Good point Brock. As an even more extreme example, suppose you are betting one dollar (payoff is even money) on the flip of a coin, but this is a special coin that is in your favor by 1 percent. Your starting bankroll is one dollar. You're playing a winning game, after all the coin is in your favor, but you are not properly funded. Your RoR is well over 50%.
Considering Coloring up for the last time
- Thread starter Ferretnparrot
- Start date
if the odds were win 51% and lose49% and you bet your whole bank one one bet you would have a 49% to fail and 51% to double, its totally not unlikely that you would fail, but it would still be a safe bet to take.
i think the problem with the typical ror to double numbers may have to do with the fact that the deviation upward is combined with the small advantage you have and would put some winnings over doubling and arnt included in the total tablulation of al the trials. I understand the regularity of upswings and down swings, i just think that the neumber of upswings will be the same as the number of downswings, and even more slightly if your playing at an advantage. so even if you bet your whole bank hypothetically on one hand at an advantage, your ror would never go below 50% since you have an equal chance of catching an upswing or downswing, regardless of how severe it is.
The only way i see a ror in blackjack goign above 50% is if you only have enough money to play so few hands that the likely hood of being dealt a blackjack or hand that you can doubl down on is less that 50%. As those are what tips you ever the 100% return level. in which case the dealer winning more hands than you would wipe you out in the short run.
i think the problem with the typical ror to double numbers may have to do with the fact that the deviation upward is combined with the small advantage you have and would put some winnings over doubling and arnt included in the total tablulation of al the trials. I understand the regularity of upswings and down swings, i just think that the neumber of upswings will be the same as the number of downswings, and even more slightly if your playing at an advantage. so even if you bet your whole bank hypothetically on one hand at an advantage, your ror would never go below 50% since you have an equal chance of catching an upswing or downswing, regardless of how severe it is.
The only way i see a ror in blackjack goign above 50% is if you only have enough money to play so few hands that the likely hood of being dealt a blackjack or hand that you can doubl down on is less that 50%. As those are what tips you ever the 100% return level. in which case the dealer winning more hands than you would wipe you out in the short run.
Maybe This Will Work!
You can easily have a 100% ROR. Even give your example. If you bet all your money on one hand and repeat you only get to play a couple hands before you are broke!!!!!!!!!
Chances of doubling your bank before losing half is 66%. IIIIIIIf betting properly!!!!!!!!!!!
I agree 100% with what Brock Windsor stated above about betting 10% of your bank in a 1% game and going broke!!!!!!
Try this:
Take 1 deck of cards and take out a 2,3,4,5,6,7. A huge advantage!
Get a piece of paper, pen and calculator and start with a $10,000 bankroll
Bet 20% on every hand
You are going to recalculate your bank after every hand.
You are going to play each hand and then shuffle while keeping out the low cards.
You can do this during commercials of TV.
Your bank will drop so quickly that it will become apparent that it will be very difficult to catch up and eventually you will go broke!
You can easily have a 100% ROR. Even give your example. If you bet all your money on one hand and repeat you only get to play a couple hands before you are broke!!!!!!!!!
Chances of doubling your bank before losing half is 66%. IIIIIIIf betting properly!!!!!!!!!!!
I agree 100% with what Brock Windsor stated above about betting 10% of your bank in a 1% game and going broke!!!!!!
Try this:
Take 1 deck of cards and take out a 2,3,4,5,6,7. A huge advantage!
Get a piece of paper, pen and calculator and start with a $10,000 bankroll
Bet 20% on every hand
You are going to recalculate your bank after every hand.
You are going to play each hand and then shuffle while keeping out the low cards.
You can do this during commercials of TV.
Your bank will drop so quickly that it will become apparent that it will be very difficult to catch up and eventually you will go broke!
I think this is kinda an invers martingale for advantage play, and would amplifie your chances of going broke most of the time, just as a regular martingale manipulates the likelyhood of making a small amount of money most of the time, but loses a lot a small percentage of the time. I think that while it may be true that by reducing your bet after a loss it increases the rate at which you tap out of money, if you followed this example and increased your bet after everytime you won, on rare occasions you would make a lot of money and the net product of all of it would still be in your favor ie 90% of the time you lose 10g, 10% of the time you make 100g but when viewed from the area of loss to double it would apear that you lose 90% of the time and only double 10%
You seem to all be focusing only on the negetive fluxuations
do you really have to add !!!!!!!!!!!!!!!! to all of your posts?
I will say that in looking at the example you suggested where you place all of your monmey on a bet everytime you will without a doubt eventually lose all your money, it woudl probibly be safe to say that from and ideal betting plan with minimal risk that your ror would aproach and eventually reach this as your betting plan got weaker. but i still cant help but wonder if the true math of this example would be undifined, as with infinate trials one bank would goto infinity and ofset evrything.
When i calculate my ror when im curious i woudl typically ignore my small bets and look only at my max bet, i would figure out how the range of deviation i was in deduct that from 100 and divide it by. so say my bank represented a number of deiviations that occurs 75% of the time , 25% of the time the deviation woudl eb greater than my bank but only 12.5% of the time would it be in the direction of destroying my bank. The formula woudl be
[100-(probibility of deviation totalling bank in percent)]/2
so a number less than 100 divided by 2 would always be less than 50 If your playing with an advantage it would be less than that.
You seem to all be focusing only on the negetive fluxuations
do you really have to add !!!!!!!!!!!!!!!! to all of your posts?
I will say that in looking at the example you suggested where you place all of your monmey on a bet everytime you will without a doubt eventually lose all your money, it woudl probibly be safe to say that from and ideal betting plan with minimal risk that your ror would aproach and eventually reach this as your betting plan got weaker. but i still cant help but wonder if the true math of this example would be undifined, as with infinate trials one bank would goto infinity and ofset evrything.
When i calculate my ror when im curious i woudl typically ignore my small bets and look only at my max bet, i would figure out how the range of deviation i was in deduct that from 100 and divide it by. so say my bank represented a number of deiviations that occurs 75% of the time , 25% of the time the deviation woudl eb greater than my bank but only 12.5% of the time would it be in the direction of destroying my bank. The formula woudl be
[100-(probibility of deviation totalling bank in percent)]/2
so a number less than 100 divided by 2 would always be less than 50 If your playing with an advantage it would be less than that.
Last edited:
Yes that's true. However I meant to imply in my post that you should KEEP PLAYING, just like you keep playing blackjack, until you have a huge fortune or go broke. I guess I should have stated more explicitly that you continue to bet if you win the first bet. I could be wrong here, but I don't think Ferret really understands what RoR means if he thinks it has to be less than 50% if you're playing a winning game. RoR is the likelyhood of going broke at SOME POINT. So in the case I described, with the coin 1% in your favor, betting 1 dollar every hand with a 1 dollar bankroll, your RoR is indeed well over 50% (I think it's actually over 90%).
In the case of RoR the offset doesn’t matter because you are probably broke already. If a strategy has a 75% RoR then 25% of the people who use it will make lots of money, which will offset the results of the losers (assuming they all have the same advantage). However, that won’t make you feel any better when you go broke. Knowing that somebody somewhere is possibly getting rich won’t ease the pain. As you said, it’s like a reverse Martingale. You have a small probability of winning a fortune very quickly but a huge probability of going completely broke. Compare that to a smart player who has a small chance of going broke and a huge chance of winning money for the rest of his life. Would you rather be very likely to earn a lot of money over a long time or be very unlikely to win a lot very quickly? If you prefer the latter then you should think about playing the lottery instead of BJ. What good is having an advantage if you are most likely going to lose all your money anyway? Even though you have an advantage, the house still has the edge.
The whole purpose of getting an advantage is so that you don’t have to rely on luck anymore. When you are skillful you want to eliminate the luck as much as possible because it swings both ways unexpectedly. If you manage your risk properly then you will make money slowly but infinitely. If you just want to take a shot at winning a big jackpot then you can find plenty of strategies for that as well, but it will most likely end badly. The choice is yours, but I think most people would prefer a 99% chance of winning a million dollars over 20 years to a 25% chance of winning a million dollars today, especially when the alternative involved going broke.
You can’t use standard deviation to estimate you RoR because SD only applies to the results at the end of the time period. You are very likely to lose more than the standard deviation would suggest sometime during that time which will make any further results useless. If I remember correctly, Schlesinger showed that the true RoR is about twice as big as the SD for the same period. If you are using SD instead of RoR then you are severely underestimating the level of risk that you are taking.
-Sonny-
The whole purpose of getting an advantage is so that you don’t have to rely on luck anymore. When you are skillful you want to eliminate the luck as much as possible because it swings both ways unexpectedly. If you manage your risk properly then you will make money slowly but infinitely. If you just want to take a shot at winning a big jackpot then you can find plenty of strategies for that as well, but it will most likely end badly. The choice is yours, but I think most people would prefer a 99% chance of winning a million dollars over 20 years to a 25% chance of winning a million dollars today, especially when the alternative involved going broke.
You can’t use standard deviation to estimate you RoR because SD only applies to the results at the end of the time period. You are very likely to lose more than the standard deviation would suggest sometime during that time which will make any further results useless. If I remember correctly, Schlesinger showed that the true RoR is about twice as big as the SD for the same period. If you are using SD instead of RoR then you are severely underestimating the level of risk that you are taking.
-Sonny-
That's exactly what I get using my ror calculator for a fixed EV and fixed bet of 1 unit. 
Input 01 for EV with Pos radio button selected (equivalent to 1% EV in the program), 1 unit for bank, and 1000000 units for goal as a representation of playing virtually forever.
ror peaks at slightly over 98%. (Dead link: http://www.bjstrat.net/cgi-bin/ror.exe) _ fixed EV ror calculator_.
As Sonny mentioned, for blackjack it is different and probably greater.
k_c
Input 01 for EV with Pos radio button selected (equivalent to 1% EV in the program), 1 unit for bank, and 1000000 units for goal as a representation of playing virtually forever.
ror peaks at slightly over 98%. (Dead link: http://www.bjstrat.net/cgi-bin/ror.exe) _ fixed EV ror calculator_.
As Sonny mentioned, for blackjack it is different and probably greater.
k_c
i put my units for 15$ units at 700 units for my 10000 some odd dollars i have to play with in that calculator and it gives me a probibility of failure below 1:1000 that doesnt seem in the balpark fo what iv read about, should i perhaps make my unit my average bet instead? so maybe a 40$ to acoomodate for bet spread?
Since I Can't Shake You!!!!!!!!!!!!! LOL
You are in a virtual room with how many people? All of them are questioning your judgement. You ask for opinions but they seem to mean nothing to you. The people who commented have all stated about the same thing. DON'T OVERBET!!!! They probably have far more blackjack experience and life experience then you do. We are showing concern for a total stranger, you should listen carefully to what we are saying! I imagine many of those who have commented have read Griffin, Schlesinger, Snyder and Wong among others and they probably know a little about blackjack and they talk about the perils of overbetting. If you have not heard of all of those names then that is another indication that your knowledge base may be lacking.
quote from Wong "Professional Blackjack"
pg 103
It can be proved mathmatically that an overbettor who consistently bets twice the optimal amount WILL BREAK EVEN OVER THE LONG HAUL. The overbettor who bets twice the optimal amount Will GO DOWN instead of up in the long run"
I hope you never need surgery because I am afraid you will perform it yourself!!!!!
You are in a virtual room with how many people? All of them are questioning your judgement. You ask for opinions but they seem to mean nothing to you. The people who commented have all stated about the same thing. DON'T OVERBET!!!! They probably have far more blackjack experience and life experience then you do. We are showing concern for a total stranger, you should listen carefully to what we are saying! I imagine many of those who have commented have read Griffin, Schlesinger, Snyder and Wong among others and they probably know a little about blackjack and they talk about the perils of overbetting. If you have not heard of all of those names then that is another indication that your knowledge base may be lacking.
quote from Wong "Professional Blackjack"
pg 103
It can be proved mathmatically that an overbettor who consistently bets twice the optimal amount WILL BREAK EVEN OVER THE LONG HAUL. The overbettor who bets twice the optimal amount Will GO DOWN instead of up in the long run"
I hope you never need surgery because I am afraid you will perform it yourself!!!!!
This gets us back to the risk of ruin argument we had in the other thread. I think people are jumping down your throat because they're using weird risk of ruin calculations. Are you saying that you're using a 700 unit bank? If so, are you using a normal spread, like 1-8 backcounting, or 1-15 or 1-20 play all? If so, your risk of ruin is certainly below 5% before expenses.
I have a weird attack plan that involves play all except negetive but minimizing hands at neutral counts and playing two at high counts when there are other peopel at the table only, i try to let other players eat up the cards by playing less at low counts and tow sometimes three at high counts.
but if the count falls and is still in my favor, i keep my max bets out there to get more money into action and weigh toward more advantage
-<-0.5-0
-0.5---(0.5 units)1 unit every other hand
0------1 unit
1------2 hands 1.5 unit average (due to occasional progressive cover and supsticious bet raising)
2------2 hand 6 unit
3------2 hand 9 unit avg
4+-----2 hand 10 unit
but if the count falls and is still in my favor, i keep my max bets out there to get more money into action and weigh toward more advantage
-<-0.5-0
-0.5---(0.5 units)1 unit every other hand
0------1 unit
1------2 hands 1.5 unit average (due to occasional progressive cover and supsticious bet raising)
2------2 hand 6 unit
3------2 hand 9 unit avg
4+-----2 hand 10 unit
Oh yeah - no reason whatsoever to assume anything else than spreading $15-$120 while backcounting, $15-$225 or $15-300$ in play-all, all with the same $10K bankroll and using normal spreads, would all certainly be less than 5% ROR.
Pretty much identical scenarios after all.
Care to share any of your "non-weird" ROR calcs? Maybe too intuitively obvious to bother?
Perhaps best, as a start, would be to, after ignoring all small bets and only using max bets, as originally described by poster, all satndard stuff, use the standard, non-weird "[100-(probibility of deviation totalling bank in percent)]/2" formula as suggested.
It's late. I'm sorry. But really. If I'm missing something, someone straighten me out.
Pretty much identical scenarios after all.
Care to share any of your "non-weird" ROR calcs? Maybe too intuitively obvious to bother?
Perhaps best, as a start, would be to, after ignoring all small bets and only using max bets, as originally described by poster, all satndard stuff, use the standard, non-weird "[100-(probibility of deviation totalling bank in percent)]/2" formula as suggested.
It's late. I'm sorry. But really. If I'm missing something, someone straighten me out.
It's only a game
Maybe you take this game too serious or have a serious problem. It's only a game and you are trying to beat the casinos. In light of some of the past cases of gamblers who lose it maybe you do need a break from. The senior members that have a wealth of both knowledge and oppinions can tell you everybody needs a break from tables lest you lose your edge. blackchipjim
Maybe you take this game too serious or have a serious problem. It's only a game and you are trying to beat the casinos. In light of some of the past cases of gamblers who lose it maybe you do need a break from. The senior members that have a wealth of both knowledge and oppinions can tell you everybody needs a break from tables lest you lose your edge. blackchipjim
Edit: I was going to go off on Kasi. But, nevermind. We have enough flamewars on this board as it is. Kasi, the formula you propose is fine; but I've seen ROR numbers quoted when there was nowhere near enough info to have made that calculation. Bottom line: I think risk of ruin vs. chance of doubling is much more useful, especially for a team.
Last edited:
I'm somewhat disappointed since all I really ever want is a discussion anyway.
I sometimes say things in an extreme way just to provoke discussions as I guess I did in your case. But how I say them doesn't change what I think.
But I try to not make personal crticisms and if I sometimes cross the line in your mind my profuse apologies. In my mind I'm just expressing an opinion on something someone said.
So feel free to flame away. Hopefully in the spirit of debating opinions, facts etc rather than character assassination. Of which, I agree with you and then some ,too much of that going on lately.
So, in that spirit, flame away!
And I do also agree with you that often, heck, almost always, there is woefully inadequate info given to make a meaningful conclusion. If there's anyone that believes God is in the details, it's me.
So, if you want to fill in any more, and enough, details as to why you considered the 3 games and spreads I mentioned all have less than 5% ROR's, great. To me, under almost anything I can think of, they would basically be apples and oranges - overbetting in most playall scenarios (high ROR) and underbetting (low ROR) in a backcounting scenario. But, like you say, it was basically a guess with assumptions based on the usual inadequate info.
Or maybe jump in with why you think that formula is fine, with or without ignoring all bets less than a max bet, that would be great. In all honesty I could use some enlightenment on it because I don't understand it at all but I may have just been thrown off by that ignoring all small bets stuff.
This ROR stuff is very hard to get a grasp on, for me anyway, coming in so many different flavors as it does. Lifetime. Trip. Losing whatever before winning whatever, etc. So having so many points of view on it is not surprising.
Who knows, maybe one of us, or someone reading, you never know, might get a new thought on something,
I sometimes say things in an extreme way just to provoke discussions as I guess I did in your case. But how I say them doesn't change what I think.
But I try to not make personal crticisms and if I sometimes cross the line in your mind my profuse apologies. In my mind I'm just expressing an opinion on something someone said.
So feel free to flame away. Hopefully in the spirit of debating opinions, facts etc rather than character assassination. Of which, I agree with you and then some ,too much of that going on lately.
So, in that spirit, flame away!
And I do also agree with you that often, heck, almost always, there is woefully inadequate info given to make a meaningful conclusion. If there's anyone that believes God is in the details, it's me.
So, if you want to fill in any more, and enough, details as to why you considered the 3 games and spreads I mentioned all have less than 5% ROR's, great. To me, under almost anything I can think of, they would basically be apples and oranges - overbetting in most playall scenarios (high ROR) and underbetting (low ROR) in a backcounting scenario. But, like you say, it was basically a guess with assumptions based on the usual inadequate info.
Or maybe jump in with why you think that formula is fine, with or without ignoring all bets less than a max bet, that would be great. In all honesty I could use some enlightenment on it because I don't understand it at all but I may have just been thrown off by that ignoring all small bets stuff.
This ROR stuff is very hard to get a grasp on, for me anyway, coming in so many different flavors as it does. Lifetime. Trip. Losing whatever before winning whatever, etc. So having so many points of view on it is not surprising.
Who knows, maybe one of us, or someone reading, you never know, might get a new thought on something,