Risk of Ruin help

[Sonny]

Can some authority figure please settle this?

I thought we had. No, a lifetime RoR is not necessarily 100%. The only way to have a 100% RoR is to play without an advantage or to significantly overbet your advantage. This is covered in Schlesinger’s book.

-Sonny-

 

[moo321]

Can some authority figure please settle this? Is moo right? Do all non-resizing betting schemes have 100% RoR because the possibility of any sized negative flux is always there? Can Ken or Fred settle this? Anyone?

You'd have to have someone with a lot more math knowledge than I have to settle it. But I can tell you that for an infinite number of hands, your risk of ruin is 100%. If you counted cards in hell, and never re-sized your bets, eventually you'd lose a million hands in a row.

It's also an asinine debate; even if I'm wrong, or I'm right, it doesn't really matter. The "lifetime" risk of ruin numbers that are being offered are rigid, and not really useful. Risk of ruin vs. doubling, and frequency distributions are much more useful. Do you care if the stock you're investing in has a 5% or 3% or 100% risk of ruin over the next 2000 years? No, you care about the next few quarters, because that's the only thing you can meaningfully predict.

 

[Kasi]

But I can tell you that for an infinite number of hands, your risk of ruin is 100%.

But you could never actually say that, let's even assume a -EV game, could you? All you would know for sure is somebody lost all their money in a billion or trillion hands. And that every time so far the first billion people eventually lost all their money in the first trillion hands. But you would never know for sure the next person couldn't play "forever" until he actually went broke. You'd assume it's real, real close to 100% but you'd never know for sure. Likewise, in a lifetime ROR we only know it's real real close to that ROR %age.

No big deal. If you want to know your ROR before doubling roll before either going bust or doubling it, just because you think it's more important, and I'm not saying it's not, that's fine but even then what does it mean to you if I tell you your ROR is 40%. How do we know what will happen after a billion hands when one has neither doubled or lost his roll? If he plays potentially "forever", does he have 100% of doubling or 100% of losing it all?

Would it satisfy you if it was defined that exactly a billion people each played a billion hands at most and that as it actually happened 13% of those billion people either had gone bust during the billion hands while the other 87% still had money that 13% ROR is a pretty good number for a lifetime ROR?

If 40% of those billion people had either doubled their roll at some point and then quit playing but the other 60% lost their roll in the first billion hands before doing so?

Just BSing lol.

But from now on, since you're so specific, just ask specific stuff like what are my chances of losing my roll in the next billion hands. Or, if I play at most a billion hands, what are the chances I will have either quit after doubling my roll or lost it all.

Or what are the chances of me doubling my roll in the next 1000 hands or losing it all or neither?

Just please stop saying your lifetime ROR is always 100%, in a +EV game no less, no matter what.

It makes you seem like you're a member of the Flat-Earth Society or something lol.

 

[godeem23]

But you could never actually say that, let's even assume a -EV game, could you? All you would know for sure is somebody lost all their money in a billion or trillion hands. And that every time so far the first billion people eventually lost all their money in the first trillion hands. But you would never know for sure the next person couldn't play "forever" until he actually went broke. You'd assume it's real, real close to 100% but you'd never know for sure. Likewise, in a lifetime ROR we only know it's real real close to that ROR %age.

No big deal. If you want to know your ROR before doubling roll before either going bust or doubling it, just because you think it's more important, and I'm not saying it's not, that's fine but even then what does it mean to you if I tell you your ROR is 40%. How do we know what will happen after a billion hands when one has neither doubled or lost his roll? If he plays potentially "forever", does he have 100% of doubling or 100% of losing it all?

Would it satisfy you if it was defined that exactly a billion people each played a billion hands at most and that as it actually happened 13% of those billion people either had gone bust during the billion hands while the other 87% still had money that 13% ROR is a pretty good number for a lifetime ROR?

If 40% of those billion people had either doubled their roll at some point and then quit playing but the other 60% lost their roll in the first billion hands before doing so?

Just BSing lol.

But from now on, since you're so specific, just ask specific stuff like what are my chances of losing my roll in the next billion hands. Or, if I play at most a billion hands, what are the chances I will have either quit after doubling my roll or lost it all.

Or what are the chances of me doubling my roll in the next 1000 hands or losing it all or neither?

Just please stop saying your lifetime ROR is always 100%, in a +EV game no less, no matter what.

It makes you seem like you're a member of the Flat-Earth Society or something lol.

If you played "forever" in a negative EV game you would have to go broke, or else it wouldn't be negative EV. You would have to go broke at some point.

 

[EasyRhino]

You'd have to have someone with a lot more math knowledge than I have to settle it. But I can tell you that for an infinite number of hands, your risk of ruin is 100%. If you counted cards in hell, and never re-sized your bets, eventually you'd lose a million hands in a row.

No no.

Let's take an extreme example. You're playing a game with a small edge (1%), but the variance is zero. That is, for every unit you wager, you will receive only and exactly 1.01 units in return. Your risk of ruin is 0%. Even if an uber-infinite number of wagers are made.

Now, let's take a game with a small advantage, but with normal variance, and an uber-infinite number of hands are played. And let's say that out of a thousand people, most are betting fixed bets, but very conservative compared to their starting point. Risk of Ruin can never be 0%, because a few of the poor saps really are going to lose a million hands in a row and bust out. But a very large majority of these guys are not. Or more accurately, they may lose a million in a row, but they will have won two million before. The variance truly can't catch up with them. As Don Schlesigner put it, they've got to get you early, or they aren't going to get you at all.

The reason that some people can reach escape velocity and never bust the bankroll is the same reason that players in a negative expecation game are required to have a 100% ROR, eventually the advantage or disadvantage dominates the variance.

 

[moo321]

No no.

Let's take an extreme example. You're playing a game with a small edge (1%), but the variance is zero. That is, for every unit you wager, you will receive only and exactly 1.01 units in return. Your risk of ruin is 0%. Even if an uber-infinite number of wagers are made.

Now, let's take a game with a small advantage, but with normal variance, and an uber-infinite number of hands are played. And let's say that out of a thousand people, most are betting fixed bets, but very conservative compared to their starting point. Risk of Ruin can never be 0%, because a few of the poor saps really are going to lose a million hands in a row and bust out. But a very large majority of these guys are not. Or more accurately, they may lose a million in a row, but they will have won two million before. The variance truly can't catch up with them. As Don Schlesigner put it, they've got to get you early, or they aren't going to get you at all.

The reason that some people can reach escape velocity and never bust the bankroll is the same reason that players in a negative expecation game are required to have a 100% ROR, eventually the advantage or disadvantage dominates the variance.

We're arguing about this semantically, and while it's useful, it's not a mathematical proof. I know that the proof involves things that I can't calculate, so unless anyone on here has a math degree, I think we're not going to have a proof.

 

[godeem23]

We're arguing about this semantically, and while it's useful, it's not a mathematical proof. I know that the proof involves things that I can't calculate, so unless anyone on here has a math degree, I think we're not going to have a proof.

Do you have a link to a proof or a source who is saying the same things you are about this?

 

[Kasi]

If you played "forever" in a negative EV game you would have to go broke, or else it wouldn't be negative EV. You would have to go broke at some point.

Actually all we would (likely) know is that after a trillion people went broke within a trillion hands is that 100% of those trillion people went broke in the first trillion hands. Does that mean the trillionth + 1 person will eventually go broke with 100% certainty?

Do we know for sure that the turtle one inch away from the finish line and getting half-way closer with each step will ever reach the finish line? All we know is he'll be really really close to the finish line after a trillion steps.

And, if one believes, for some silly reason, that one will always go broke with 100% certainty at some point in a -EV game, why wouldn't one believe that one will always win with 100% certainty in a +EV game?

It's just a theoretical number when you get down to it lol. Not exact, but really really close the closer you get to "infinity" lol.

 

[zengrifter]

Deep. No one gets out alive. zg

 

[Kasi]

I know that the proof involves things that I can't calculate, so unless anyone on here has a math degree, I think we're not going to have a proof.

At least you seem to acknowledge that a proof exists while acknowledging you wouldn't know it if it bit you in the ass.

If all you need is a math degree, eventually you will discover that it's people with math degrees that figured this stuff out.

I wonder what life was like for Copernicus having proved everything does not actually revolve around the earth.

This is simple stuff - there's only so many ways 416 cards can be arranged.

 

[QFIT]

You'd have to have someone with a lot more math knowledge than I have to settle it. But I can tell you that for an infinite number of hands, your risk of ruin is 100%. If you counted cards in hell, and never re-sized your bets, eventually you'd lose a million hands in a row.

This is only true with a negative EV game. Yes you might lose a million hands in a row. But, in a positive EV game, you are more likely to have won enough hands in a row prior to losing a million hands in a row to withstand the loss and remain in the game. OTOH, in a negative EV game, your risk is 100% -- even with a Martingale and infinite bankroll.

 

[EasyRhino]

Uh oh, an infinite number of bets with an infinite bankroll.

My mind just exploded.

 

[sagefr0g]

This is only true with a negative EV game. Yes you might lose a million hands in a row. But, in a positive EV game, you are more likely to have won enough hands in a row prior to losing a million hands in a row to withstand the loss and remain in the game. OTOH, in a negative EV game, your risk is 100% -- even with a Martingale and infinite bankroll.

well i'm confused as usual but just keep slogging along anyway lol.
so my thought is for a game where a counter can realize a positive EV and then your talkin him playing lots and lots of blackjack where the question of interest becomes lifetime ROR well being the case where so many hands are played that this player is gonna reach his N0 if he doesn't bust his bank. and by then the player should be ahead by one standard deviation or at least still have enough bank left to play lol. but the N0 is inside of this lifetime play to where Moo i think can relate to that. and well probably should be able to figure the likelyhood of reaching some goal by the N0. and say you do get ahead by one standard deviation by the N0 then wouldn't one have more confidence of being within the projected lifetime ROR and even more so after another N0 and on and on. to where you'd be able to say about you lifetime ROR that well they didn't get me early so they just might not get me more than my lifetime ROR projects?

 

[godeem23]

Actually all we would (likely) know is that after a trillion people went broke within a trillion hands is that 100% of those trillion people went broke in the first trillion hands. Does that mean the trillionth + 1 person will eventually go broke with 100% certainty?

We're not talking about a trillion hands. We're talking about infinitely many hands, eternity. We're also not running simulations to see what a trillion people will do. We're using mathematics. Mathematically, in a negative EV game, as the number of hands approaches infinity, bankroll approaches zero (starting with positive bankroll obviously). This is fact. This is the very meaning a negative EV game. Streaks of luck exist. Anyone can win a hundred or a million hands in a row if given enough time, but we're talking about infinite time. No streak of good or bad luck lasts indefinitely.

And, if one believes, for some silly reason, that one will always go broke with 100% certainty at some point in a -EV game, why wouldn't one believe that one will always win with 100% certainty in a +EV game?

Because there's no upper bound on how much money one can have but there is a lower bound.

 

[rukus]

jesus man, this was settled 2 weeks ago by sonny's posts quoting the formulas put into print by Don S and others using SOUND mathematical formulas, backed up by blackjack avenger's posts, and now you just received a response from norm w, who wrote the CV products.

if you want further proof verifying the session/trip or LIFETIME/INFINITE RoR forumlas, for the love of g-d read blackjack attack if you havent (or re-read it if you have) - chapter 8 - don compares the formula's predictions to actual simulations of short sessions, medium sized sessions, and essentially infinite sessions.

i dont have a math degree per se, but graduated with an engineering degree from MIT (im not joking), and lets just say i learned a bit about math. the arguments presented to you are correct. im no authority on this but i promise you the people who developed the formulas and their applications are. Sonny and BJ avenger have described how to apply them properly.

leave it or take, i guess.