A tale of two games

[h2oski48]

Interesting results from our last three years of play. We have played only quarter triple play, two of us playing as a team all the time. We started keeping good records at the beginning of 2003. Played JOB as we felt like our bankroll was on the short side and thought JOB might be the best situation. We have been able to maintain a CB and BB ratio between .9 and 1% of our cash in so the game is positive for us. Here are our results from 1-03 thru 9-04: Cash in $283k 226,400 hands (counting a "play" on triple play as 3 hands)
Loss rate of 1.73% ($3900). We had 4 royals which is short but doesn't explain the loss rate. During the last part of this play we started to track all of our hands of quads and above. In 61,416 hands we had 111 quads, 34 less than expected, 4 SF, 3 less than expected and no royals. It had always seemed like we were getting hammered on the quads but we didn't know for sure until we started tracking them. We knew the game perfectly and don't believe we were making mistakes to explain the losses.
At this point our bankroll was about 70% of where we started, our CB really helped us to hang in. We took a month off to practice NSUD and decided to try a game with a little higher return knowing that the volatility may finish us off.

Started playing NSU on the same games in the same casinos in Oct 04. We immediately started to win, and as a result,
we have been playing more. Here are the results 10-04 to now: Cash in $316k 253,141 hands with a win rate of 1.38% ($4370). We continued to track all hands of 5OK and above and here are those results:
  Hand Actual Results Expected Results
  5OK 784 788
  WRF 478 483
   4D 52 47
   RF 10 6

Maybe our luck would have changed if we had continued to play JOB. In my opinion there may be a difference in the two games where the final hand in Deuces is influenced more by the draw than JOB. In JOB, it seems like if you start out with a bad deal the hand does not improve with the draw as often as it will with Deuces. If there is any validity to that, then multiplay might be better for Deuces than JOB?

Other observations:
We learned from Bob and Jean the importance of playing the promotions to maximize our CB/BB. That is of the utmost when playing a negative game. Collecting that 1%, win or loose, really adds up through the year. I think I would rather play NSU with 1% CB than FPDW with no CB.
I feel like playing as a team of two tends to make the "ride" a little smoother.
Since this was discussed earlier, we have had royals twice on the same machine this year. One was about 1.5 hours later and the most recent was less than 30 minutes after the first. NSU triple play machines are limited and when we get two together we don't move if we can help it.
I realize that tracking winning hands takes some time but I find it to be very worthwhile info when analyzing your play and results.
Anyone else care to comment on their results with these two games on multiplay? Terry

[Jeff_Cole]

on multiplay? Terry

···

--- In vpFREE@yahoogroups.com, "Terry B" <h2oski48@w...> wrote:

Anyone else care to comment on their results with these two games

=============================================================
No comment, but thanks for sharing your results and analysis.
Jeff

[Jean_Scott1]

<<We learned from Bob and Jean the importance of playing the promotions to
maximize our CB/BB. That is of the utmost when playing a negative game.
Collecting that 1%, win or loose, really adds up through the year. I think
I would rather play NSU with 1% CB than FPDW with no CB.
I feel like playing as a team of two tends to make the "ride" a little
smoother.>>

I agree 100% with your points made above . We also play mostly negative
games, with strong CB and/or BB and/or promotions. CB and BB are always
positive numbers so this factor will somewhat smooth out the roller coaster
ride of the game itself. Brad and I have been playing as one "team" for 21
years, 15 of those at VP - and we have always felt that this smoothed out
the ride in the short term, as well as getting us to the long term faster.

I don't keep exact enough records to discuss the differences between NSUD vs. JoB in our experiences - but NSUD is more volatile than JoB and perhaps your results have caught the upswing part of NSUD's volatility for a long good streak.

···

________________________________________
Jean $�ott - http://www.FrugalGambler.biz
  for VP software and strategy cards.
  NEW - Blackjack Strategy Cards +
              Updated Second Edition of
             "The Frugal Gambler."

[Harry_Porter]

Terry B wrote:

Maybe our luck would have changed if we had continued to play JOB.
In my opinion there may be a difference in the two games where the
final hand in Deuces is influenced more by the draw than JOB. In
JOB, it seems like if you start out with a bad deal the hand does
not improve with the draw as often as it will with Deuces. If there
is any validity to that, then multiplay might be better for Deuces
than JOB?

Interesting observation. I'm going to assume that the comparison is
between NSUD and JOB. This is borne out in the respective variance
numbers for the two games. Dealt Deuces hands typically have a
greater opportunity to improve to paying hands than Jacks (in great
measure to the high hand potential presented by wild cards, when
present). But, on the other hand, when we're talking about the vast
majority of hands that pay 50 credits or less, Jacks has a 6% edge in
return. You're more dependent upon the potential to improve your hand
on the draw in Deuces than Jacks.

Bottom line, there are two key considerations when it comes to
multiplay: game ER and variance. If you were to equalize returns of
the two games via different cb rates (factoring in any other cash
incentives), Deuces would be put at a disadvantage due to its higher
variance ... this translates to greater downside risk and higher
bankroll requirement.

Given equivalent cb rates, clearly NSUD has an ER advantage. However,
assuming cb yields a positive opportunity in both games, Jacks
potentially could still be the better play vs. NSUD when bankroll
limitations are considered.

The variance increase in moving from Jacks to NSUD is magnified for
multiplay vs. single line play. (The difference in variance sourced
from the draw in each game increases with each additional line
played.) The increased variance for Deuces vs. Jacks may place the
player in a position where ROR considerations makes Deuces an
imprudent play while Jacks is still within tolerable constraints.

Anyone who has approached multiplay of each game has grasped how the
loss potential threat becomes much more acute with deuces.

Other observations:
We learned from Bob and Jean the importance of playing the
promotions to maximize our CB/BB. That is of the utmost when
playing a negative game. Collecting that 1%, win or lose, really
adds up through the year. I think I would rather play NSU with 1%
CB than FPDW with no CB.

I grasp the appeal that, loosely paraphrasing, 1% in the hand is worth
more than (at least) 1% in the bush.

But while that 1% serves to reasonably equalize returns, it does
nothing to increase the assurance that you'll realize a given return
from your play. The variance of each game roughly translates to the
likelihood your play return will approximate the ER of the game and
the variance of these games is very close.

Given the option between these two plays in two difference casinos,
I'd be roughly indifferent to the plays and make my selection solely
upon my casino preference.

I feel like playing as a team of two tends to make the "ride" a
little smoother.

I understand this sentiment and it's been expressed by others. For
myself, I describe my wife as being my "anchor" in the casino. When
my play is going rough, there's a strong chance that her play is
smooth or turning a profit.

But this is merely a matter of selective perception. The truth is
that play of $20K of a game by two people at a given denomination is
going to be equally volatile as the same amount of play by a single
person at the same denom. It only stands to reason.

Of course, if you consider $20K play achieved by the single player
through play at twice the denomination, no doubt the completion of
that play by two at half the denomination and twice as many hands will
smooth things out. It's the "twice as many hands" that's the key.

- Harry

[Suzan_Sabyl_Landrum]

Hi,

I will confess that I am not an expert in statistics and issues of
variance, but from my understanding, the more lines you play, the less
one will notice the variance increase moving from jacks to NSUD and
the more likely they will be to have their results reflect the norm.
So if you normally play 100-play jacks and switch to 100-play NSUD,
the difference in variance between the two games will be less apparent
and one is more likely to experience results closer to the expected EV
than if one were to move from single line jacks to single line deuces.
So basically the variance will be higher playing single line than
playing 100-play. However, the jump from single line $.25 to
multi-line $.25 is a jump in limits. So it might feel like multi-line
has the larger variance. Where you will see the results more clearly
is if you were to play both $1 5 coin and 100 line penny for 5 coin.
The bet amount total is the same, but your variance will be much, much
lower on the penny 100 line, all else being equal.

If anyone is more adept at these issues, please correct me if I am
wrong on this.

Sabyl

···

On Thu, 13 Oct 2005 09:32:55 -0000, Harry wrote:

The variance increase in moving from Jacks to NSUD is magnified for
multiplay vs. single line play. (The difference in variance sourced
from the draw in each game increases with each additional line
played.) The increased variance for Deuces vs. Jacks may place the
player in a position where ROR considerations makes Deuces an
imprudent play while Jacks is still within tolerable constraints.

Anyone who has approached multiplay of each game has grasped how the
loss potential threat becomes much more acute with deuces.

[Harry_Porter]

sabyl55 wrote:

I will confess that I am not an expert in statistics and issues of
variance, but from my understanding, the more lines you play, the
less one will notice the variance increase moving from jacks to NSUD
and the more likely they will be to have their results reflect the
norm.
So if you normally play 100-play jacks and switch to 100-play NSUD,
the difference in variance between the two games will be less
apparent and one is more likely to experience results closer to the
expected EV than if one were to move from single line jacks to
single line deuces.
So basically the variance will be higher playing single line than
playing 100-play. However, the jump from single line $.25 to
multi-line $.25 is a jump in limits. So it might feel like
multi-line has the larger variance. Where you will see the results
more clearly is if you were to play both $1 5 coin and 100 line
penny for 5 coin. The bet amount total is the same, but your
variance will be much, much lower on the penny 100 line, all else
being equal.

There is a mix of perceptions here. Your final comment comparing $1
single play and penny 100 play is right on. And, as you earlier
suggest, it is the case that for a given wager (such as $5 in this
example), multiline play can be expected to more closely approximate
play ER than single line for any given number of plays.

But neither of these two facts bear on a comparison of JB and NSUD
variances. The difference between NSUD and JB variance increases with
each additional line played in multiline and, for any given number of
plays, NSUD results can be increasingly expected to vary from ER by a
greater degree than JB.

As I stated before, those who actively dabble in multiline play will
attest that the perceived differences in bankroll fluctuation between
NSUD and JB play become more acute as a greater number of lines are
played.

For further reference, you may wish to link to Michael Shackleford's
article on this topic at:
http://wizardofodds.com/videopoker/appendix3.html

- Harry