A SCORE is assigned to a particular game (playing all or back-counting) setting the bankroll and RoR as constants ($10K and 13.5%). Your winrate at that particular game may be more than the SCORE because you could have a bigger bank and/or a higher RoR; the point of the SCORE is to show you how lucrative your game & style of play are relative to other games.

Any confirmation?

And how is RoR being kept at a constant 13.5% throughout all the games listed in the tables? It says in the SCORE chapter that an optimal betting schedule would be used in all games. An optimal schedule should remain constant regardless of rules and play all vs. wonging, shouldn't it (granted, the edges yielded by specific true counts would vary from rule set to rule set)? If so, how is it that different bet spreads can be employed in the same games while keeping the RoR constant? If not, how do you figure out the bet unit that is assumed for a particular SCORE?

Some help from Don, maybe?

[Don Schlesinger]

> What is a SCORE (as it applies to evaluating
> a game)? Reading BJA3, I thought it was just
> the hourly win that a particular game will
> get you (playing all or back-counting), but
> I'm guessing that I missed the whole point
> of the SCORE.

Before one consults the Chapter 10 tables of BJA3, the entire SCORE chapter (9) precedes them. I moved that chapter to be in front of Chapter 10 for BJA3, because I thought it was more logical to present the material that way. If you reread pp. 152-159, you will see a very thorough discussion of the concept of SCORE, exactly what it means, and how it is calculated.

Don

[Don Schlesinger]

> Actually, let me see if I have this right. A
> SCORE is assigned to a particular game
> (playing all or back-counting) setting the
> bankroll and RoR as constants ($10K and
> 13.5%).

So far, so good! :-)

> Your winrate at that particular game
> may be more than the SCORE because you could
> have a bigger bank and/or a higher RoR; the
> point of the SCORE is to show you how
> lucrative your game & style of play are
> relative to other games.

> Any confirmation?

Right on!

> And how is RoR being kept at a constant
> 13.5% throughout all the games listed in the
> tables? It says in the SCORE chapter that an
> optimal betting schedule would be used in
> all games. An optimal schedule should remain
> constant regardless of rules and play all
> vs. wonging, shouldn't it (granted, the
> edges yielded by specific true counts would
> vary from rule set to rule set)?

An optimal betting schedule remains constant for the game and rules set being played. But it varies from one such set to another. Just look at all the different bet ramps and bet amounts in the Chapter 10 charts.

> If so, how
> is it that different bet spreads can be
> employed in the same games while keeping the
> RoR constant?

Because as you change the bet spread, you reconfigure the optimal bet ramp and the unit size.

> If not, how do you figure out
> the bet unit that is assumed for a
> particular SCORE?

See above. Each optimization is different for each set of rules and bet ramp.

> Some help from Don, maybe?

Done!

Don

[pm]

I'm sure I'll have some other questions to bother you with.

[Don Schlesinger]

> I'm sure I'll have some other questions to
> bother you with.

That's what we're here for.

Don

[paranoid android]

Norm, I see in your software you use the term c-SCORE sometimes rather than SCORE. What is the c for? I suppose it's explained in the manual, but we all know that manuals are for wimps. ;-)

[Don Schlesinger]

> Norm, I see in your software you use the
> term c-SCORE sometimes rather than SCORE.
> What is the c for? I suppose it's explained
> in the manual, but we all know that manuals
> are for wimps. ;-)

After I made a big deal about preserving the sanctity of the SCORE appellation (see "More on SCORE," in Chapter 9 of BJA3), Richard Reid suggested that we call hourly win rates that aren't exactly SCOREs, in the strictest sense of the term, "c-SCOREs," where the "c" stood for "custom." But, I lost this battle, as reported in the aforementioned article, and SCORE has become somewhat generic, being synonymous to many with "hourly win rate," no matter what the terms.

Don

[paranoid android]

> hourly win rates that aren't exactly SCOREs,
> in the strictest sense of the term,
> "c-SCOREs,"

Why aren't they exactly SCOREs? Still a bit confused.

[Cacarulo]

> Why aren't they exactly SCOREs? Still a bit
> confused.

There are some fixed parameters in the term SCORE like:

$10,000 bankroll
13,53% ROR
I18
100 RPH (rounds per hour)

For comparison purposes those parameters are ok but you can use others. For example, my "scores" (note the lower case) are based on C22. Since one of the parameters is not repected it is no longer correct to still call it SCORE. For that reason some people use the term c-SCORE instead.
Others like myself use simply "score".

Regards,
Cac

[Sun Runner]

> For comparison purposes those parameters are
> ok but you can use others. For example, my
> "scores" (note the lower case) are
> based on C22. Since one of the parameters is
> not repected it is no longer correct to
> still call it SCORE. For that reason some
> people use the term c-SCORE instead.
> Others like myself use simply
> "score".

With all due respect, you don't use lower case "score."

As you may or may not know I have been searching all your posts recently regarding KO related information. As late as this week, in a posting string you again referred to the work you previously posted and referred to the SCOREs.

A post you put up regarding this issue in January this year referred to your work as SCORE.

Understand, your work has been a great resource for me personally, and you always (or most always) identify the parameters that make up your SCORE .. but the point here is, even when you (we) think we are not treading on the concept, we do.

I personally believe that SCORE (a fixed set of parameters) has value as a measuring tool. And when you see it referred to, you can know what you are looking at.

I also think the c-SCORE idea has merit when posting anything else.

Just my observation and thanks for the continuing good work.

[Cacarulo]

> With all due respect, you don't use lower
> case "score."

You've got me on this one. The thing is that I've unconciously incorporated the C22 to the SCORE definition (instead of the I18). That's why all of my posts refer to the term SCORE. But, when I change other parameters like speed (RPM) or ROR I do use "score".

> I personally believe that SCORE (a fixed set
> of parameters) has value as a measuring
> tool. And when you see it referred to, you
> can know what you are looking at.

Correct.

> I also think the c-SCORE idea has merit when
> posting anything else.

> Just my observation and thanks for the
> continuing good work.

You're welcome.

Sincerely,
Cac