bankroll requirements for VP versus blackjack

[Donald_Ross]

If a person had a goal of winning $20.00/hour playing either video poker or blackjack, which game would require a larger bankroll? Let's assume the game specifics are $1.00 FPDW with perfect play and 6 deck, hit on soft 17, double on any first 2 cards, double after splitting, and 75% penetration. Let's also assume a 5% risk of ruin. I'm guessing the video poker bankroll needs to be larger. I think this will make for an interesting discussion.

···

---------------------------------
Yahoo! Mail
Bring photos to life! New PhotoMail makes sharing a breeze.

[Non-text portions of this message have been removed]

[mklpryy24]

If a person had a goal of winning $20.00/hour playing either video

poker or blackjack, which game would require a larger bankroll? Let's
assume the game specifics are $1.00 FPDW with perfect play and 6 deck,
hit on soft 17, double on any first 2 cards, double after splitting,
and 75% penetration. Let's also assume a 5% risk of ruin. I'm
guessing the video poker bankroll needs to be larger. I think this
will make for an interesting discussion.

******************************************************************
  
Yes, but no $fpdw left, but assuming they are, you will need to play

800 hands an hour perfectly, not impossible, to average $28 an hour,
but it is a positive play so a large enuf bankroll will reduce or
eliminate the ROR,

The BJ conditions you describe are good , but are stil about a .33%
edge for the house , the hit on soft 17 & no surrender are the biggest
reasons & for any negative game you can not bring enuf money to avoid
the risk of going broke nor can you assume any hourly win can be had if
you flat bet every hand.

BUT if you are counting cards & adjusting both your bets AND your
playing stratergy to the count & can avoid being notice , then it is
possibile but i have no idea how you would caculate the bankroll needed.
a lot would depend on the amount of your base bet.
Maybe BJ21.com can be a help in that

Thanks for reminding me why i switched to VP

M J

···

--- In vpFREE@yahoogroups.com, Donald Ross <hedonist144@...> wrote:

[Dunbar]

The first part of your question is easy to answer with Dunbar's Risk
Analyzer for Video Poker. For a 5-coin $1 FPDW game, you need very
close to $23,000 to have just a 5% RoR. To make $20/hour, you would
need to play a very achievable 525 hands/hr. These numbers would of
course change substantially if there were cashback, player errors,
or tipping involved. For example, if you tip $20 for an RF and $10
for quads, the required bankroll rises to $24,400 and the speed that
you need goes up to 562 hands/hr.

To find a precise bankroll for the blackjack game, you would have to
define how much of a bet spread you are willing to use as well as
the speed of your game. If I were at home, I could look up some
reasonable EV and SD figures in Arnold Snyder's "Beat The 6-Deck
Game" book, or Don Schlesinger's "Blackjack Attack", but I am on an
extended trip away from my own library. Also, one of the good
simulation programs would give you the necessary ev and sd figs.

Looking at the only blackjack book I have with me, Stanford
Wong's "Professional Blackjack", it looks like a game with 6 decks
with your rules but with 80% penetration (62 cards cut off) would
return roughly $12/100 hands to someone willing to bet between $10
and $100/hand. The difference between 52 cards and 62 cards is
about $3/100 hands. So I expect the difference from 62 to 78 cards
(75%) would drop the ev to roughly $10/100 hands. Betting twice
those amounts would deliver the desired $20/hour if you can average
100 hands/hr. (that's about the speed you expect with 3 players at
a table.) I'm estimating the standard deviation of this game
(betting $20 to $200) at $845/100 hands. Putting those figures into
Dunbar's Risk Analyzer for Blackjack, I get a bankroll of $53,000 to
have a longterm 5% RoR.

So it looks to me that FPDW requires about half the bankroll of the
blackjack game that you described. But blackjack is VERY sensitive
to rules, penetration, and bet spread. With a better blackjack
game, blackjack might require a smaller bankroll than the same-ev VP
game. In addition, I've made a LOT of approximations in calculating
the ev and standard deviation of your blackjack game. If someone
has EV and standard deviation data for a particular blackjack game,
I'll be glad to post the bankroll/RoR relationship for that game.

--Dunbar

>
> If a person had a goal of winning $20.00/hour playing either

video

poker or blackjack, which game would require a larger bankroll?

Let's

assume the game specifics are $1.00 FPDW with perfect play and 6

deck,

hit on soft 17, double on any first 2 cards, double after

splitting,

and 75% penetration. Let's also assume a 5% risk of ruin. I'm
guessing the video poker bankroll needs to be larger. I think

this

···

--- In vpFREE@yahoogroups.com, "mklpryy24" <mklpryy24@...> wrote:

--- In vpFREE@yahoogroups.com, Donald Ross <hedonist144@> wrote:
will make for an interesting discussion.
>

******************************************************************
  

> Yes, but no $fpdw left, but assuming they are, you will need to

play

800 hands an hour perfectly, not impossible, to average $28 an

hour,

but it is a positive play so a large enuf bankroll will reduce or
eliminate the ROR,

The BJ conditions you describe are good , but are stil about

a .33%

edge for the house , the hit on soft 17 & no surrender are the

biggest

reasons & for any negative game you can not bring enuf money to

avoid

the risk of going broke nor can you assume any hourly win can be

had if

you flat bet every hand.

BUT if you are counting cards & adjusting both your bets AND

your

playing stratergy to the count & can avoid being notice , then it

is

possibile but i have no idea how you would caculate the bankroll

needed.

a lot would depend on the amount of your base bet.
Maybe BJ21.com can be a help in that

Thanks for reminding me why i switched to VP

M J

[Dunbar]

The BJ conditions you describe are good , but are stil about

a .33%

edge for the house , the hit on soft 17 & no surrender are the

biggest

reasons & for any negative game you can not bring enuf money to

avoid

the risk of going broke nor can you assume any hourly win can be

had if

you flat bet every hand.

The BJ conditions are not particularly good. The house edge is
about 0.6%. (The "hit on soft17" rule is very costly).

There is a nice blackjack basic strategy calculator at
www.wizardofodds.com/blackjack

BUT if you are counting cards & adjusting both your bets AND

your

playing stratergy to the count & can avoid being notice , then it

is

possibile but i have no idea how you would caculate the bankroll

needed.

a lot would depend on the amount of your base bet.
Maybe BJ21.com can be a help in that

My program, Dunbar's Risk Analyzer for Blackjack, will give you that
kind of info, but you need to know your EV and the standard
deviation of one hand (or 100 hands). DRA-BJ will calculate the ev
and SD for you, if you know how often you are making bets of each
size. Or, you can get that kind of EV and SD info from the books I
mentioned in my first post in this thread or from sim programs.

--Dunbar

[nightoftheiguana2000]

a blackjack calculator is here:
http://www.bjstats.com/bjre.asp
it reports EV and N0, but Kelly bankroll is N0 x EV
another one is here:
http://www.bjstats.com//bjbr.asp
blackjack is a lot more complicated than most people think, if you
expect to win, most modern games have poor penetration particularly on
hand dealt games where many dealers will preferential shuffle a
favorable deck, then there is mindplay:
http://www.google.com/search?q=mindplay+blackjack

···

--- In vpFREE@yahoogroups.com, "dunbar_dra" <h_dunbar@...> wrote:

>
> The BJ conditions you describe are good , but are stil about
a .33%
> edge for the house , the hit on soft 17 & no surrender are the
biggest
> reasons & for any negative game you can not bring enuf money to
avoid
> the risk of going broke nor can you assume any hourly win can be
had if
> you flat bet every hand.

The BJ conditions are not particularly good. The house edge is
about 0.6%. (The "hit on soft17" rule is very costly).

There is a nice blackjack basic strategy calculator at
www.wizardofodds.com/blackjack

> BUT if you are counting cards & adjusting both your bets AND
your
> playing stratergy to the count & can avoid being notice , then it
is
> possibile but i have no idea how you would caculate the bankroll
needed.
> a lot would depend on the amount of your base bet.
> Maybe BJ21.com can be a help in that

My program, Dunbar's Risk Analyzer for Blackjack, will give you that
kind of info, but you need to know your EV and the standard
deviation of one hand (or 100 hands). DRA-BJ will calculate the ev
and SD for you, if you know how often you are making bets of each
size. Or, you can get that kind of EV and SD info from the books I
mentioned in my first post in this thread or from sim programs.

--Dunbar

[Dunbar]

--- In vpFREE@yahoogroups.com, "nightoftheiguana2000"
<nightoftheiguana2000@...> wrote:

a blackjack calculator is here:
http://www.bjstats.com/bjre.asp
it reports EV and N0, but Kelly bankroll is N0 x EV
another one is here:
http://www.bjstats.com//bjbr.asp
blackjack is a lot more complicated than most people think, if you
expect to win, most modern games have poor penetration particularly on
hand dealt games where many dealers will preferential shuffle a
favorable deck, then there is mindplay:
http://www.google.com/search?q=mindplay+blackjack

Great link. I agree that blackjack is much more difficult than video
poker. While I was able to answer the original poster's bankroll
question immediately with regard to FPDW, there is a huge range of
possible answers to the blackjack game that he or she specified. The
bankroll required for a $20/hr win rate will depend on the count used,
the spread, the ramp (how the spread is used), the number of players at
the table, and the amount of cover that is used. If we assume optimum
play, then we can assume an optimum bet ramp and no cover.

As I said earlier, I need the standard deviation of a game to calculate
the Risk vs Bankroll numbers. I can get the sd from "NO". Using
the "NO" calculator at the link you provided to the CVCX (Casino
Verite) site, I chose an optimized 1-12 bet range, using HiLo with the
so-called Illustrious 18 play variation indeces. For this game, and
the specifications of the original poster, NO is 87,616. The proposed
benchmark goal was to make $20/hr. At 100 hands/hr, that would be ev=
0.2/hand.

NO = sd^2/ev^2. So, sd^2=NO*ev^2
sd^2 = 87616 * 0.2^2
sd^2 = 3504
sd=SQRT(3504)= 59.2/hand

Putting ev=.2 and sd=59.2 into Dunbar's Risk Analyzer for Blackjack, I
find that a bankroll of $26,300 is enough for a 5% RoR.

The $26,300 blackjack bankroll is a little more than the bankroll
required for a $20/hr (about 500 hands/hr) player of $1 FPDW. But even
slight modifications of my assumptions would yield a bankroll that was
smaller than the VP bankroll. For example, if we assume a heads-up
player, we would then get at least 200 hands/hr, and our ev would only
need to be 0.1/hand to make $20/hr. Our sd/hand is now 1/2 what it was
before, or 29.6.

Under THOSE conditions, DRA-BJ says you only need a $13,200 bankroll
for a 5% RoR.

Bottom line? There are a lot more variables in a blackjack game to
consider. Good blackjack games will require a smaller bankroll than
FPDW; poor blackjack games will require a bigger bankroll.

But then we haven't considered cashback!

--Dunbar