Thread Rating:

I know the EV doesn't change with odds bets and more odds increases bankroll variance, but how do the odds bets change the risk of ruin probability?

One way to determine what that figure is, is to run a simulation. You could also work out a markov chain.

Edit: Misinterpreted OP’s question.

Edit 2: It might increase ROR or might decrease. Depends on the variables.

my experience I like to use a transition matrix and run a simulation, taking odds increases the chance of hitting a win target/target goal/stop-win limit for both the pass and don't pass.Quote:crapstossawayHow do you calculate a risk of ruin for a stop-win limit of a given bankroll when odds are taken? How does the risk of ruin change with no odds, 1x odds, 2x odds etc?

that has to lower the chance of ruin compared to no odds, most times, depending on the wagers and targets, of course (comparing about the same average bet)

and taking odds lowers the number of trials to target or ruin.

I have lots of data on this.

if u give me an example bankroll, target, max number of bets if required, I can show some data.

Sally

added: you can see some data in my blog here

https://wizardofvegas.com/member/mustangsally/blog/#post1094

without an equal average bet, of course risk of ruin increases.

but a $5 pass and no odds vs. $5 pass with $50 odds over an hour of play

is not even close to being a fair comparison, if that matters, and it should.

I like to play about 30 minutes to an hour once a week with a $200 buy in. $5 minimum pass line and 2 come bets with odds working on come out rolls. I'm not looking to make significant money, but I enjoy playing and hope to make maybe $25 and see it as a win even if I just break even.I don't see it as a grind and the thrill for me is to try and get lucky. I do however wish to structure my odds bets, stop limits and time per session in such a way that I reduce to a minimum my risk of ruin on the bankroll, willing to lose it all in a 1 hour session.

I haven't been able to deduce the optimal amount of odds (up to 20x) to minimize my risk of ruin or structure a betting system with fluctuating odds as my bankroll changes in relation to my $25 goal. I'm open to suggestions and changes to my stop win limit and odds to minimize risk of ruin on $200 bankroll.

five dollar pass line is difficult to find.

10x odds is easy to find but emotionally and financially difficult to sustain, particularly for a full hour.

what is difference at ten dollar pass line bet versus ten dollar pass bet with 2x odds?

I just don't see a point where I should stop playing as long as the chips in the rack are increasing and I am in control of my play. For example, If I start off with $400, I am out when I am felted. If I get above $400 by a couple hundred dollars, then it is time to leave even or better. If things keep going well, the number to leave at goes up as the chips are accumulated.

I am playing a negative expectation game; when it runs positive for me for those few minutes or hours...I am trying to make a few dollars...

Quote:RonCI get the stop-loss points, but why have a stop-win point?

Because if you don't stop when you are winning you won't stop when you are losing, you will play catch-up all the way down to your starting bankroll and you will keep going after that into negative territory.

Quote:FleaStiffBecause if you don't stop when you are winning you won't stop when you are losing, you will play catch-up all the way down to your starting bankroll and you will keep going after that into negative territory.

I have an automatic stop-loss when I am losing...the felt. I also have a daily budget that I do not exceed. Once I am up >$100 in a "session", my stop-loss moves up accordingly. Since I don't play back down to the felt and the stop-loss moves up as I win, I don't go back into negative territory. I guess as a casual gambler I see it differently than others...

fixed itQuote:FleaStiffSally, I had difficulty with the link.

the average bet is different for one over an hour (30 bets) of play.Quote:FleaStiffwhat is difference at ten dollar pass line bet versus ten dollar pass bet with 2x odds?

the OP Q seems to me to be about hitting a win target with a starting bankroll

and playing just for a small amount of time(bets)

for Ur direct question

say a stake of $200, flat betting $10, trying for $400 or bust

from here: section 2r. calculating in units

stake:20, bet:1,target:40,gain:20

https://sites.google.com/view/krapstuff/risk-of-ruin

> gambler.ruin(20,20, 244/495, "Craps pass line 0x odds")

[1] "Craps pass line 0x odds. Stake:20, Target:40"

[1] "p(goal):0.36223094, p(ruin):0.63776906"

[1] "mean Trials:389.69, mean given Goal:389.69, mean given Ruin:389.69"

using R (not public)

bank p(target) p(ruin) avg trials

20.0 0.45954944 0.54045056 51.893986

for that example

taking odds increases the chance of hitting $400

and decreases chance of busting out the $200 stake

but, that should have been expected

Sally

Quote:mustangsally

the OP Q seems to me to be about hitting a win target with a starting bankroll

and playing just for a small amount of time(bets)

Sally

Bingo. I was looking at your blog post charts, but I'm not sure exactly how to interpret the percentages as it relates to given stop win limit. Any data on how different odds and increases average bets contribute to risk of ruin with a given stop win limit?

Or you just use a simple formula :Quote:mustangsallyfor Ur direct question

say a stake of $200, flat betting $10, trying for $400 or bust

from here: section 2r. calculating in units

stake:20, bet:1,target:40,gain:20

https://sites.google.com/view/krapstuff/risk-of-ruin

> gambler.ruin(20,20, 244/495, "Craps pass line 0x odds")

[1] "Craps pass line 0x odds. Stake:20, Target:40"

[1] "p(goal):0.36223094, p(ruin):0.63776906"

[1] "mean Trials:389.69, mean given Goal:389.69, mean given Ruin:389.69"

x^b / (x^b + y^b)

x = 244/495

y = 1 - x

b = 20 units

Which gives 36.2230940439679 % chance of doubling a 20 unit stake before busting.

Quote:FleaStiffThe odds bet is subject to the house edge as well as the contract bet.

Really? Are you John Patrick??

To reiterate and rephrase, I'm hoping to discover whether it is in my best interest to bet with or without x number of odds on a $5 pass line 3 point molly with working odds on come out rolls and 40x min bet bankroll and stop win limit of around 1.125% and approx 1 hour duration.

What amout of odds lends itself to the lowest risk of ruin?

So I would hazard to guess that the answer to your question is not a constant, but maybe some kind of curve or function of your distance from winGoal or stopLoss (or zero if you're just going to deplete your stash).

For instance, if you are not far from your winGoal, bet enough odds just to reach your winGoal if you win. That could be the best strategy in an -EV game. If you lose, you can try doing that calculation again. But if you lose too many in a row, the next loss will send you to zero. These are just some thoughts - a simulation would be useful here.

Quote:kfmfe04I suspect that odds will be helpful when you are close to your winGoal and detrimental when you are close to blowing out

I question the merits to this. I could be wrong, but the way I see it is there are 2 sides in balance - amount of bankroll subject to HA (more when less odds are taken, less when more odds are taken) vs "variance of the mean". in other words, fewer attempts at a negative expected outcome vs giving yourself more chances to make your way back toward the expected outcome toward the middle and away from the swings (beneficial for when you are down in bankroll).

Yes, simulations can be helpful, but only in proving a theory. I'd be interested in knowing how changes in the variables -most importantly odds, but also stop-win limits, iterations at the game i.e. time and bankroll and what impact these changes have on risk of ruin. A formula which includes odds a variable would be the solution.

Quote:unJonI’d suggest a simplification. Ask the question as pretending no odds, what should my bet sizes be in order to optimize the goal. Then for each bet size, you can only do better by minimizing amount on PL and maximizing odds based on table min and odds allowed.

What is the bet size to optimize the goal?

Quote:mustangsally

without an equal average bet, of course risk of ruin increases.

but a $5 pass and no odds vs. $5 pass with $50 odds over an hour of play

is not even close to being a fair comparison, if that matters, and it should.

MustangSally, why does the risk of ruin increase with the $50 odds in the above example, if a stop-limit is given?

The Odds bet, more properly known as the free odds bet, is a bet with no player edge or house edge. That is why they let you take it down anytime you want.

However, to the extent that you choose to put down "x" amount as an odds bet, it is indeed part of YOUR bankroll and is subject to being lost based upon an event taking place that is indeed subject to a very modest house edge.

If you are up ten grand, a fifty dollar odds bet probably has some mathematically quantifiable effect on risk of ruin but to a practical player it has no effect at all on risk of ruin.

If you are just starting your roll the funds you commit to an odds bet can be lost and would be both emotionally and financially effective to diminish your ability to continue making Line Bets.

Given the above, and the crux of my post, what impact do the odds play on risk of ruin? In other words, since odds bets are paid with true odds (by definition) what causes their play to impact your chances of losing it all or hitting a stop-win limit?

Quote:FleaStiffLet me rephrase things:

The Odds bet, more properly known as the free odds bet, is a bet with no player edge or house edge. That is why they let you take it down anytime you want.

However, to the extent that you choose to put down "x" amount as an odds bet, it is indeed part of YOUR bankroll and is subject to being lost based upon an event taking place that is indeed subject to a very modest house edge.

If you are up ten grand, a fifty dollar odds bet probably has some mathematically quantifiable effect on risk of ruin but to a practical player it has no effect at all on risk of ruin.

If you are just starting your roll the funds you commit to an odds bet can be lost and would be both emotionally and financially effective to diminish your ability to continue making Line Bets.

Imagine a VP machine with two paytables. One has a 98% RTP and the second has 100%RTP. Here’s the catch, you have to play one hand on 98% paytables but you can then add up to five times the hand amount to play that same hand at 100% paytable.

Should you:

(A) Sit there only paying 1x and getting 98% RTP or

(B) Play max bet so you have 5x money getting 100% RTP.

If you answer (A), then I suggest you are playing too high of a VP denomination for your bankroll and you should go to a cheaper machine where you can do (B).

It’s the same as craps. If your bankroll can’t afford $10 plus $50 odds on the PL, that’s fine. Play $10 with no odds. But if you could find a $2 table, you should immediately quit playing at the $10 table and go play $2 PL with 4x or 5x odds.

And I understand the casinos don’t offer that. It doesn’t change the fact that odds are good. Just sometimes the good makes your variance higher and your bankroll is too small to handle the swings.

My post remains unsolved and I'll explain. The more of the bankroll that's put into odds means fewer iterations or chances to hit the win goal. The more iterations, the more likely you are to reduce the variance or swings away from the mean. You can buy more chances to hit your win goal before the possibility of rapidly depleting your bankroll with large downswings on odds.

However, the more iterations you see, the greater your chances are of losing because your subjecting yourself to a greater amount to a house edge. I'd like to better understand the relationship between this concept of many iterations on a house edge to reduce swings vs larger average bets with odds as it relates to risk of ruin.

If someone else doesn't, I probably figure out how to get wincraps to hyper-drive multiple sessions and compare betting strategies of odds vs no odds with a 3 point molly.

As more time passes and I think about it more, my inkling is ultimately max odds wins heads up against multiple iterations because the house edge comes knocking down the door and slowly but surely progressively lowers your chances of hitting your win goal irregardless of the pros of having more chance of returning to center if down. This would act in the same as best chances of hitting your goal is to bet it all at once instead of multiple rolls. I'm not sure if this fallacy (if its a fallacy) has a name, but it would be useful to reference in the future.

Bolded part is exactly right. It’s always true in a negative expectation game. And no sorry, I’m not the guy to work out the formula for you on your specific question.Quote:crapstossawayDeMango, honestly, I think I could figure out how to answer this definitively in Wincraps simulations with enough time and dedication. The work involved is significant for me due to my novice experience with it yet, but that's just my personal barrier to entry and prioritization.

If someone else doesn't, I probably figure out how to get wincraps to hyper-drive multiple sessions and compare betting strategies of odds vs no odds with a 3 point molly.

As more time passes and I think about it more, my inkling is ultimately max odds wins heads up against multiple iterations because the house edge comes knocking down the door and slowly but surely progressively lowers your chances of hitting your win goal irregardless of the pros of having more chance of returning to center if down. This would act in the same as best chances of hitting your goal is to bet it all at once instead of multiple rolls. I'm not sure if this fallacy (if its a fallacy) has a name, but it would be useful to reference in the future.

Casino games tend to have much more complicated, discrete rules than a simple barrier option, so I doubt you will find a simple formula. Casino game simulations tell you what to expect, if you take a given set of conditions to be true (fair dice, good shuffles/no special clumping, a balanced roulette wheel, no machinations, no AP play like dice control, etc...). Of course, to the extent that real conditions deviate from those ideal conditions, the results of any calculations or simulations will NOT apply.

How returns change as a function of odds, stop/win limits, number of iteration, etc... and their impact on risk of ruin can all be found via simulations. I've written a baccarat simulator that does spew out these kinds of statistics, but no craps simulator yet (that's in the works). In fact, on another thread, in another forum, I just recently analyzed the impact on risk of ruin using a negative progression system like Labouchere vs a flat betting system, on top of a complicated set of pattern matching rules (something that most baccarat players seem to be obsessed about).

My comments previously about odds are just an opinion based on the fact that odds have zero EV. So their immediate impact is mostly on leverage and volatility, which will affect the chance of you hitting a barrier. Of course, I could be wrong. It would take a simulation to suss that out.

Quote:DeMangoThe obvious answer to all this is a computer simulation. Wincraps comes to mind. Too much work?

Agreed. You can probably try running it for 1x odds, then 2x odds ... 100x odds. But I don't have Wincraps so I don't know how flexible its simulation engine is.

I'm more curious about variable odds.

What would happen if I increase odds when I get close to winGoal (just enough odds to hit the winGoal on ONE WIN), but decrease it when I get close to busting out (to give me more chances for the random walk to bring me back to life).

I think I will try this out on my Baccarat simulator.

The casinos are ripe for harvest, but the skilled laborers are few.Quote:DeMangoThe casinos are ripe for harvest, but the laborers are few.

Quote:crapstossawayDeMango, honestly, I think I could figure out how to answer this definitively in Wincraps simulations with enough time and dedication. The work involved is significant for me due to my novice experience with it yet, but that's just my personal barrier to entry and prioritization.

If someone else doesn't, I probably figure out how to get wincraps to hyper-drive multiple sessions and compare betting strategies of odds vs no odds with a 3 point molly.

As more time passes and I think about it more, my inkling is ultimately max odds wins heads up against multiple iterations because the house edge comes knocking down the door and slowly but surely progressively lowers your chances of hitting your win goal irregardless of the pros of having more chance of returning to center if down. This would act in the same as best chances of hitting your goal is to bet it all at once instead of multiple rolls. I'm not sure if this fallacy (if its a fallacy) has a name, but it would be useful to reference in the future.

Still sounds like a curve, kind of like how a full Kelly betting is constantly re-calculated for optimal sizing. So I don't think there's one answer, though I too thought unJon was correct in saying you might be overbetting your bankroll at those odds.

Maybe you should start with knowns.

You want to play for 1/2 hour. If a typical table, that's 15 decisions or so.

You want to br at $200. So,

$5 table 1x odds, 2 come bets, 1x odds. $30. (Maybe $31 or 32 depending on the points). That funds less than 7 losing decisions, should you go on a bad run. Maybe some give-back if they PSO and your 1st or 2nd come gets paid. But not even half what you need to bet that way if it goes south, to play for 30 minutes.

And there's a constant small hit of the HE on each base bet. You can improve that percentage by betting higher odds, but it can go south much more quickly - you could be felted in 5-10 minutes.

I think you're under-funded for your play. But short-term, it could make your modest win goal, too. So why not play how you like.

Disclaimer. I am not an expert. Just thinking out loud.