Bankrolls and risk of ruin

If I was playing the below games what should I be taking as a bankroll when considering RoR?

$50 Blackjack 6 deck with surrender, resplit, double on any, hit soft 17
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
70 x 0.0075 x 50 x 4 x 7

$25 Craps pass line (& odds), place 6 & 8
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
48 x 0.0141 x 75 x 4 x 7

$50 Baccarat banker
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
72 x 0.0106 x 50 x 4 x 7

$50 Roulette 00 inside spread
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
38 x 0.0526 x 50 x 4 x 7

Apologies in advance if the hands per hour are out but these are only estimates.

search for calculators online

I've had a look but couldn't find any sites other than ones stating x50-100 your average stake.

These sites didn't seem to account for time spent playing which I would have thought would definitely make a difference.

Could anyone recommend a site for this?

I've had a look but couldn't find any sites other than ones stating x50-100 your average stake.

These sites didn't seem to account for time spent playing which I would have thought would definitely make a difference.

Could anyone recommend a site for this?

is this usable?

https://wizardofodds.com/games/blackjack/appendix/12/

Quote: Jimmyburke1927

If I was playing the below games what should I be taking as a bankroll when considering RoR?
$25 Craps pass line (& odds), place 6 & 8
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
48 x 0.0141 x 75 x 4 x 7.

Risk of Ruin is expressed per session and is generally calculated at the ten percent level. It is calculated on the basis of no tips, no mistakes, no toke bets, no tips to waitresses. So if you drink but don't tip and start to make errors from fatigue or alcohol, that changes your risk. If you do Odds, its more money. If you Dark Side, its slightly more, if you place six and eight, its pretty much the same.

So to just stand there for four hours of 25.00 Pass Line bets with no odds, no Place Bets, no tips, no nothin'., you need to have 20 units, for the combinations of bets you suggest. A four hour session with your full play and reasonable tips and errors would require you to have about 2,200.00, but please remember I'm required to use my fingers to get past ten, so calculate yourself if possible.

The concept of risk of ruin is that normal variation can get very impressive on the plus or minus sign. RoR tells you what you need to have a ninety percent chance that normal variation will allow you to still be in the game rather than totally wiped out.

Quote: Jimmyburke1927

If I was playing the below games what should I be taking as a bankroll when considering RoR?

$25 Craps pass line (& odds), place 6 & 8
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
48 x 0.0141 x 75 x 4 x 7

$50 Roulette 00 inside spread
Hands per Hour X House Edge X Avg Bet X Hours Played X Days
38 x 0.0526 x 50 x 4 x 7

Apologies in advance if the hands per hour are out but these are only estimates.

estimates are fine.
but more important is how the bets are made, and increasing of the bets and with Craps what odds?

I have lots of data on this but are you talking about 2X odds always or 345x etc
and the place bets. $24 each. are you pressing up?
If the point is 8 do you take down the Place 8 and make the Place 6 $48?

The Roulette "inside spread" is that $5 on 10 numbers straight up?

the concept I use, with simulations, is to use expected value and standard deviation.
RoR can be exactly calculated, approximated with a small error, or by those that say to use 10X your large bet or some other "rule of thumb" who knows how accurate that can be?
Sally

Quote: FleaStiff

So to just stand there for four hours of 25.00 Pass Line bets with no odds, no Place Bets, no tips, no nothin'., you need to have 20 units, for the combinations of bets you suggest. A four hour session with your full play and reasonable tips and errors would require you to have about 2,200.00, but please remember I'm required to use my fingers to get past ten, so calculate yourself if possible.

How many bets did you base your session from?
OP wants to use at first 48*4 for a one day session of 192 bets
I get about 20% RoR just with a 20 unit bankroll and 192 bets (20: 0.19628452)
10% comes in at around 128 bets (20: 0.102354587)

If you show how you did yours and I can show how I did mine

OP is off to gamble!
Sally

Quote: mustangsally

If you show how you did yours and I can show how I did mine, Sally

If I show you my calculations you are likely to die laughing at my grasp of arithmetic or what I refer to as "higher mathematics".

I used the standard 20 units of Craps bet for Four Hours of PASS line betting with no other bets, no tips, no tokes, no mistakes by dealers, no mistakes by player. (I obviously do not think this is a valid real world situation, no matter how stingy a player might be).

I used as a source, Vegas Click which used to have a slightly more informative graphic on this issue but I assume that the information is unchanged and so I was able to obtain if from some notes derived from the random recesses of my computer's hard drive and random recesses from my biological hard drive that were originally obtained from the same site.

Their site used to reveal the average "hands" for various games and the quick calculation formulas the house used to use for comp calculations even if actual play was not quite the easily calculated theoretical play.

I figured if its 20 units for a simple pass line bet with no odds, then it would be about the same for Placing the Six and Eight repeatedly since those bets have "about" the same house edge. I also figured the odds bets would be a good bit more. I allowed tips to dealers and tips to waitresses, the effects of fatigue and the effects of alcohol to equal a separate bet roughly approximating perfect play at the Pass Line. I figure if he is betting 25.00 on the line he would throw in two greens for the place bets to be positioned as 48.00 in bets and two dollars as a constant drain of annoying "ones'' that he will probably use as tips/tokes/etc or waste on scattered impulse bets such as center bets for himself or the boys and girls.

So this amounted to the player subjecting his bankroll to five bets (letting odds count double) with a bit of a fudge for tips and errors. This lead to 100 units needed for 4 hours of play which since he is basically "betting Greens" would work out to 2,500, but since tips and errors are not all that much, I reduced it to 2200.00 needed.

I told you you would laugh at my math skills.

Quote: FleaStiff

If I show you my calculations you are likely to die laughing at my grasp of arithmetic or what I refer to as "higher mathematics".

I did not die laughing. I actually had to read your post a few times.

Quote: FleaStiff

I used as a source, Easy Vegas

I see
so i went to
https://Easy.Vegas/gambling/bet-size
and
https://Easy.Vegas/gambling/average-loss

He uses different values from what I can tell in his rounds per hour. Not just 48
The 48 rounds comes from the Wizard table here
https://wizardofodds.com/gambling/house-edge/

I had also only pointed to probabilities of a 20 unit bankroll just making $25 pass line bets with no odds.

waiting for the OP to see his parameters that he would like to use instead of showing a bunch of data that would not apply to him.

still the concept of using math with expected value and standard deviation to find Bankroll and RoR is fun, IMO.
and like most math easy once you see how it is done

Quote: FleaStiff

I told you you would laugh at my math skills.

Not laughing yet.
I really have not figured out what you did from what you said because I found those two pages at vegasclick.com very interesting
as reading more about the man William Feller and reading his first book and working through some of his examples.
in other words, I pushed you to the end of the line for the moment

Sally

On craps... I assume you mean $85 total against the house edge. $25 line and $30 on the 6 and 8 each - if you meant 24 on the 6 and 8 each... the same numbers below pretty much apply. If you never press and play full odds at a 3,4,5 table... my gut tells me it's north of $2500 to play for four hours. The biggest issue is that your odds bets on the 6 and 8 can DESTROY you.

If this seems high... let me just point out that you will lose $150-200 per PSO and over the course of four hours you will see a handful of PSO streaks. $2500 is on the low end.

When it comes to roulette, it entirely depends on your playstyle. If you're throwing $50 on a single number you only have a 50% chance to hit it over the course of 26 spins. In other words, it's within reason (7%) that you could lose $5000 without ever hitting your number. However, if you bet it on black every time, you can probably survive on $2500 or so without any issues. I know that you said inside numbers, but I'm using the extremes to make a point.

Quote: ahiromu

On craps... I assume you mean $85 total against the house edge. $25 line and $30 on the 6 and 8 each - if you meant 24 on the 6 and 8 each... the same numbers below pretty much apply. If you never press and play full odds at a 3,4,5 table... my gut tells me it's north of $2500 to play for four hours. The biggest issue is that your odds bets on the 6 and 8 can DESTROY you.

so can the 4 and 10 not hitting for 6 or 7 times in a row.
This still brings up the difference between the "gut" feelings about bankroll requirements versers just simulationg it or calculating it.

The concept of using ev (expected value) and variance (actually standard deviation) to determine bankroll and the RoR values has been discussed before.
here is one of many
https://wizardofvegas.com/forum/questions-and-answers/math/10123-risk-of-ruin/

Will wait to see if OP comes back a winner first!

Quote: ahiromu

When it comes to roulette, it entirely depends on your playstyle. If you're throwing $50 on a single number you only have a 50% chance to hit it over the course of 26 spins.

Only a 50% chance is like flipping a fair coin.
and it should be only a 50% chance of winning at least 1 time
at least 2 times = 0.1488421 and a very nice profit of at least $2,300

That 26 spins BYW can come from this formula
# of trials =LOG(1-pct)/LOG(1-p)
pct = 0.50
p= 1/38
# of trials = 25.99

Quote: ahiromu

In other words, it's within reason (7%) that you could lose $5000 without ever hitting your number.
However, if you bet it on black every time, you can probably survive on $2500 or so without any issues. I know that you said inside numbers, but I'm using the extremes to make a point.

probably survive = ?. I hope you should survive. Maybe $2500 is too high?
how do we know??

38 spins per hour and 4 hours each day = 152 spins day total
50 units bankroll (2500/50 if flat betting)
the chance of busting a 50 unit bank in 152 spins = ?

win 51 lose 101 would do
but so would win 46 and lose 96. that is only 142 spins
it can not be that high? my gut tells me

maybe this formula by the math greats (Feller included)

oh,oh
that 2ⁿ = 2^152 to start out.
never fear cos and sin and a little pi thrown in too. easy stuff once you see how to use it.

how about just ev and sd values.
maybe too easy?
how about a sim?
can't sim? or too lazy? I get lazy at the strangest times I am told

recursion in Excel? yes, already done


I do love pie
Sally
seconds

So this kind of is along the same line of a question I've been pondering lately for VP. I'm relatively new to knowing terms like full pay and risk of ruin, haven't been actually studying video poker strategy for that long, but now that I'm well on my way to being able to competently use optimal play strategy here's my big question on RoR vs. full pay:

Do you play a full pay machine if the minimum betting denomination is such that you have a small number of hands you could consecutively play before ruin or do you play a non-full pay machine with a lower minimum betting denomination?

My specific dilemma is that I typically will bring a $300 bankroll for video poker with me to Vegas. At the casino I prefer to go to (Luxor), the only full-pay machines they have are in the high limit room and I would have to bet $1 minimum. Assuming I bet 5 units so that the royal flush bonus is in play, which based on all I've read I don't understand why you wouldn't, that gives me 60 hands I could play before ruin which doesn't seem like a lot at all.

My alternative would be to play a non-full pay machine, this casino has both 8/6 and 9/5 available in $0.25 denomination, with that same bankroll which would give me 240 hands to play as opposed to 60 but of course the ER is certainly not going to be 99.54% due to not being full pay. I'm not willing to travel to another casino just to find a full-pay machine with a lower betting threshold.

I tried to simulate the 60 hand scenario to get an idea on the ruin % using VideoPoker! on iPhone/iPad and so far I have only busted out 1/5 times with those limits in place. Still, I have to wonder if that's not a great way to bust out my bankroll within the first hour or two of play and then have nothing left to play with for the duration of the trip. I can't say I would be devastated, since I don't gamble with money I'm not happy with losing, but I would be disappointed if I busted out that quickly since gambling is part of what I enjoy about going to Vegas.

Quote: scubatim84

Do you play a full pay machine if the minimum betting denomination is such that you have a small number of hands you could consecutively play before ruin or do you play a non-full pay machine with a lower minimum betting denomination?

My specific dilemma is that I typically will bring a $300 bankroll for video poker with me to Vegas.

I like your questions
sounds fun but
your first few statements are confusing to me.

I myself try to play the best pay tables but at times here in SoCal that will not happen every time so I play what I can play

Quote: scubatim84

At the casino I prefer to go to (Luxor), the only full-pay machines they have are in the high limit room
and I would have to bet $1 minimum.
Assuming I bet 5 units so that the royal flush bonus is in play, which based on all I've read I don't understand why you wouldn't,
that gives me 60 hands I could play before ruin which doesn't seem like a lot at all.

I disagree with you on this
I say 60 hands played at $5 = $300 and I would say you would not be ruined after that few hands played.
do you see whys?

Quote: scubatim84

I tried to simulate the 60 hand scenario to get an idea on the ruin % using VideoPoker! on iPhone/iPad and so far I have only busted out 1/5 times with those limits in place.

that program must be off.
you can not bust a $300 bankroll playing $5 per hand for 60 hands.
not possible unless the game is rigged.

using Video Poker for Winners I show this bust rate for X number of hands played (calculated values)
9/6 $1 JOB
120: .02% (or abouts 1 in 5000)
240: 3.12% (or abouts 1 in 32)
500: 20.95% (or abouts 1 in 4.8)
1000: 45.15% (or abouts 1 in 2.2)

go, play and have fun
I say
Sally

I'd bring at least a royals worth, half if you don't mind a short session. I usually carry this much and only tapped out once, and I've been playing all night.

Quote: mustangsally

I like your questions
sounds fun but
your first few statements are confusing to me.

I myself try to play the best pay tables but at times here in SoCal that will not happen every time so I play what I can play
I disagree with you on this
I say 60 hands played at $5 = $300 and I would say you would not be ruined after that few hands played.
do you see whys?

that program must be off.
you can not bust a $300 bankroll playing $5 per hand for 60 hands.
not possible unless the game is rigged.

using Video Poker for Winners I show this bust rate for X number of hands played (calculated values)
9/6 $1 JOB
120: .02% (or abouts 1 in 5000)
240: 3.12% (or abouts 1 in 32)
500: 20.95% (or abouts 1 in 4.8)
1000: 45.15% (or abouts 1 in 2.2)

go, play and have fun
I say
Sally

I get what you're saying about ruin...it's not going to happen off 60 hands but the better way to say it I suppose would have been that given 400-600 hands per hour, that doesn't leave me with a whole lot of playing time at that denom. Based on those numbers, I run a 45% chance of busting out after about 2 hours, which isn't too bad I suppose but this last trip I played for 5-6 hours. Although to be fair, this was on $0.25 8/6 JOB for part of the time and $0.25 9/6 DDB for the rest of the time (I thought 9/6 was full pay for DDB...lol).

How did you get those numbers off Video Poker for Winners btw? I'm looking at getting that program and that's a pretty nifty simulation to calculate before I go on the next trip.

Djatc, even at $0.25 a royal's worth would be $1,000. I just don't gamble with that much money...never have, and given that the ER is < 100% so inevitably I'll lose that money, the idea of dropping $4,000 is just crazy to me. I could almost go to the Caribbean for that much money.

Quote: scubatim84

given 400-600 hands per hour, that doesn't leave me with a whole lot of playing time at that denom. Based on those numbers, I run a 45% chance of busting out after about 2 hours, which isn't too bad I suppose

well, if playing VP = fun then playing more VP = more fun

in my opinion , to make your starting bankroll possibly last longer
play slower, track your starting hands in a notebook and people watch between hands
works well for me and I still have fun

also, there is no law saying you can not play the lower denoms at the lower pay tables.
7 of my 9 Lifetime Royals have come from those horrible non-full play games.
(other VPers have told me this and even said I should just give my winnings back to the casino)
I have also been called many names and all not flattering
by playing those less desirable pay tables by the so-called (self-proclaimed) VP ex-spits (experts),
and I still had fun playing them too.
ands
I still have my winnings

so in my opinion
be careful of other VP players opinions
they could be taken very lightly (because of fun) as, in my opinion,
VP ER only exists in the long run (playing forever or 10s of millions of hands - maybe) as you must agree
https://wizardofodds.com/ask-the-wizard/video-poker/deuces-wild/
"Perhaps a better question is how many hands would I have to play to have a 90% probability of getting to 100.66%?

The standard deviation of full pay deuces wild is 5.08.
The standard deviation of the sample mean is 5.08/sqr(n), where n is the number of hands.
I’ll skip (skip skip) over the rest of the math to the answer, which is 42,383,720 hands. "

Quote: scubatim84

How did you get those numbers off Video Poker for Winners btw?

I looked at them
there are other programs I use and actually got them from my Excel calculations that matched VPW, so one of us had to do the calculations correctly, in my opinion
The math is simple to do once you see it done and specialized VP programs are not needed in my opinion
but are OK to have taking out the guess work

Sally

Quote: mustangsally

be careful of other VP players opinions
they could be taken very lightly (because of fun) as, in my opinion,
VP ER only exists in the long run (playing forever or 10s of millions of hands - maybe) as you must agree

Exactly.

If your goal is to grind VP indefinitely for profit, then you really must look for the highest paytables.

If your goal is to play a while, and either accept a modest loss or quit while you're ahead, lower paytables - while not ideal - are reasonable.

Personally, I find 5c 6/5 3 hand JoB to be strangely soothing, even at 1 coin per hand. I can get in a ridiculous number of hands for $5, and probably make my play last longer than an equivalent dollar amount on PacMan.

I'm not playing it to make money, I'm playing it to relax. Of course, every once in a while, I'll find myself going crazy and betting 2 (or - gasp - 3) coins a hand, flopping a full house or 4oak, and walking away with an extra $5.

Strangely enough, there's hardly ever any competition for those machines... except about an hour and a half before the tour bus picks up the old people for the ride back to the retirement home.

OR, you could get away from the strip and find plenty of FP games at any denomination you like. VPFREE2 is a good reference. The crapy machines exist because people are willing to play them. Think of it this way, on a 99% machine you lose money twice as fast as 99.5% machine. Suit yourself.

Anyone planning on making money solely playing FP VP, GL & GB.

For entertainment purposes go for it. Other than that Forghedaboudit.

If you are planning on losing a certain amount and you only play once in a while, just play whatever game you like best.

If you are retired and want to earn comps and keep busy and sharp FP VP is great. Older VP players seem to keep sharp.

Quote: Sabretom2

The crapy machines exist because people are willing to play them.

yes sir
Playing VP = fun

and many times only the crapy machines are available to play and players win on crapy machines too.
Even 2 Royals in 4 hands

Quote: Sabretom2

Think of it this way, on a 99% machine you lose money twice as fast as 99.5% machine. Suit yourself.

That is playing the same denom, I thinks.
scubatim84 is looking for extended playing time for his $300 bankroll, I thinks
and 25 cents games I would think so in my opinion would suit him well

so let us see if your argument suits him well too.

$1 9/6 JOB
1000 hands played
EV = 1000 * $5 * -.004559 = -$22.795 eh
Chance of a $300 bankroll ruin = abouts 45.15% (a little tight around the neck this one)
Chance of a session profit = abouts 32.02% not great

25 cent 8/5 JOB
easy to find these I say
1000 hands played
EV = 1000 * $1.25 * -.027 = -$33.75 eh
Chance of a $300 bankroll ruin = abouts 0.01% ahhhh
Chance of a session profit = abouts 22.37% not great

not even twice as fast as your claim (and I used a reals crappppppy pay table we both must agreesssss)
maybe my math is wrongss
please show your math on this

Sally

Quote: mustangsally

yes sir
Playing VP = fun

and many times only the crapy machines are available to play and players win on crapy machines too.
Even 2 Royals in 4 hands

That is playing the same denom, I thinks.
scubatim84 is looking for extended playing time for his $300 bankroll, I thinks
and 25 cents games I would think so in my opinion would suit him well

so let us see if your argument suits him well too.

$1 9/6 JOB
1000 hands played
EV = 1000 * $5 * -.004559 = -$22.795 eh
Chance of a $300 bankroll ruin = abouts 45.15% (a little tight around the neck this one)
Chance of a session profit = abouts 32.02% not great

25 cent 8/5 JOB
easy to find these I say
1000 hands played
EV = 1000 * $1.25 * -.027 = -$33.75 eh
Chance of a $300 bankroll ruin = abouts 0.01% ahhhh
Chance of a session profit = abouts 22.37% not great

not even twice as fast as your claim (and I used a reals crappppppy pay table we both must agreesssss)
maybe my math is wrongss
please show your math on this

Sally

Yeah as everyone points out, the non-full pay are much easier to find and really to play an equivalent amount each hand its either $1 full pay only betting 1 unit per hand (so no royal flush bonus in play), at which point if I DID hit a royal flush I'd cry because I'd only win $250, vs. playing $1.25 non-full pay at 9/5 or 8/6 (both available at Luxor although I believe 9/5 is slightly better on ER) where if I did hit a royal flush I'd be one happy camper. :)

I actually was curious how you got to run a simulation that spit out those numbers in VPW...I just downloaded it last night to sharpen up my knowledge of the optimal play strategy on wizardofodds' web site and don't see where I can get that data. Wasn't questioning the numbers themselves.

Definitely given that either way the chance of a session profit is low, because 99.54% even at full pay still means you'll lose money over time, I really like the .01% chance of ruin vs. 45%, lol. Can't imagine why. :)

Quote: scubatim84

I actually was curious how you got to run a simulation that spit out those numbers in VPW...I just downloaded it last night to sharpen up my knowledge of the optimal play strategy on wizardofodds' web site and don't see where I can get that data.

It is not a simulation like this would be
http://www.beatingbonuses.com/simulator_java.htm

The program actually calculates the values.
I thinks by convolution, that would explain the slowness compared to a program that uses matrix multiplication

after selecting your denom to play visit the menu Analyze/Bankroll...
the Help explains it too in my version

have fun in Vegas!
I be there in 2 weeks
Sally

Quote: mustangsally

It is not a simulation like this would be
http://www.beatingbonuses.com/simulator_java.htm

The program actually calculates the values.
I thinks by convolution, that would explain the slowness compared to a program that uses matrix multiplication

after selecting your denom to play visit the menu Analyze/Bankroll...
the Help explains it too in my version

have fun in Vegas!
I be there in 2 weeks
Sally

Got it, yeah this program is pretty cool. Thanks and good luck!