Don Schlesinger: Announcing Contest #1

[Don Schlesinger]

Recently, on these pages, J Morgan challenged you to find the closest basic strategy play, given any combination of standard casino rules. For the purposes of this contest, let's define our rules options as follows: 1, 2, 4, 6, or 8 decks; h17 or s17; we may resplit three times, to a total of four hands; das or ndas; late surrender or no surrender; doubling on any two cards, or perhaps just on certain totals; aces may be split once only, and split aces receive one card only.

By "closest play," we mean the play with the smallest numerical difference of EV between the correct basic strategy play and the second-best play, as determined by precise combinatorial analysis. Of course, you may use any means to determine your answers, however, our judges' answers will be determined by CA.

To be eligible to win Contest #1, we'd like to challenge you to find not only the closest such play, but the TOP FIVE closest BS plays, in order!

For each play, you must list the player's holding (each of the two cards, and not just the two-card total), the dealer's up-card, and the precise combination of rules, as per above, of the game being played. It is understood that those rules may change from play to play; you do not need to have the same set of rules assumptions for all five plays. To make things interesting (variety is the spice of life!), we want the five holdings to all be different, that is, once you've used a certain hand against a dealer's up-card, that same hand may not appear again, with different rules assumptions. So, you'll need to determine which set of rules makes for the closest decision, and use that holding one time only.

DO NOT POST your answers to this page!! Please submit your responses to [email protected]. The winner will be the person who will have submitted the correct answer (or, failing a perfect answer, the answer that lists the most correct responses from among the five). In the event of a tie, accuracy of the rankings will be used as the tie-breaker. If there are still ties, the winner will be picked in a random drawing conducted by our Webmaster, Viktor Nacht.

Deadline for submission of your entry is 11:59 p.m., PT, September 22, 2002.

The prize you are playing for is: Any book or video from the Fall 2002 RGE catalog; any piece of software contributed by a Master; or a one-year subscription to Trackjack or Las Vegas Slot Club and Promotions Guide.

Good luck to one and all!

The Masters

[MathProf]

I had some questions about the rules for the contest. First, I assume that BS means the first round off the top of the pack. That is, we can't use cut card effects.

Second Question: Must we use an integral number of decks. Could do something 2.5 decks? In BJ, suits have significance, so there is really no mathematical reason why we couldn't play the game with 10 quarter decks, or 11, or 13, or something else.

> Recently, on these pages, J Morgan
> challenged you to find the closest basic
> strategy play, given any combination of
> standard casino rules. For the purposes of
> this contest, let's define our rules options
> as follows: 1, 2, 4, 6, or 8 decks; h17 or
> s17; we may resplit three times, to a total
> of four hands; das or ndas; late surrender
> or no surrender; doubling on any two cards,
> or perhaps just on certain totals; aces may
> be split once only, and split aces receive
> one card only.

> By "closest play," we mean the
> play with the smallest numerical difference
> of EV between the correct basic strategy
> play and the second-best play, as determined
> by precise combinatorial analysis. Of
> course, you may use any means to determine
> your answers, however, our judges' answers
> will be determined by CA.

> To be eligible to win Contest #1, we'd like
> to challenge you to find not only the
> closest such play, but the TOP FIVE closest
> BS plays, in order!

> For each play, you must list the player's
> holding (each of the two cards, and not just
> the two-card total), the dealer's up-card,
> and the precise combination of rules, as per
> above, of the game being played. It is
> understood that those rules may change from
> play to play; you do not need to have the
> same set of rules assumptions for all five
> plays. To make things interesting (variety
> is the spice of life!), we want the five
> holdings to all be different, that is, once
> you've used a certain hand against a
> dealer's up-card, that same hand may not
> appear again, with different rules
> assumptions. So, you'll need to determine
> which set of rules makes for the closest
> decision, and use that holding one time
> only.

> DO NOT POST your answers to this page!!
> Please submit your responses to
> [email protected]. The winner will be the
> person who will have submitted the correct
> answer (or, failing a perfect answer, the
> answer that lists the most correct responses
> from among the five). In the event of a tie,
> accuracy of the rankings will be used as the
> tie-breaker. If there are still ties, the
> winner will be picked in a random drawing
> conducted by our Webmaster, Viktor Nacht.

> Deadline for submission of your entry is
> 11:59 p.m., PT, September 22, 2002.

> The prize you are playing for is: Any book
> or video from the Fall 2002 RGE catalog; any
> piece of software contributed by a Master;
> or a one-year subscription to Trackjack or
> Las Vegas Slot Club and Promotions Guide.

> Good luck to one and all!

> The Masters

[Karel]

Yes, I guess we assume an integral number of decks.

> Second Question: Must we use an integral
> number of decks. Could do something 2.5
> decks? In BJ, suits have significance, so
> there is really no mathematical reason why
> we couldn't play the game with 10 quarter
> decks, or 11, or 13, or something else.

[ET Fan]

I thought he was pretty clear about your second question...

let's define our rules options as follows: 1, 2, 4, 6, or 8 decks...

Sounds like you're planning to win, in which case my optimal strategy for the contest becomes very clear: don't bother!

[Don Schlesinger]

> I thought he was pretty clear about your
> second question...

> let's define our rules options as follows:
> 1, 2, 4, 6, or 8 decks... Sounds like
> you're planning to win, in which case my
> optimal strategy for the contest becomes
> very clear: don't bother!

The game begins with 1, 2, 4, 6, or 8 freshly shuffled decks. BS, by conventional definition, is the optimal play off the top of the pack.

As for the optimal contest strategy, well . . . there may be a better way! :-)

Don

[MathProf]

First, I apologize for not reading the directions more clearly. I do agree with ET Fan that there were spelled out clearly in the first post. Sorry to have troubled you.

Also, I don't think I will much time to participate. I may have time to make a guess and send in an entry. By make a guess, I mean I guess at what he top 7 or 8 plays are, compute and rank them, and then send in the top 5. But it would really only be a guess. I don't think I will have time to set up an exhaustive search.

> The game begins with 1, 2, 4, 6, or 8
> freshly shuffled decks. BS, by conventional
> definition, is the optimal play off the top
> of the pack.

> As for the optimal contest strategy, well .
> . . there may be a better way! :-)

> Don

[Don Schlesinger]

> I don't think I will have time to set
> up an exhaustive search.

I'm beginning to wonder if we made the first contest too hard. I don't want to divulge how one might go about ascertaining the answers, but I'm a bit surprised that we have had almost 100 people read the contest, yet none have sent in responses yet.

Get to it, guys!!

Don