NO Formula (In a Table Format)

[brucem]

I just finished rereading the NO explanation on VPfree. The formula
is shown below.
Hands=Variance/(er-1+cb)^2

It occured to me it would be easy to setup a NO table where the NO
value is already computed so people who find it difficult to solve
formulas like NO, would not have to do it. And it may simplify
calculating NO inside a casino. A person would need to know
the variance of the games they play, and how to calculate (er-1+cb)
^2. That's fairly easy using VP software and casino CB info.

One axis would have variance values (in increments, beginning around
18 and up I suppose. The other axis would have the (er-1+cb)^2
component, also in increments starting at .1 and up, at .1 increments
(or maybe less). Cells within the table would be the NO value
(number of hands).

I don't know how useful this would be ... maybe not useful at all.

[brucem]

In my earlier post I suggested a way to create a table of NO
values, to eliminate the need to solve the NO equation.

However, while the idea of a NO table is (I believe) useful, I
think there is a much better way to implement the table than I
initially proposed.

In the "new improved" method the vertical axis in the table is a list
of the 10-15 best VP games (9/6JB, 10/7DDB, NSUD, FPDW, etc.). One
way to order these games is by increasing variance, but it's better I
think to order the games according to EV, from lowest to highest.

The horizontal axis would be CB values, beginning with 0 and
increasing in 15-20 small increments along the axis.

The NO formula is solved for each table cell. That is, each table
cell would contain the NO "number of games" value.

This table design is better because the player needs to know only two
things to use it: the game that interests them, and the CB for the
casino with that game. That's it!

A NO table, basically, eliminates the need to compute NO, which for
some isn't easy to do. But, it seems to me, the table serves another
purpose too ... to evaluate alternative playing choices.

To give a very simple example, assume you play 2-4 different games
in 2 casinos. On certain days one casino offers 3x CB, but the
games have a lower EV (or higher variance) than the games in the
other casino. Which casino to visit? This is where the table comes
in handy. To decide, check the NO value for the game(s) in the
casino with 3x CB. Then check the NO value for the game(s) in the
other casino. Go to the casino (and play the game) with the lowest
NO value.

That is the general rule when using the table. Play the game with
the lowest NO value ... because NO correlates three important
criteria: EV, CB, and Variance, expressed in terms of a "number of
games".

I'm not going to try to create a NO table. If I tried, I'd probably
screw it up. However, if someone wants to try I'd like to see the
finished product, to see if it serves a useful purpose.

···

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

[Dunbar]

I think that's a useful suggestion, brumar. I've created a table
with the NO for 17 games and cashback up to 1% in increments of
0.1%. Here is a bit of the table, though I think it won't maintain
table format in a post:

   VALUES FOR "NO" FOR VARIOUS GAMES AND LEVELS OF CASHBACK

                                   CASHBACK
GAME var ev 0.00% 0.10% 0.20%
Loose Deuces 70.3 100.97% 748,000 615,000 515,000
Deuces Wild 25.8 100.76% 446,000 348,000 280,000
All-American 26.8 100.72% 515,000 397,000 316,000
JokerWild,KOB 26.2 100.65% 629,000 472,000 367,000

I'm rounding the values of NO up to the nearest 1000. My figures
for variance and ev come from WinPoker.

If the vpFREE Administrator likes the idea, I can send the entire
table as an email attachment.

--Dunbar

In my earlier post I suggested a way to create a table of NO
values, to eliminate the need to solve the NO equation.

However, while the idea of a NO table is (I believe) useful, I
think there is a much better way to implement the table than I
initially proposed.

In the "new improved" method the vertical axis in the table is a

list

of the 10-15 best VP games (9/6JB, 10/7DDB, NSUD, FPDW, etc.).

One

way to order these games is by increasing variance, but it's

better I

think to order the games according to EV, from lowest to highest.

The horizontal axis would be CB values, beginning with 0 and
increasing in 15-20 small increments along the axis.

The NO formula is solved for each table cell. That is, each table
cell would contain the NO "number of games" value.

This table design is better because the player needs to know only

two

things to use it: the game that interests them, and the CB for the
casino with that game. That's it!

A NO table, basically, eliminates the need to compute NO, which

for

some isn't easy to do. But, it seems to me, the table serves

another

purpose too ... to evaluate alternative playing choices.

To give a very simple example, assume you play 2-4 different games
in 2 casinos. On certain days one casino offers 3x CB, but the
games have a lower EV (or higher variance) than the games in the
other casino. Which casino to visit? This is where the table

comes

in handy. To decide, check the NO value for the game(s) in the
casino with 3x CB. Then check the NO value for the game(s) in the
other casino. Go to the casino (and play the game) with the

lowest

NO value.

That is the general rule when using the table. Play the game with
the lowest NO value ... because NO correlates three important
criteria: EV, CB, and Variance, expressed in terms of a "number of
games".

I'm not going to try to create a NO table. If I tried, I'd

probably

···

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

--- In vpFREE@yahoogroups.com, "brumar_lv" <brumar_lv@> wrote:
screw it up. However, if someone wants to try I'd like to see the
finished product, to see if it serves a useful purpose.

[Peter]

Interesting ordering...

DW (.2) 280
AA (.2) 316
DW (.1) 348
JW (.2) 367
AA (.1) 397
DW (.0) 446
JW (.1) 472
AA (.0) 515
LD (.2) 515
LD (.1) 615
JW (.0) 629
LD (.0) 748

.....BL

···

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

I think that's a useful suggestion, brumar. I've created a table
with the NO for 17 games and cashback up to 1% in increments of
0.1%.

[Dunbar]

The piece of the table I posted earlier showed up better than I
expected. So here's the whole thing.

      VALUES OF "NO FOR VARIOUS GAMES AND LEVELS OF CASHBACK

CASHBACK

GAME var ev 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.8% 1.0% 1.5% 2.0%
Loose Deuces 70.3 100.97% 748,000 615,000 515,000 437,000 375,000
326,000 286,000 225,000 182,000 116,000 80,000 Deuces Wild 25.8 100.76%
446,000 348,000 280,000 230,000 192,000 163,000 140,000 106,000 84,000
51,000 34,000 All-American 26.8 100.72% 515,000 397,000 316,000 257,000
213,000 180,000 154,000 116,000 91,000 55,000 37,000 Joker Wild--Kings
or better 26.2 100.65% 629,000 472,000 367,000 294,000 240,000 200,000
169,000 126,000 97,000 57,000 38,000 Double Double Jackpot 38.2 100.35%
3,173,000 1,913,000 1,278,000 913,000 685,000 533,000 427,000 291,000
211,000 113,000 70,000 Deuces Deluxe 26.0 100.32% 2,490,000 1,453,000
951,000 670,000 498,000 384,000 306,000 207,000 149,000 79,000 49,000
Double Bonus 28.3 100.17% 9,496,000 3,806,000 2,037,000 1,266,000
863,000 625,000 474,000 299,000 206,000 102,000 60,000 Pick 'Em 15.0
99.95% net loss 60,023,000 6,670,000 2,401,000 1,225,000 742,000 497,000
267,000 167,000 72,000 40,000 Four Joker Poker 23.6 99.89% net loss net
loss 30,291,000 6,653,000 2,837,000 1,564,000 989,000 498,000 299,000
123,000 67,000 Aces and Eights 21.7 99.78% net loss net loss net loss
32,470,000 6,574,000 2,736,000 1,491,000 642,000 356,000 133,000 69,000
Not-So-Ugly-Deuces 25.8 99.73% net loss net loss net loss 321,896,000
15,662,000 4,947,000 2,392,000 924,000 487,000 171,000 87,000 Double
Jackpot Poker 22.4 99.63% net loss net loss net loss net loss
298,258,000 13,796,000 4,331,000 1,226,000 569,000 177,000 85,000
Double Joker Poker 22.4 99.63% net loss net loss net loss net loss
298,258,000 13,796,000 4,331,000 1,226,000 569,000 177,000 85,000 Five
Joker Poker 16.8 99.59% net loss net loss net loss net loss net loss
21,452,000 4,735,000 1,116,000 487,000 143,000 67,000 Jacks or Better
(full-pay) 19.5 99.54% net loss net loss net loss net loss net loss
101,259,000 9,425,000 1,651,000 660,000 180,000 82,000 Bonus Deuces
Wild 32.7 99.45% net loss net loss net loss net loss net loss net loss
129,613,000 5,218,000 1,612,000 362,000 156,000 Bonus Poker 20.9 99.17%
net loss net loss net loss net loss net loss net loss net loss net loss
7,587,000 472,000 154,000

"NO" is the number of hands at which the average profit will be one
standard deviation away from zero. In other words, "NO" is the point at
which you have approximately an 82% chance of being ahead.

Deuces Wild is highlighted because it has the lowest value for NO for
all cashback up to 2%. However, if there is 3% cashback in Heaven, then
Pick 'Em would have a lower NO than Deuces Wild.

--Dunbar

Note: ev and variance figures are from WinPoker.

— In vpFREE@yahoogroups.com <mailto:vpF…@…com> ,
"bornloser1537" <bornloser1537@…> wrote:

Interesting ordering...

DW (.2) 280
AA (.2) 316
DW (.1) 348
JW (.2) 367
AA (.1) 397
DW (.0) 446
JW (.1) 472
AA (.0) 515
LD (.2) 515
LD (.1) 615
JW (.0) 629
LD (.0) 748

.....BL

— In vpFREE@yahoogroups.com <mailto:vpF…@…com> ,

"dunbar_dra" <h_dunbar@> wrote:

>
> I think that's a useful suggestion, brumar. I've created a table
> with the NO for 17 games and cashback up to 1% in increments of
> 0.1%.

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

[nightoftheiguana2000]

for the heck of it, I compressed it down into a form that should keep
its shape:
GAME var ev 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.8% 1.0% 1.5% 2.0%
L_D 70 +.97% 748, 615, 515, 437, 375, 326, 286, 225, 182, 116, 80,
D_W 26 +.76% 446, 348, 280, 230, 192, 163, 140, 106, 84, 51, 34,
A_A 27 +.72% 515, 397, 316, 257, 213, 180, 154, 116, 91, 55, 37,
J_K 26 +.65% 629, 472, 367, 294, 240, 200, 169, 126, 97, 57, 38,
DDJ 38 +.35% 3E6, 2E6, 1E6, 913, 685, 533, 427, 291, 211, 113, 70,
D_D 26 +.32% 2E6, 1E6, 951, 670, 498, 384, 306, 207, 149, 79, 49,
D_B 28 +.17% 9E6, 3E6, 2E6, 1E6, 863, 625, 474, 299, 206, 102, 60,
P_E 15 -.05% NEG, 6E7, 7E6, 2E6, 1E6, 742, 497, 267, 167, 72, 40,
4_J 24 -.11% NEG, NEG, 3E7, 7E6, 3E6, 3E6, 989, 498, 299, 123, 67,
A&8 22 -.22% NEG, NEG, NEG, 3E7, 7E6, 3E6, 1E6, 642, 356, 133, 69,
NSU 26 -.27% NEG, NEG, NEG, 3E8, 2E7, 5E6, 2E6, 924, 487, 171, 87,
D_J 22 -.37% NEG, NEG, NEG, NEG, 3E9, 1E7, 4E6, 1E6, 569, 177, 85,
2_J 22 -.37% NEG, NEG, NEG, NEG, 2E9, 1E7, 4E6, 1E6, 569, 177, 85,
5_J 17 -.41% NEG, NEG, NEG, NEG, NEG, 2E7, 5E6, 1E6, 487, 143, 67,
JoB 20 -.46% NEG, NEG, NEG, NEG, NEG, 1E8, 9E6, 2E6, 660, 180, 82,
BDW 33 -.55% NEG, NEG, NEG, NEG, NEG, NEG, 1E8, 5E6, 2E6, 362, 156,
B_P 21 -.83% NEG, NEG, NEG, NEG, NEG, NEG, NEG, NEG, 8E6, 472, 154,

···

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

The piece of the table I posted earlier showed up better than I
expected. So here's the whole thing.

      VALUES OF "NO FOR VARIOUS GAMES AND LEVELS OF CASHBACK

CASHBACK

GAME var ev 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.8% 1.0% 1.5% 2.0%
Loose Deuces 70.3 100.97% 748,000 615,000 515,000 437,000 375,000
326,000 286,000 225,000 182,000 116,000 80,000 Deuces Wild 25.8 100.76%
446,000 348,000 280,000 230,000 192,000 163,000 140,000 106,000 84,000
51,000 34,000 All-American 26.8 100.72% 515,000 397,000 316,000 257,000
213,000 180,000 154,000 116,000 91,000 55,000 37,000 Joker Wild--Kings
or better 26.2 100.65% 629,000 472,000 367,000 294,000 240,000 200,000
169,000 126,000 97,000 57,000 38,000 Double Double Jackpot 38.2 100.35%
3,173,000 1,913,000 1,278,000 913,000 685,000 533,000 427,000 291,000
211,000 113,000 70,000 Deuces Deluxe 26.0 100.32% 2,490,000 1,453,000
951,000 670,000 498,000 384,000 306,000 207,000 149,000 79,000 49,000
Double Bonus 28.3 100.17% 9,496,000 3,806,000 2,037,000 1,266,000
863,000 625,000 474,000 299,000 206,000 102,000 60,000 Pick 'Em 15.0
99.95% net loss 60,023,000 6,670,000 2,401,000 1,225,000 742,000 497,000
267,000 167,000 72,000 40,000 Four Joker Poker 23.6 99.89% net loss net
loss 30,291,000 6,653,000 2,837,000 1,564,000 989,000 498,000 299,000
123,000 67,000 Aces and Eights 21.7 99.78% net loss net loss net loss
32,470,000 6,574,000 2,736,000 1,491,000 642,000 356,000 133,000 69,000
Not-So-Ugly-Deuces 25.8 99.73% net loss net loss net loss 321,896,000
15,662,000 4,947,000 2,392,000 924,000 487,000 171,000 87,000 Double
Jackpot Poker 22.4 99.63% net loss net loss net loss net loss
298,258,000 13,796,000 4,331,000 1,226,000 569,000 177,000 85,000
Double Joker Poker 22.4 99.63% net loss net loss net loss net loss
298,258,000 13,796,000 4,331,000 1,226,000 569,000 177,000 85,000 Five
Joker Poker 16.8 99.59% net loss net loss net loss net loss net loss
21,452,000 4,735,000 1,116,000 487,000 143,000 67,000 Jacks or Better
(full-pay) 19.5 99.54% net loss net loss net loss net loss net loss
101,259,000 9,425,000 1,651,000 660,000 180,000 82,000 Bonus Deuces
Wild 32.7 99.45% net loss net loss net loss net loss net loss net loss
129,613,000 5,218,000 1,612,000 362,000 156,000 Bonus Poker 20.9 99.17%
net loss net loss net loss net loss net loss net loss net loss net loss
7,587,000 472,000 154,000

"NO" is the number of hands at which the average profit will be one
standard deviation away from zero. In other words, "NO" is the point at
which you have approximately an 82% chance of being ahead.

Deuces Wild is highlighted because it has the lowest value for NO for
all cashback up to 2%. However, if there is 3% cashback in Heaven, then
Pick 'Em would have a lower NO than Deuces Wild.

--Dunbar

Note: ev and variance figures are from WinPoker.

— In vpFREE@yahoogroups.com <mailto:vpF…@…com> ,
"bornloser1537" <bornloser1537@> wrote:
>
> Interesting ordering…
>
> DW (.2) 280
> AA (.2) 316
> DW (.1) 348
> JW (.2) 367
> AA (.1) 397
> DW (.0) 446
> JW (.1) 472
> AA (.0) 515
> LD (.2) 515
> LD (.1) 615
> JW (.0) 629
> LD (.0) 748
>
>
> …BL
>
> — In vpFREE@yahoogroups.com <mailto:vpF…@…com> ,
"dunbar_dra" <h_dunbar@> wrote:
> >
> > I think that's a useful suggestion, brumar. I've created a table
> > with the NO for 17 games and cashback up to 1% in increments of
> > 0.1%.
>

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

[brucem]

Thank you for the feedback, and taking the time to actually create a
NO table! I don't have the knowledge needed to select the best 10-15
games, or to choose appropriate CB increments, so I couldn't have
done it without help.

It was mentioned that FPDW has the best NO (up to a point). That's
where comparing NO's seems like a good idea to me, because the CB can
vary from one casino to another, so on any given day a different game
may have a better NO "score" than FPDW. What do you think of my
suggestion that comparing NO scores serve as a way to choose which
game to play? Is that valid, or does it put too much emphasis on NO,
instead of the usual emphasis on the EV+CB?

···

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

I think that's a useful suggestion, brumar. I've created a table
with the NO for 17 games and cashback up to 1% in increments of
0.1%. Here is a bit of the table, though I think it won't maintain
table format in a post:

   VALUES FOR "NO" FOR VARIOUS GAMES AND LEVELS OF CASHBACK

                                   CASHBACK
GAME var ev 0.00% 0.10% 0.20%
Loose Deuces 70.3 100.97% 748,000 615,000 515,000
Deuces Wild 25.8 100.76% 446,000 348,000 280,000
All-American 26.8 100.72% 515,000 397,000 316,000
JokerWild,KOB 26.2 100.65% 629,000 472,000 367,000

I'm rounding the values of NO up to the nearest 1000. My figures
for variance and ev come from WinPoker.

[nightoftheiguana2000]

It works, you could also use the traditional Sharpe Ratio:

(er-1+cb)/sqrt(variance)

http://www.google.com/search?q="Sharpe+Ratio"

(notice: N0=1/(Sharpe Ratio)^2)

···

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

What do you think of my
suggestion that comparing NO scores serve as a way to choose which
game to play? Is that valid, or does it put too much emphasis on NO,
instead of the usual emphasis on the EV+CB?

[Dunbar]

Very nice. It has as much in the way of significant digits as would
be useful in all but the rarest cases.

--Dunbar

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

for the heck of it, I compressed it down into a form that should

keep

···

its shape:
GAME var ev 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.8% 1.0% 1.5% 2.0%
L_D 70 +.97% 748, 615, 515, 437, 375, 326, 286, 225, 182, 116, 80,
D_W 26 +.76% 446, 348, 280, 230, 192, 163, 140, 106, 84, 51, 34,
A_A 27 +.72% 515, 397, 316, 257, 213, 180, 154, 116, 91, 55, 37,
J_K 26 +.65% 629, 472, 367, 294, 240, 200, 169, 126, 97, 57, 38,
DDJ 38 +.35% 3E6, 2E6, 1E6, 913, 685, 533, 427, 291, 211, 113, 70,
D_D 26 +.32% 2E6, 1E6, 951, 670, 498, 384, 306, 207, 149, 79, 49,
D_B 28 +.17% 9E6, 3E6, 2E6, 1E6, 863, 625, 474, 299, 206, 102, 60,
P_E 15 -.05% NEG, 6E7, 7E6, 2E6, 1E6, 742, 497, 267, 167, 72, 40,
4_J 24 -.11% NEG, NEG, 3E7, 7E6, 3E6, 3E6, 989, 498, 299, 123, 67,
A&8 22 -.22% NEG, NEG, NEG, 3E7, 7E6, 3E6, 1E6, 642, 356, 133, 69,
NSU 26 -.27% NEG, NEG, NEG, 3E8, 2E7, 5E6, 2E6, 924, 487, 171, 87,
D_J 22 -.37% NEG, NEG, NEG, NEG, 3E9, 1E7, 4E6, 1E6, 569, 177, 85,
2_J 22 -.37% NEG, NEG, NEG, NEG, 2E9, 1E7, 4E6, 1E6, 569, 177, 85,
5_J 17 -.41% NEG, NEG, NEG, NEG, NEG, 2E7, 5E6, 1E6, 487, 143, 67,
JoB 20 -.46% NEG, NEG, NEG, NEG, NEG, 1E8, 9E6, 2E6, 660, 180, 82,
BDW 33 -.55% NEG, NEG, NEG, NEG, NEG, NEG, 1E8, 5E6, 2E6, 362, 156,
B_P 21 -.83% NEG, NEG, NEG, NEG, NEG, NEG, NEG, NEG, 8E6, 472, 154,

[nightoftheiguana2000]

N0 is more useful, because it gives you the actual number of hands you
have to play to get at least an 84% chance of net winning, play less
than that and your chances are less, play 4xN0 hands and your chances
of net winning go up to 98%. Of course, this assumes you have
sufficient bankroll and don't bust out.

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

···

--- In vpFREE@yahoogroups.com, "brumar_lv" <brumar_lv@> wrote:
> What do you think of my
> suggestion that comparing NO scores serve as a way to choose which
> game to play? Is that valid, or does it put too much emphasis on NO,
> instead of the usual emphasis on the EV+CB?

It works, you could also use the traditional Sharpe Ratio:

(er-1+cb)/sqrt(variance)

http://www.google.com/search?q="Sharpe+Ratio"

(notice: N0=1/(Sharpe Ratio)^2)

[deuceswild1000]

Very nice. It has as much in the way of significant digits as

would

be useful in all but the rarest cases.

--Dunbar

--- In vpFREE@yahoogroups.com, "nightoftheiguana2000"
<nightoftheiguana2000@> wrote:
>
> for the heck of it, I compressed it down into a form that should
keep
> its shape:
> GAME var ev 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.8% 1.0% 1.5%

2.0%

> L_D 70 +.97% 748, 615, 515, 437, 375, 326, 286, 225, 182, 116,

80,

> D_W 26 +.76% 446, 348, 280, 230, 192, 163, 140, 106, 84, 51, 34,
> A_A 27 +.72% 515, 397, 316, 257, 213, 180, 154, 116, 91, 55, 37,
> J_K 26 +.65% 629, 472, 367, 294, 240, 200, 169, 126, 97, 57, 38,
> DDJ 38 +.35% 3E6, 2E6, 1E6, 913, 685, 533, 427, 291, 211, 113,

70,

> D_D 26 +.32% 2E6, 1E6, 951, 670, 498, 384, 306, 207, 149, 79, 49,
> D_B 28 +.17% 9E6, 3E6, 2E6, 1E6, 863, 625, 474, 299, 206, 102,

60,

> P_E 15 -.05% NEG, 6E7, 7E6, 2E6, 1E6, 742, 497, 267, 167, 72, 40,
> 4_J 24 -.11% NEG, NEG, 3E7, 7E6, 3E6, 3E6, 989, 498, 299, 123,

67,

> A&8 22 -.22% NEG, NEG, NEG, 3E7, 7E6, 3E6, 1E6, 642, 356, 133,

69,

> NSU 26 -.27% NEG, NEG, NEG, 3E8, 2E7, 5E6, 2E6, 924, 487, 171,

87,

> D_J 22 -.37% NEG, NEG, NEG, NEG, 3E9, 1E7, 4E6, 1E6, 569, 177,

85,

> 2_J 22 -.37% NEG, NEG, NEG, NEG, 2E9, 1E7, 4E6, 1E6, 569, 177,

85,

> 5_J 17 -.41% NEG, NEG, NEG, NEG, NEG, 2E7, 5E6, 1E6, 487, 143,

67,

> JoB 20 -.46% NEG, NEG, NEG, NEG, NEG, 1E8, 9E6, 2E6, 660, 180,

82,

> BDW 33 -.55% NEG, NEG, NEG, NEG, NEG, NEG, 1E8, 5E6, 2E6, 362,

156,

> B_P 21 -.83% NEG, NEG, NEG, NEG, NEG, NEG, NEG, NEG, 8E6, 472,

154,

>

Perhaps I do not know how to read this chart, but I think for the
benefit of a first time reader it needs to say that the above
numbers have to be multiplied by 1000 to get the number of hands.
Perhaps it did, if so I missed it or like I said reading it wrong.

Another suggestion would be to title and document it also.

All things considered, I have hard copied it and added it to my
library with notations.

···

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

[Harry_Porter]

brumar_lv wrote:

What do you think of my suggestion that comparing NO scores serve as
a way to choose which game to play? Is that valid, or does it put
too much emphasis on NO, instead of the usual emphasis on the
EV+CB?

I'd suggest that relative ROR's of the plays should be considered as
well. The problem with NO is that it ignores bankroll limitations.

- H.

[nightoftheiguana2000]

N0 is related to ROR:
N0=variance/(er-1+cb)^2 hands
Est. Kelly Bankroll=variance/(er-1+cb) bets

···

--- In vpFREE@yahoogroups.com, "Harry Porter" <harry.porter@...> wrote:

brumar_lv wrote:
> What do you think of my suggestion that comparing NO scores serve as
> a way to choose which game to play? Is that valid, or does it put
> too much emphasis on NO, instead of the usual emphasis on the
> EV+CB?

I'd suggest that relative ROR's of the plays should be considered as
well. The problem with NO is that it ignores bankroll limitations.

- H.

[Harry_Porter]

nightoftheiguana2000 wrote:

N0 is related to ROR:
N0=variance/(er-1+cb)^2 hands
Est. Kelly Bankroll=variance/(er-1+cb) bets

Understood. This is the distinction I'm drawing between the two:

NO suggests that one of two plays is always more favorable than the other.

As ROR changes under different bankroll asumptions, when given a
stated ROR tolerance, one play may be more attractive under one
bankroll but the second play is more attractive under a larger bankroll.

- H.

[nightoftheiguana2000]

You could rank games by ROR of your particular bankroll. The Kelly
optimal game will have an ROR of around 11%, if you play with double
the Kelly bankroll (conservative), your ROR is about 1%. ROR
approaches zero with Kelly strategy if you can go down in denomination.

For example, Loose Deuces has a higher ER than FPDW, but if your
bankroll equals the Kelly number for FPDW, you would be playing LD
suboptimally. If you happened to get lucky on FPDW and your bankroll
grew to the Kelly number for LD, then that game would become optimal,
conversely if you had the Kelly bankroll to play LD, but your bankroll
shrunk, at some point FPDW would become the optimal game. Some people
consider Kelly strategy too agressive, in which case just double up,
and your risk goes down an order of magnitude, or pick some inbetween
multiple.

nightoftheiguana2000 wrote:
> N0 is related to ROR:
> N0=variance/(er-1+cb)^2 hands
> Est. Kelly Bankroll=variance/(er-1+cb) bets

Understood. This is the distinction I'm drawing between the two:

NO suggests that one of two plays is always more favorable than the

other.

···

--- In vpFREE@yahoogroups.com, "Harry Porter" <harry.porter@...> wrote:

As ROR changes under different bankroll asumptions, when given a
stated ROR tolerance, one play may be more attractive under one
bankroll but the second play is more attractive under a larger bankroll.

- H.

[brucem]

--- In vpFREE@yahoogroups.com, "bornloser1537" <bornloser1537@...>
wrote:

Interesting ordering...

DW (.2) 280
AA (.2) 316
DW (.1) 348
JW (.2) 367
AA (.1) 397
DW (.0) 446
JW (.1) 472
AA (.0) 515
LD (.2) 515
LD (.1) 615
JW (.0) 629
LD (.0) 748
.....BL

Looking at statistics like those above makes me wonder why 9/6JB was
ever the standard "conservative" way to play VP.

For example, the 9/6JB cashback must exceed .4% to reach a break even
point and at CB=1%, NO = 660,000 games! That's a lot of games at a
high CB. By comparison, for FPDW and .0% cashback, NO = 446,000
games, about 2/3rds the NO for 9/6JB at 1%CB.

It makes me wonder why anyone would play 9/6JB. What am I missing?
Is it simply that games like FPDW can't be found about $1?