Thread Rating:

drjohnny
drjohnny
  • Threads: 32
  • Posts: 170
Joined: Sep 2, 2012
August 2nd, 2013 at 11:01:43 AM permalink
I just came across an easy to use Risk of Ruin Calculator.

Let's say there are 2 craps players who keep playing until they double a 100 unit bankroll or bust out trying.

Player A bets 1 unit on the pass line with no odds every time (1.41% house edge) - his risk of ruin is a whopping ~94%.

Player B bets 1 unit on the pass line with 10x odds every time (0.184% house edge) - his risk of ruin drops down to ~51%.
7craps
7craps
  • Threads: 18
  • Posts: 1977
Joined: Jan 23, 2010
August 2nd, 2013 at 11:47:58 AM permalink
Quote: drjohnny

I just came across an easy to use Risk of Ruin Calculator.

Nice. Same one I have in Excel.
looks like they use the Classic Gambler's Ruin formula.
Been around for 300 years or so, maybe longer

https://en.wikipedia.org/wiki/Gambler's_ruin

This is accurate when the wagers are the same.

trying to use an average wager or bets with high odds payoffs introduces an error I have seen up to 10%
Quote: drjohnny

Let's say there are 2 craps players who keep playing until they double a 100 unit bankroll or bust out trying.

OK
$5 wagers with $500 bankroll sounds good to me (Joe and Moe)
Quote: drjohnny

Player A bets 1 unit on the pass line with no odds every time (1.41% house edge) - his risk of ruin is a whopping ~94%.

Player B bets 1 unit on the pass line with 10x odds every time (0.184% house edge) - his risk of ruin drops down to ~51%.

This is not a fair comparison.
PlayerA (Joe) average bet is $5
PlayerB (Moe) average bet is $38.33 (agree?)
of course PlayerB has a better chance, way better, to double his stake...
bet more win more

Give PlayerA $40 wagers and the formula should show ruin at 58.75% (success at 41.25%)
he could also have a RoR of just 50.7% by betting all $500 on the next outcome. Bold play

PlayerB ruin is slightly higher than what the calculator shows but only by maybe 1 to 2% max
(looking at my thumb without doing a sim or math 1- (($500/$1000)-.0141) should be close)

The difference really becomes in how many trials (time) it takes (or can take)
until target or ruin is reached
that requires a different formula for the average or distribution of the number of trials
but I think can be found easily with search (Feller)
winsome johnny (not Win some johnny)
dicesitter
dicesitter
  • Threads: 32
  • Posts: 1157
Joined: Jan 17, 2013
August 2nd, 2013 at 11:57:58 AM permalink
dr



If you start with $100 and bet $5 pass line and $50 odds and your number is 5 and you dont hit it how much
do you have left for the next bet.


Dicesetter
drjohnny
drjohnny
  • Threads: 32
  • Posts: 170
Joined: Sep 2, 2012
August 2nd, 2013 at 12:04:00 PM permalink
Quote: 7craps

PlayerB (Moe) average bet is $38.33 (agree?)


How did you come up with that figure?

I would think Player B's average bet is $55.

Quote: dicesitter

If you start with $100 and bet $5 pass line and $50 odds and your number is 5 and you dont hit it how much
do you have left for the next bet.


Dicesetter


I would only make $50 odds bets with a $1K bankroll.
7craps
7craps
  • Threads: 18
  • Posts: 1977
Joined: Jan 23, 2010
August 2nd, 2013 at 12:51:01 PM permalink
Quote: drjohnny

How did you come up with that figure?

3*$5 (flat bet) + 2* $50 (odds bet)
$15+$100 all divided by 3
$115/3 or $38 and 1/3

or
1/3 of the time the $5 pass line bet is resolved for $5 on the come out roll
never gets the chance to make the odds bet.
2/3 now we have the full $55
(1/3)*$5 + (2/3)*$55 =
$5/3 + $110/3 = $115/3
Quote: drjohnny

I would think Player B's average bet is $55.

That is correct for the average bet on the don't pass
$Flat bet * (Odds factor + 1)
$5*(10+1) = $5*11 = $55
winsome johnny (not Win some johnny)
EvenBob
EvenBob
  • Threads: 441
  • Posts: 28575
Joined: Jul 18, 2010
August 2nd, 2013 at 1:01:16 PM permalink
So whats an acceptable ROR then.
"It's not called gambling if the math is on your side."
drjohnny
drjohnny
  • Threads: 32
  • Posts: 170
Joined: Sep 2, 2012
August 2nd, 2013 at 1:24:23 PM permalink
Quote: EvenBob

So whats an acceptable ROR then.


I like to keep my RoR per session under 25%.
dicesitter
dicesitter
  • Threads: 32
  • Posts: 1157
Joined: Jan 17, 2013
August 2nd, 2013 at 2:43:36 PM permalink
thats correct

YOu dont even have enough money to play two hands.

You cant expect to go to the table and win every hand you start
with.

We have all seen 3-4-5-6-7-8 guys toss a hand in a row without
a pass line winner.

The risk of ruin is so high if you over bet your bank roll.

Dicesetter
drjohnny
drjohnny
  • Threads: 32
  • Posts: 170
Joined: Sep 2, 2012
August 2nd, 2013 at 2:52:54 PM permalink
dicesitter, what risk of ruin does your team play with?

Have you guys kept track of your bust out rate?
blade78
blade78
  • Threads: 1
  • Posts: 2
Joined: Aug 2, 2013
August 2nd, 2013 at 10:17:13 PM permalink
HI,
I am BRAND new to this site and I just have a question
I don't understand the return rate?
Wizard says in Jacks or Better, the return rate is 99%+

what's the mean? Does it mean if I follow the guide I wil win 99% of the time?

Why dont I just go sit i in front of a Jacks or better machine then?

Thanks
AxelWolf
AxelWolf
  • Threads: 164
  • Posts: 22272
Joined: Oct 10, 2012
August 2nd, 2013 at 10:46:38 PM permalink
that means the game returns only 99%+ of your money back in the long run. for every $1 you bet you will get back 99 cents (with perfect play)
its really 99.54% return 54% of the time that you bet you end up with nothing no pay back at all. I think you are reading it as if 99 percent of the time you will win a hand or session.
♪♪Now you swear and kick and beg us That you're not a gamblin' man Then you find you're back in Vegas With a handle in your hand♪♪ Your black cards can make you money So you hide them when you're able In the land of casinos and money You must put them on the table♪♪ You go back Jack do it again roulette wheels turinin' 'round and 'round♪♪ You go back Jack do it again♪♪
dicesitter
dicesitter
  • Threads: 32
  • Posts: 1157
Joined: Jan 17, 2013
August 2nd, 2013 at 11:54:25 PM permalink
Well I bet according to the amount of money I have set aside to play craps with,
by that i mean i have an account that i play craps from, my winnings go back in and i play from
there.

I dont have a risk of ruin because i dont over bet my bank roll, and i understand my weakness's
and my strengths as a player.

dicesetter
Mission146
Mission146
  • Threads: 142
  • Posts: 16832
Joined: May 15, 2012
August 3rd, 2013 at 6:31:23 AM permalink
It means that, over the long run, for every dollar you bet you will lose less than $0.01.

It has nothing to do with your actual probability of winning in a given session, whatever your win goal, the higher it gets, the lower your probability of winning an individual session will be.
https://wizardofvegas.com/forum/off-topic/gripes/11182-pet-peeves/120/#post815219
dicesitter
dicesitter
  • Threads: 32
  • Posts: 1157
Joined: Jan 17, 2013
August 3rd, 2013 at 8:15:28 AM permalink
Well the long run is made up of alot of short runs.

MOst people including myself understand the math of the game, and it is math
in the looooooooooooooooooooooooooong run that indicates stuff.

However unless you have an endless amount of money, the short run can make
sure you never reach the long run.

The idea that if you bet a pass line bet with odds you will only lose $1.41 out
of every $100 you bet or if you place only the 6 & 8 you will lose only $1.52
for every $100 you bet sounds really good, hell this is not bad at all.

BUT......... what if you only have $200 and you bet $22 starting and the first 13 players
dont make a hand of 3 rolls and no one makes a pass line winner, what was the
per centage of your loss, was it a combined 1.46% ... no it was 100%

While the math of the game is important, the risk of ruin has to do with how much
you have to play with and how much of that you bet at one time, if you bet to
much relative to what you have, the math of the game will not save you.

In one of the craps books,i am not sure it may be Franks or Sharpshooters. there is a
line which says the only difference between a person that plays craps, and one that plays
craps alot is that the person that plays alot....... can loss...it all!!!!!

That is a very true statement.

Dont over bet your pocketbook.


Dicesitter
EvenBob
EvenBob
  • Threads: 441
  • Posts: 28575
Joined: Jul 18, 2010
August 3rd, 2013 at 1:04:58 PM permalink
Quote: Mission146



It has nothing to do with your actual probability of winning in a given session, whatever your win goal, the higher it gets, the lower your probability of winning an individual session will be.




Definition of 'Risk Of Ruin'

The probability of an individual losing.. gambling money.. to the point at which continuing on is no longer considered an option to recover losses. Risk of ruin is calculated by taking into account the probability of winning.. the probability of incurring losses, and the portion of an individual's capital base that is in play or at risk. Also known as the "probability of ruin".


http://www.investopedia.com/terms/r/risk-of-ruin.asp
"It's not called gambling if the math is on your side."
EvenBob
EvenBob
  • Threads: 441
  • Posts: 28575
Joined: Jul 18, 2010
August 3rd, 2013 at 1:16:17 PM permalink
Calculators like this are misleading. It says if you
have 50 units to risk and want to make a 1 unit
profit and have win rate of 50.5%, you have about
a 1% ROR. What it doesn't explain is the variance
involved. You could easily play for days in that
scenario before you were ahead by one unit.

If you have 10 units to risk to make 1 unit and a 65%
win rate, your ROR is also about 1%. But the variance
is very low, you could be one unit ahead in the first
15min most of the time. So all things considered, the
calculator doesn't mean much.
"It's not called gambling if the math is on your side."
7craps
7craps
  • Threads: 18
  • Posts: 1977
Joined: Jan 23, 2010
August 6th, 2013 at 6:15:54 PM permalink
Quote: EvenBob

Calculators like this are misleading.

The Ruin probabilities are accurate if you use them properly.
even money payouts and equal wagers.

Just not telling or showing the whole or complete picture that includes the time element.
Alan Shank brought this up in many of his posts too.
Just like all the house edge and ev only pushers that also leave out the variance.
Exactly

Quote: EvenBob

It says if you
have 50 units to risk and want to make a 1 unit
profit and have win rate of 50.5%, you have about
a 1% ROR.

sounds about right
My Excel

Quote: EvenBob

What it doesn't explain is the variance
involved. You could easily play for days in that
scenario before you were ahead by one unit.

The formula can be found here for the average number of trials
http://catlin.casinocitytimes.com/article/calculating-a-hold-percentage-1230

of course this is a 'wait time' distribution so 'average' can mean almost anything is not intuitive.
average is about 40.9 trials by the formula and about 88% chance it happens by the 40th trial
table showing part of the distribution (table created by a transition matrix Markov Chain)

always a chance one never hits the target or hits ruin.
hit targetruintrialin progress
0.505010.495
0.505020.495
0.631237030.368763
0.69435050.30565
0.7613970100.238603
0.8319460200.168054
0.8639580300.136042
0.8832110400.116789
0.8964080500.103592
0.9293061.53E-071000.070693847
0.9526150.0001007752000.047284225
0.9731770.003166695000.02365631
0.983090.008107710000.0088023

2.4% probability of still going at 500 trials
also can be done easily in Excel
N	N or less prob
1 0.505
2 0.505
3 0.631237375
4 0.631237375
5 0.694349751
6 0.694349751
7 0.733791041
8 0.733791041
9 0.761397183
10 0.761397183
11 0.782099719
12 0.782099719
13 0.798364371
14 0.798364371
15 0.811578079
16 0.811578079
17 0.822588402
18 0.822588402
19 0.83194624
20 0.83194624
21 0.840027201
22 0.840027201
23 0.847097335
24 0.847097335
25 0.853351058
26 0.853351058
27 0.858934182
28 0.858934182
29 0.863958491
30 0.863958491
31 0.868511315
32 0.868511315
33 0.872662005
34 0.872662005
35 0.876466423
36 0.876466423
37 0.879970142
38 0.879970142
39 0.883210758
40 0.883210758
41 0.8862196
42 0.8862196
43 0.889023014
44 0.889023014
45 0.891643335
46 0.891643335
47 0.894099639
48 0.894099639
49 0.896408335
50 0.896408335

Quote: EvenBob

If you have 10 units to risk to make 1 unit and a 65%
win rate, your ROR is also about 1%.

no. I think you misread the value.


Quote: EvenBob

But the variance
is very low, you could be one unit ahead in the first
15min most of the time. So all things considered, the
calculator doesn't mean much.

yep. average is about 3.3 trials by the formula
table showing part of the distribution
sure nice to see the whole picture that one might encounter.
110exact trialin progress1 intrial
0.6500.650.352.861
0.65000.352.862
0.79787500.1478750.2021254.953
0.797875000.2021254.954
0.86515800.0672830.1348427.425
0.865158000.1348427.426
0.90342500.0382670.09657510.357
0.903425000.09657510.358
0.92780200.0243770.07219813.859
0.9278022.75855E-0500.07217041513.8610
0.9444382.75855E-050.0166360.05553441518.0111
0.9444388.40667E-0500.05547793318.0312
0.9563348.40667E-050.0118960.04358193322.9513
0.9563340.00015973600.04350626422.9914
0.9651290.0001597360.0087950.03471126428.8115
0.9651290.00024418600.03462681428.8816
0.9717980.0002441860.0066690.02795781435.7717
0.9717980.00032960600.02787239435.8818
0.9769570.0003296060.0051590.02271339444.0319
0.9769570.00041105500.02263194544.1920

2.3% probability of still going at 20 trials
also done in Excel
(cumulative distribution for # of games until target or ruin)
N	N or less prob
1 0.65
2 0.65
3 0.797875
4 0.797875
5 0.865158125
6 0.865158125
7 0.903425402
8 0.903425402
9 0.927801658
10 0.927829243
11 0.944466038
12 0.944522519
13 0.956417827
14 0.956493496
15 0.96528859
16 0.96537304
17 0.972042652
18 0.972128073
19 0.977287018
20 0.977368466
21 0.981422687
22 0.981497459
23 0.984724726
24 0.984791644
25 0.987387617
26 0.987446459
27 0.989552396
28 0.989603493
29 0.991323607
30 0.991367578

but for many, they are very very happy with one value from any calculator
that tells them everything THEY wants to know.

you and I are in the majority here.
thanks, now my new computer is all set up the way I like it
winsome johnny (not Win some johnny)
katastrof78
katastrof78
  • Threads: 0
  • Posts: 1
Joined: Jan 25, 2014
January 26th, 2014 at 6:59:15 AM permalink
Dear; 7craps



You excel spreadsheet has been very revealing.

could you please Will you send me your excel spreadsheet ?

thank you for now...
  • Jump to: