CRAPS SYSTEM - WHAT'S WRONG WITH IT?

Quote: JORDAN1

I saw your computer simulation, and it is close to what I do, but still there is one variant that, to me seems to be part of the program, which probably doesn't matter. If a shooter makes any of my don't numbers, I stop betting against that shooter, take all of the odds on all the numbers I have established don't bets with, and wait until another shooter turn arrives.

That reduces your average hourly loss, but only because you're betting less (putting less money on the table). The house edge isn't affected. It's still 0% on the Odds bets and ~1.4% on everything else.

Quote: JORDAN1

Another question would be just a single coin toss bet. No house advantage, just a 50/50 bet. Is there a way to come out ahead on such a game?

No. The house edge on such a game is zero, so you're asking how to make 0 become greater than 0. Can't be done. Zero = 0. Zero is not >0.

Quote: JORDAN1

Everyone hates sessions, but don't they deserve some merit, especially in a game of coin toss?

I address the significance of sessions in my article on betting systems.

Quote: JORDAN

What would happen if you played a coin toss game with four simple rules.

1 -Whenever you lose three coin tosses in a row, stop a session. (Could be a hundred flips occur but you only stop when you lose three in a row)

2 - If you lose three in a row starting out a session, stop the session.

3- Lose a total of three tosses (like this -1, 0, -1, -2, -1, -2 ,-3) stop the session.

4 - If you get down to -2 and get the losses back to 0 (even) stop the session.

The long-term average is dead even. Stopping based on results doesn't help or hurt you (besides that you're no longer putting money at risk). I cover this in detail in my article on the Gambler's Fallacy.

Quote: JORDAN1

Is there any way that one could turn this into a winning game?

No. Again, I refer you to my article on betting systems, which provides four different pieces of evidence for why betting systems don't overcome the house edge.

Quote: JORDAN1

The reason behind my question if the house odds can get down to .0004, isn't the game as close to a coin toss as possible? So if it is, then couldn't sessions be used to possibly obtain a possible (ever so slightly) advantage?

No. If you're playing a game with an expected result of -0.04%, then the game is negative, period. A negative number is not a positive number.

You're not the first person to have these ideas, but they're very old ideas. I mean literally centuries old. They've been studied for that long, and the conclusion is always the same: Changing the amount, frequency, or criteria of bets doesn't change the outcome. No one has ever shown otherwise.

Quote: JORDAN1

If a shooter makes any of my don't numbers, I stop betting against that shooter, take all of the odds on all the numbers I have established don't bets with, and wait until another shooter turn arrives.

that is a 'feel good' play only and NO such math or
simulation results shows that it makes you WIN more money.

IF you bet less, on average, you lose less
PERIOD

Quote: JORDAN1

The shooter that knocked down one of my don't numbers might roll for 30 minutes or more, but I don't dare make another bet.

LOL. Playing SCARED never wins.
OR the shooter might just 7 out on the very next roll. IF a point is established it is the single most probable outcome of any one roll - a 7.

There have been some craps gaming writers, Frank Scoblete, to name one, that says to lose only one bet against a random shooter...
He NEVER has shown ANY math or simulation data to back his FALSE claim and he is one that has used math guys - abortions to their work - to make his ideas, THAT ARE GARBAGE, smell sweeter.

IMO, play Craps to have fun as you really will not win over a lifetime of play.
You only get one lifetime session.

of course, believe what you want and 'flip the bird' to the rest IF that makes you feel even better while playing Craps.

love this tee:...........................................................................

So where can we get dice tee shirts?

Quote: DeMango

So where can we get dice tee shirts?

you can get the one I posted at this link but I don't know anything about the company and can't vouch for them:


https://www.teepublic.com/t-shirt/4124937-vintage-the-dice-giveth-and-taketh-away?utm_source=paid_search&utm_medium=ppc&utm_campaign=PLA&utm_content=tshirt-unisex&feed_sku=4124937D1V&ar_clx=yes&ar_channel=google_ads&ar_campaign=1510062400&ar_adgroup=57012254846&ar_ad=352852398878&ar_strategy=search&gclid=EAIaIQobChMIlJmQtpnO5AIVE4vICh1URAgEEAQYAyABEgJFx_D_BwE&gclsrc=aw.ds


here's another tee I like:

Quote: JORDAN1

Another question would be just a single coin toss bet. No house advantage, just a 50/50 bet. Is there a way to come out ahead on such a game?

For this no-edge scenario there is zero reason to run a simulation. Your question is answered completely by OnceDear's rule of Thumb
Or. TLDR. No. You cannot get an edge. Yes, you could probably come out ahead: Pointless.

Quote: JORDAN1

Everyone hates sessions, but don't they deserve some merit.

There it is, my cue that I can stop reading a system thread.

Quote: JORDAN1

Another question would be just a single coin toss bet. No house advantage, just a 50/50 bet. Is there a way to come out ahead on such a game?

Only if you have unlimited time and unlimited budget.

This has been discussed before here. Time to break out my "50/50 game analyzer" that counts the number of coin flips needed to end up ahead.

After a run of 100,000 games, there were 16 instances where you needed more than 1,000,000 flips before you finally came out ahead; in four of them, you needed more than a billion flips, and in all four cases, there was a point where your losses exceeded 45,000 bets (in two of them, they exceeded 75,000).

Thanks for all the insights and the lengthy work on that simulator you made to evaluate my "system." Funny thing, a couple of clicks into running the sim, my bankroll turned from $5000 to $13000. The overall loss was over $30,000 for a million rolls, but if I was playing my "system" and ran $5000 to $13000, there would be no way I would continue playing all the way back to zero dollars or even a loss. If $13,000 was hit, I would stop when I got it back down to $10,000. I know, I know, and oh yeah, I know... Sessions are good for nothing.

Another fun thought. The casino wins with their house edge over a bunch of bets over a long period of time. What if... You ONLY gamble 5 times in your life? Each time you go gambling you take $5000. You play games that offer as close to a 50/50 bet as possible. You bet it all at one time.

You have these outcomes.

Lose all 5 - Total Loss $25000
Lose 4 out of 5 - Total Loss $20,000
etc...

or

Win all 5 - Total gains $25,000
Win 4 out of 5 - Total gains $20,000
etc...

Not a bad way to go at it. Losing $25,000 over a life-time of $1 bets is the same as the 5 shots described above, but the house advantage hardly comes into play with this way of betting. Will anyone try this? No, but this is probably the best way to gamble in one's life. Better than Martindales and Fibonacci systems, anyway.

Thanks again.

Fibonacci sequence is 1, 1, 2, 3, 5, 8, 13, 21, etc... Similar to a Martindale system, but in my little opinion a bit more stable.

Quote: JORDAN1

Thanks for all the insights and the lengthy work on that simulator you made to evaluate my "system." Funny thing, a couple of clicks into running the sim, my bankroll turned from $5000 to $13000. The overall loss was over $30,000 for a million rolls, but if I was playing my "system" and ran $5000 to $13000, there would be no way I would continue playing all the way back to zero dollars or even a loss. If $13,000 was hit, I would stop when I got it back down to $10,000.

stop for how long?

Quote: JORDAN1

I know, I know, and oh yeah, I know... Sessions are good for nothing.

and you
still only get one lifetime session

Quote: JORDAN1

Funny thing, a couple of clicks into running the sim, my bankroll turned from $5000 to $13000. The overall loss was over $30,000 for a million rolls, but if I was playing my "system" and ran $5000 to $13000, there would be no way I would continue playing all the way back to zero dollars or even a loss. If $13,000 was hit, I would stop when I got it back down to $10,000.

Yes, systems don't lose every session, but for that matter neither does flat-betting. When I have time, I'll add a feature to the sim to have it test thousands of 1-hour sessions so the reader can see that the overall result is a loss, even if some of the sessions are winners.

Yes, I agree that if you get a big win, you should set a loss limit and stop playing! Some folks on this forum pooh-pooh the idea of quitting while you're ahead, saying it doesn't work, but it can't be disputed that if you get ahead and quit, you can't lose the money that you're not gambling. Quitting while you're ahead doesn't mean that you have a long-term advantage over the house, but you certainly benefit from stopping your play and not giving it back to the casino.

Quote: JORDAN1

Another fun thought. The casino wins with their house edge over a bunch of bets over a long period of time. What if... You ONLY gamble 5 times in your life? Each time you go gambling you take $5000. You play games that offer as close to a 50/50 bet as possible. You bet it all at one time.

Yes, by minimizing the number of bets, you reduce your exposure to the house edge. I have an article about a similar idea, my Halfies System. It doesn't overcome the house edge (nothing does), but it gives you a way better shot at doubling your bankroll in the short term than flat-betting does.

Quote: JORDAN1

Will anyone try this? No, but this is probably the best way to gamble in one's life. Better than Martindales and Fibonacci systems, anyway.

Yes, and better than flat-betting, if the purpose is to double a bankroll or better. But if the goal is to get a lot of play time, obviously it doesn't fare so well.

Quote: MichaelBluejay

Yes, systems don't lose every session, but for that matter neither does flat-betting. When I have time, I'll add a feature to the sim to have it test thousands of 1-hour sessions so the reader can see that the overall result is a loss, even if some of the sessions are winners.

this goes against the majority of gaming writers that only state the (-)ev of any session played.

It is pounded into the brain over and over you always lose x% of all your bets made.
negative numbers add up to a negative number.

Once a person finds out that is not true for every session (not one lifetime session), the concept of sessions becomes way more important to them as they have proved to themselves that one CAN WIN in any one session, and it could be the end of their lifetime session.

The reason so many gaming writers use ev and not
risk of ruin is
the math is way more difficult for the later and most can't go there even if they wanted to.

Alan Krigman comes to mind as one who expressed session win probabilities
all others just quote -ev and you lose
done

Quote: 7craps

this goes against the majority of gaming writers that only state the (-)ev of any session played....The reason so many gaming writers use ev and not
risk of ruin is
the math is way more difficult for the later and most can't go there even if they wanted to. Alan Krigman comes to mind as one who expressed session win probabilities
all others just quote -ev and you lose
done

Well, as I like to say, I'm not most gambling writers. :) Someday I'll devote a whole article to the various ways I buck conventional wisdom, and even where I disagree with the Wizard.

Like Krigman, I do indeed post session win probabilities, and plan to post more for various systems when I have time. I also explain how just because a system *ever* wins in the short term, the overall result of all short sessions is a loss.

Quote: MichaelBluejay

Halfies System. It doesn't overcome the house edge (nothing does), but it gives you a way better shot at doubling your bankroll in the short term than flat-betting does.

I read your blog about your halfies system in which you state this:

"you place your entire bankroll (the amount you bring to gamble) on a single low-edge, even-money bet, such as a a baccarat hand, a pass line bet on craps, or a 𝐡𝐚𝐧𝐝 𝐨𝐟 𝐛𝐥𝐚𝐜𝐤𝐣𝐚𝐜𝐤 "



your advice re using blackjack for this strategy is not good

blackjack is only a low house edge bet if you are willing to double and split when called for in basic strategy
since you are putting your entire bankroll on the bet you are not able to do that
by not having doubling or splitting available you have increased the house edge significantly in the game of blackjack


also - to say that a bet on a blackjack table is a low house edge bet is misleading - it's only a low house edge bet if you have the considerable knowledge of basic strategy in the game

Quote: lilredrooster

I read your blog about your halfies system in which you state this:

"you place your entire bankroll (the amount you bring to gamble) on a single low-edge, even-money bet, such as a a baccarat hand, a pass line bet on craps, or a 𝐡𝐚𝐧𝐝 𝐨𝐟 𝐛𝐥𝐚𝐜𝐤𝐣𝐚𝐜𝐤 "

Whoops, you're absolutely right, I just removed the reference to blackjack, thanks for catching it. I don't know why I would have mentioned blackjack, except I wrote that article nearly 20 years ago and I was more prone to those kinds of errors then.

By the way, my website is not a blog.

Quote: MichaelBluejay

I also explain how just because a system *ever* wins in the short term, the overall result of all short sessions is a loss.

this is still not correct as the conditions are not stated.

Here is a very small table of Marty probabilities.
Condition: Doubling the starting stake or stop when Marty breaks (lose Max bet)

on average, it takes a little more than 2 trials per unit stake.
for a 13 step Marty, one could be looking at over 16,000 rounds played.
lots of small sessions could add up to this. many think 16k is the long run.

step	Max unit Bet	required stake (units; 2^step-1)	double prob (craps pass line - no odds)	double prob (00 Roulette - evens)
2	2	3	41.00%	37.79%
3	4	7	37.61%	33.18%
4	8	15	35.84%	30.19%
5	16	31	34.75%	27.86%
6	32	63	33.96%	25.83%
7	64	127	33.31%	23.96%
8	128	255	32.73%	22.18%
9	256	511	32.18%	20.47%
10	512	1023	31.65%	18.83%
11	1,024	2047	31.14%	17.24%
12	2,048	4095	30.62%	15.71%
13	4,096	8191	30.11%	14.26%

the math for this is simple but not included to keep things even simpler

25% to 35% success rate to double a starting bankroll, for many, is worth the time and effort and bankroll and comps and entertainment value.

Quote: 7craps

this is still not correct as the conditions are not stated.

I'm going to have to disagree. I don't think it's necessary to state any conditions (other than "normal"). Run 100k short sessions, sum all the wins and losses, and the result will be an overall loss. Sure, if you have a $10M bankroll and you quit each session as soon as you're up $5, then the sum of all short sessions will be an overall winner, but I'm assuming that we're not using ridiculous variables that nobody is thinking of when they wonder whether systems win in the short term.