Debunking Roulette Marty with pictures

As a little exercise, I've created an Excel workbook which illustrates dramatically how a progressive Martingale will generally give a bankroll with quite a steady initial uptrend punctuated with death defying crashes.

It allows you to experiment with waiting till xNumber of Reds appear before starting a martingale bet on the possibility of a Black to break the cycle.

It's nothing sophisticated and does not use macros, so is safe to run in excel 2010 or later.

What it shows pretty convincingly is that it is very likely that a Martingaler will see a nice easy and steady increases in his bankroll until a wholly unpredictable loss will set him back. Often the setback will wipe out profits and be survivable. Often it will be a total loss of bankroll. It also shows that waiting for a sequence to get established is just wasting time.

I'll leave it online a few months. If the owners of admins here wish to, they can host it themselves, subject to leaving it unchanged and attributing to me by name.








Martingale Debunking Workbook (You may need to 'Enable editing' in your Excel application.

It extends to 100,000 spins but works even better if you extend the rows by duplication. As per the instruction, just fire off a re-calculate by selecting cell a1 and hitting delete

Very interesting work, thanks.

However we must decide if we have a theoretical discussion about the concept of increase as you lose or a very practical discussion.
Theoretically the Martingale is based on some very interesting concepts that can give much food for thought. Theoretically speaking, (with unlimited bankroll) this method is a winner.

Practically speaking, you can't lose all your bankroll in one attack, because you would have reached the table maximum sooner :-)

In my opinion Martingale is overvalued by beginners and undervalued by "experts". None really understands the true value of its concept.

Quote: Kavouras

Very interesting work, thanks.

However we must decide if we have a theoretical discussion about the concept of increase as you lose or a very practical discussion.
Theoretically the Martingale is based on some very interesting concepts that can give much food for thought. Theoretically speaking, (with unlimited bankroll) this method is a winner.

Practically speaking, you can't lose all your bankroll in one attack, because you would have reach the table maximum sooner :-)

In my opinion Martingale is overvalued by beginners and undervalued by "experts". None really understands the true value of its concept.

If you have unlimited money then you have all the money therefore there's nothing to win.

Quote: AxelWolf

If you have unlimited money then you have all the money therefore there's nothing to win.

This is exactly why I wrote that we should clarify if we are doing a theoretical or practical discussion.

There is only practical :)

If you have infinite bankroll, you can never increase or decrease it :)

Most newbies discover Marty and believe it to have practical value: It does: if you have 1000 and will be shot unless you can raise 1100, then you marty and stand a good chance of surviving. But that's about it.

As a recent newbie messaged me on this concept "Don't worry my friend we will make a lot of money". i wrote this workbook to stop him taking a path to ruin and thought it worth sharing.

My objective was not to prove that Marty is without value, only to illustrate, practically, that it is no more likely to increase your bankroll to a meaningful extent (x2) than it is to destroy it totally.

Quote: OnceDear

There is only practical :)

If you have infinite bankroll, you can never increase or decrease it :)

Most newbies discover Marty and believe it to have practical value: It does: if you have 1000 and will be shot unless you can raise 1100, then you marty and stand a good chance of surviving. But that's about it.

As a recent newbie messaged me on this concept "Don't worry my friend we will make a lot of money". i wrote this workbook to stop him taking a path to ruin and thought it worth sharing.

My objective was not to prove that Marty is without value, only to illustrate, practically, that it is no more likely to increase your bankroll to a meaningful extent (x2) than it is to destroy it totally.

I mostly agree with you.

Quote: Kavouras

I mostly agree with you.

Fascinating. Stuff the aspects we agree on. Let's argue !

I like the second (or third) graph:

"Hey guys, check me out, I started with $5k and I've already more than doubled my BR! You guys suuuuuck!"

*2 hours later*

"So uhh.....I may have just gotten a little unlucky.....I uhh, well, hmm, my BR is kind of uh....well at least I still have my health!"




Anyway, nice graphs. It's one thing to know eventually it loses or w/e....but is something else to see what the graphs look like.

Well, I can think of a situation where martingale has value especially in vegas.

A hooker wants $500 for the night and you have $450. The odds are in your favor for getting laid tonight. But you'll still wake up with less money than you started :)

Quote: sc15

Well, I can think of a situation where martingale has value especially in vegas.

A hooker wants $500 for the night and you have $450. The odds are in your favor for getting laid tonight. But you'll still wake up with less money than you started :)

I prefer the life or death scenario. In your scenario, one could have a wafty crank and keep the money :)

Quote: OnceDear

As a little exercise, I've created an Excel workbook which illustrates dramatically how a progressive Martingale will generally give a bankroll with quite a steady initial uptrend punctuated with death defying crashes.

It allows you to experiment with waiting till xNumber of Reds appear before starting a martingale bet on the possibility of a Black to break the cycle.

It's nothing sophisticated and does not use macros, so is safe to run in excel 2010 or later.

thank you for sharing
the Marty is a fun way to play for many, for sure

oh
I used it in excel 2007
with no problems

there are also
free
add-ins that turn a spreadsheet into a simulator with no code to write too
I have one and use it sometimes

======================================
i bring this up because of your statement
"How often do you double your bankroll before crashing? About half the time !!!"

this is not even close to being a true statement
in my opinion

I get simulated about 62% to bust the $5k bank (that makes 38% success rate, far far from "About half the time")

using your sheet and another one i have this to show
no color added until feeling better


color in me other simulator

Mully

Quote: mustangsally

i bring this up because of your statement
"How often do you double your bankroll before crashing? About half the time !!!"

this is not even close to being a true statement
in my opinion

I get simulated about 62% to bust the $5k bank (that makes 38% success rate, far far from "About half the time")
Mully

This is strange indeed!
You would have better results if you bet all your bankroll in the first spin.
Maybe numbers lie after all...
Can you explain you results logically?

Quote: Kavouras

This is strange indeed!
You would have better results if you bet all your bankroll in the first spin.

yepp
or just bet $2500 at a time = 47.2992701%
or $1000 at a time = 43.2824583%

Quote: Kavouras

Maybe numbers lie after all...
Can you explain you results logically?

trying to double a $5000 bankroll $2 at a time takes time
and lots of spins in the OP example
that looks abouts likes 2500 betting units

$1000 bets = about 5 betting units
and on average abouts 25 spins - to success or ruin
of course it could take more than 25 spins - only 65% chance it happens by #25

this is where more is less
and less is better

units that be
Mully

Very good point Sally. Looking at it logically we can also say that if a system allows you to double your bankroll in more than 50% of trials then you must have a mathematical advantage at the game. Furthermore the martingale demands that we make many many small wagers: the sum of our action is significantly higher than if we bet the whole bankroll using "bold strategy". The cost of making all those small bets is a reduced probability of a large win....the house edge grinds away at our probability of success.

Quote: sc15

Well, I can think of a situation where martingale has value especially in vegas.

A hooker wants $500 for the night and you have $450. The odds are in your favor for getting laid tonight. But you'll still wake up with less money than you started :)

I'm sure shell take $50 off if youre fast enough

Quote: mustangsally

...I bring this up because of your statement
"How often do you double your bankroll before crashing? About half the time !!!"
...
I get simulated about 62% to bust the $5k bank (that makes 38% success rate, far far from "About half the time")

Thanks Sally for adding to my analysis. Your observation is interesting and seems to be correct.

I'd made the intuitive assumption that the chances of doubling bankroll would be similar to the 48.65% chance of doubling with one, bankroll sized bet. To my very rough estimations .4865 is close to a half. Can you live with that level of inaccuracy :o)

But it does transpire that if bet size is small compared to bankroll, and consequently you can take many consecutive losses before busting out, then your chances of doubling bankroll seem to be further diminished. It's not obvious to me quite why this should be and frankly I cannot be bothered to figure it out. Maybe it's the impact of greater repeated exposure to house edge, or maybe it's some consequence of the partial last bet of a losing streak.

Frankly I'm not too bothered, having proven that such a Marty is not sensible in trying to make a significant gain.

So, I'll just modify the comment in my workbook to read :-
"How often do you double your bankroll before crashing? Less than half the time !!!" and leave it at that.

To any latecomers to this topic: The charts shown are just examples created from the spreadsheet. So feel free to download and try it for yourself.

Quote: OnceDear

Quote: mustangsally

...I bring this up because of your statement
"How often do you double your bankroll before crashing? About half the time !!!"
...
I get simulated about 62% to bust the $5k bank (that makes 38% success rate, far far from "About half the time")

...It's not obvious to me quite why this should be and frankly I cannot be bothered to figure it out....

I think I've figured a simple logical explanation without complicated maths:-

With one £5000 wager, either,
Sum of winning wagers = £5000 and Sum of losing wagers = £0
or
Sum of winning wagers = £0 and Sum of losing wagers = £5000

Therefore the Sum of 'Action' on the table is (£5000+0) or (0 + £5000), which is £5000 in either case.
So, with one bankroll sized bet, Sum of Expected Loss Equals (1/37 * 5000)

BUT

With a multitude of smaller bets being placed in any other system, there will be some winners that are cancelled out with some losers as we progress to the final outcome:-

Sum of winning wagers >£5000 and sum of losing wagers >£0 if we are to successfully double
or
Sum of winning wagers >£0 and sum of losing wagers >£5000 if we are to eventually fail.

So, Sum of Expected loss will still be 1/37 * (Sum of winning wagers + Sum of losing wagers), but (Sum of winning wagers + Sum of Losing wagers) must be > 5000.

Therefore with small and occasionally cancelling wagers, Sum of Expected loss Greater than 1 / 37 * Bankroll

I cannot be bothered to quantify the difference, but I can see why it arises.

two two ?

Quote: OnceDear

So, I'll just modify the comment in my workbook to read :-
"How often do you double your bankroll before crashing? Less than half the time !!!" and leave it at that.

you could also add this to excel online (for you free) and that way a person does not need to have Excel to try it
https://office.live.com/start/default.aspx


one suggestion is to add a max table bet just to be more realistic

===============================================
also, to give Marty players even more excitement at 0Roulette
turn $5k into $7.5k, not double but just 50% increase (start bet =2)

now we are at about 55% success to hit 7.5k (one time probability that is)
now a favorite to hit the win goal target!

more winners by not being greedy
many will say

(of course, many spins on avg still required abouts 3,000 of them)

now
start bet =5
success rate abouts 57%
and only 1,161 spins on average to hit 7.5k
keeps getting better (think of the comps too)
better = more fun

time is money
Mully

http://www.toxicdrums.com/doubling-system.html
nice pics to view here too

Quote: mustangsally

two two ?

I'm sorry but I don't understand that comment. Please explain.
As to your other comments about using Marty to achieve modest objectives, you are quite correct. With the caveat that using Marty multiple times with multiple modest objectives one after another will still be as risky as one continuous session. Let us not encourage Marty exponents who think that quitting a session while ahead and coming back another day is a good idea.

Quote: OnceDear

I'm sorry but I don't understand that comment. Please explain.

i replied to your post and hit the button
and it posted twice
1 second apart
maybe me fingers are too fast

was not about you that time

Quote: OnceDear

As to your other comments about using Marty to achieve modest objectives, you are quite correct. With the caveat that using Marty multiple times with multiple modest <snip>

exactly
with a one time winning probability of X
(0<X<1),
many think it is always X no matter how many times in a row needs to be won

but if tried 2 different times, or different days for example, it is now X*X which is less than X
example
55% chance to win one time
to win the next two times in a row
0.55 * 0.55 = 30.25%
now less than 55%
but it looks right
Sally

added
just having some fun

Quote:

As a recent newbie messaged me on this concept "Don't worry my friend we will make a lot of money". i wrote this workbook to stop him taking a path to ruin and thought it worth sharing.

Some people will not listen. Oh well !!! I've done my good deed for the day.

Anyone who still thinks that variations on such a Martingale are a good idea for meaningful wealth increase (relative to existing wealth) then good luck to them. Take your system and play it big-time. Knock yourself out !

I will waste no more time explaining this.

Quote: mustangsally

===============================================
also, to give Marty players even more excitement at 0Roulette
turn $5k into $7.5k, not double but just 50% increase (start bet =2)

now we are at about 55% success to hit 7.5k (one time probability that is)
now a favorite to hit the win goal target!

Or save them the trouble: Bet half bankroll on 1-12 and half bankroll on 13-24 at the same spin.

They then stand a 24/37 = 64.86% chance of hitting their target of increasing bankroll by 50% from 5,000 to 7,500 without all that waiting and messing about.

Dead favourite. What could possibly go wrong ;o)