Martingaling - Looks so good until the inevitable

And it grows!!!

Glad to know there are a lot of others with absolutely nothing better to do when market is closed than surf the factory!

.. Well now, I wouldn't sat that. Ask Thodie. I'm a busy boy today. Just taking a 10/15 minute break here and happened to catch the show.

So, let's do this...

Set the rules!!!

lol

1. You need to have a $100 live account if you are outside of the United States. Proximus needs $200, it is of no advantage to him!
2. If you are in the United States, max account size is $2000 - 50:1
3. The account has to be a live, real money account...sorry to the cheapies...you need a brownie to get into the game!
4. This is a Martingale vs Non, competition, which means, trade however you see fit!
5. Trade lot start size = .01.
6. If you blow up, you are out!
7. Reveal your strategy as Auto or Manual...and Martin or non!
8. First person to cross the finish line, wins!

Again, if allowed...I will shoot FF a clean and crisp $100 bill, USD. They can pay the winner!

Did I miss anything??

Live fully open TE maybe?

done, TE is open, i´ll start with 100$ and leverage 1:50 so disadvantage is on my side right?

Huge disadvantage on your side, dude...

But I have no way to handicap the race...

If you are good with your odds, I certainly am not about to disagree with you....

Every race needs a longshot!

if it blows it blows
if not

If not...

Magix I love the idea of non-marti against marti... My only question would be what would the finish line be? The last account standing? or reaching a certain amount of % first?

$100.

That is the end line.

You start with $100...proxi, he was afforded the same start as already has of $200....

From there, you trade to double the account...

As things aren't exactly fair in this whole world of trading, leverage for overseas is as high as 2000:1, and for the Yanks, it is max 50:1, they can have an account start of a Max $2000.

The deal is, all start trades are max, .01, or 10 cent lots.

The idea isn't necessarily about the best trader or system, but more so about longevity on gain with the independent ideologies.

Sure, there will be those who want to gamble away...but thus the reason for a REAL account being the pre-qualifier. Throw away $100, or make $100 and take a prize of another $100..

Pretty simple really...

Fresh starts.

Only skin allowed in the game!

I already know of 1 member I don't want to compete against...and he is using our house bot!

lol

LINK

Of course...as things go...it really is a matter of I will pay $100 for somebody to prove me right or wrong....

Of course I can’t speak for everybody, but I think it’s possible that the folk who are opposed to martingale are being misunderstood here. Nobody is saying that martingale isn’t the best way to ensure a consistent income from trading. The recovery mechanism that’s built into its sizing algo guarantees that it is (and is possibly the only way to gain any kind of income, for anybody who doesn’t have a proven analysis-based edge). Ultimately trading is a matter of balancing return and risk, and the argument against martingale is grounded in the aspect of risk management.

Is there anybody here who's taken out an insurance policy on their family home? Let me say this: if you have, then mathematically, you’re extremely naive. The probability that your home will be destroyed by fire, earthquake or whatever, is extremely small; and expectancy-wise, you’re playing a losing game because insurance companies invariably make more money from collecting premiums than they ever pay out in claims. You’re playing a negative expectancy game that’s rigged against you, and the probability is massively high that, over the course of your lifetime, you will lose money.

The reason that I’ve taken the mathematically stupid option of insuring my home is because I want to offload the burden of risk. Even though the risk of losing my home may be extremely small, I have taken steps to eliminate that risk, because of the near-catastrophic replacement cost.

Like me, I expect that probably all of you who are posting here, are not trading a serious sized real money account. By ‘serious’, I mean an amount that’s in the ballpark of the value of your home, say $250,000 or more. You’re also probably not a fund manager who’s legally accountable to your clients for the loss of their money. Hence you have no need to employ the same type of risk model, no need to reduce positional risk to something like the ‘textbook’ 1%–2% per trade. For example, your risk ‘backstop’ might consist of starting with a trivial amount (say, less than $10,000), the loss of which is not catastrophic; and/or you might withdraw your starting amount at some point, so that subsequent risk is limited to what you choose to see as being the ‘market’s money’. In this case, I totally agree that any kind of risk management you might deem as being personally acceptable is perfectly valid. Secondly, if a margin call can easily be shrugged off, then chances are that, if it does occur, you’ll have the necessary capital to re-fund the account, and start over. And thirdly, from an account building standpoint, the fact that martingale ensures consistent income continues to hold good.

This quote raises an interesting point. Let’s assume that the probability of the ‘death trade’ — the event, or sequence of events, that causes irretrievable drawdown (aka ‘ruin’) — is extremely small, perhaps infinitesimally so. Then statistically, you need an extremely large, or ‘infinitesimally’ large, data sample, to make it possible to obtain any statistically meaningful conclusions. Hence I would suggest that the math argues that a run of 5 years without a death trade, or 10 years, or even the length of time that forex has been around, is nowhere near enough to convey anything meaningful about whether a certain martingale is likely to succeed, or fail, within a certain time period (e.g. a trader’s lifetime). The statistical error involved is prohibitively high.

To suggest that any kind of test result is meaningful, then, is perhaps a bit like saying “I’ve been driving for 20 years, and I’ve never died in a car accident, therefore it will never happen to me”. (I think the analogy with trading is reasonable, because your ability to avoid death comes partly down to your driving skill, and partly down to whatever the road throws at you in the way of unsafe oncoming drivers. So it is with the market).

If (and I stress IF) price movement represented a completely random walk, then it’s mathematically true to say that, if you trade for long enough, you’ll eventually encounter every conceivable price pattern, and therefore including the pattern that causes the ‘death trade’ (and the longer you trade, the greater the risk). Having said that, I believe that price movement is not completely random, in that there are certain ‘truisms’ that we can apply; two of these that spring immediately to mind are (1) that price never trends in one direction forever, and (2) that price always follows a pattern of accumulation --> breakout --> trend --> accumulation --> breakout --> trend etc. Maxims like these have invariably held good in all markets, and can potentially provide an edge. Having said that, there are many different ways that these truisms can play themselves out (depending on current orderflow, market participants, heavyweight agendas etc etc). So, given this uncertainty, while I agree that techniques like averaging down help by alleviating the need to make accurate entries, I also believe that the same uncertainty makes it impossible to make statements, with 100% certainty, like:
i) a certain martingale will avoid the death trade indefinitely;
ii) price will absolutely never behave in a certain way (e.g. there will never be an X pip move without a Y pip retracement somewhere along the way);
iii) it’s impossible (for a given system) to have X consecutive losing trades.


My objections to martingale are definitely not, as ZebraSquirl suggests, that I’ve worked for a brokerage. They are purely mathematical, and I will explain them briefly below. (When I say ‘martingale’ I mean doubling without limit; of course I realize that there are less severe progressions, and ways of capping loss, etc etc).

1. Martingale is a form of MM, and no MM can alter expectancy. PipMeUp has posted a neat mathematical proof here, and over the years I’ve posted spreadsheets that allow the testing of different staking systems: martingale, anti-martingale, d’Alembert, Labouchere, Guetting, Oscar’s Grind, etc etc, and, if applied over every possible combination of outcomes (assuming that each outcome is equally likely), all of these systems always deliver an expectancy of exactly zero. In other words, they merely redistribute wins and losses, nothing more.

Increasing stake size simply to recover recent losses has no mathematical basis. However, increasing stake size because probabilities have meanwhile improved (e.g. by using analysis, or some market truism, that provides an edge) augments that edge. (That is how blackjack pros beat the casino: larger bets when the cards remaining in the deck give the player a positive expectancy over the game rules).

[I expect that somebody will reply, saying that trading and casino math are different. Not so: all that trading effectively does, is generate a distribution of wins and losses, and MM is the science that examines the application of staking systems (position sizing) to this distribution.]

2. Increasing stake (i.e. position) size increases risk of ruin EXPONENTIALLY. For example, if you increase your position risk from 1% to 2% per trade, then your risk of ruin is significantly more than doubled. Hence, everything else being equal, a trader who risks (say) a consistent 3% per trade is massively more likely to survive than a trader who risks 1%, 2%, 4%, 8% etc, because his average position size will almost certainly be lower.

[Note: As to whose return is greater (assuming both traders have the same win% and RR), the first trader will win 3x the amount (because he is risking 3%, while the martingaler is always recovering 1%), but the martingaler will always return to a 1% win, while the first trader will incur some 3% losses).]

3. If you have a proven edge (e.g. through analysis, or by exploiting a market truism), then there is no need to take on the significantly greater risk of ruin that progressive sizing creates. You can simply and safely risk a consistent X% per trade, and there is no need for ‘recovery’ trading, because over the long term your wins will outweigh your losses, as the edge plays itself out.

In the Market Wizard books, one overriding theme that comes through is that these supertraders prioritize risk management over income. Again, these are traders with seriously large account sizes, and consequently, none of them take the “all or nothing” type of risk that an amateur, who can afford to try some “$500 to a million in 12 months” strategy because the loss of $500 is inconsequential, and indeed replaceable. The latter paradigm justifies the use of a completely different risk model.

4. Market probabilities don't increase commensurately with martingale sizing: Ideally, the position size should reflect the improved probability of success in the outcome. After sizes have been doubled few times, e.g. 1, 2, 4, 8, 16, 32 — the 6th trade in the progression needs be have something approaching 32x the expectancy of the 1st trade, to justify the 32x increase. I highly doubt that this is likely to be true for most martingales. In other words, while the improved probability might warrant some kind of increase in position size, the probability of increased return is not great enough to justify the increase in risk. The focus should be on the probabilities being offered, not merely the need to recover recent losses, which is effectively asking the market to comply with the trader's P/L requirements.


Finally, some points on martingale variants. Every Martingale variant aimed at mitigating risk incorporates a tradeoff that has a commensurate downside:

i) Using a less severe sizing progression (than doubling) cripples the recovery rate, i.e. necessitates >1 winning trade to return to breakeven.

ii) Limiting the number of progression steps does likewise (e.g. if we cap the number of steps at 5, then it takes 31 wins @ 1 unit, instead of just 1 win @ 32 units, to recover from 5 consecutive losses).

iii) Starting the progression with a tiny position size reduces return in exactly the same proportion.

iv) Withdrawing money as insurance against the death trade reduces the funds needed to accommodate a given number of progression steps (and of course there's no guarantee that the death trade won't occur before the withdrawal). And if position size is based on account equity, then it will lower future returns also.


FWIW, that's my summary on how I see the pros and cons of martingale. There is no return without some kind of risk, and all trading systems represent an attempt to skew the risk/return balance in a favorable way, that the trader feels personally comfortable with. Yes, martingale offers the guarantee of a more consistent income than ‘textbook’ trading. Whether you’re willing to take on the risk of a margin call is likely going to depend on your starting account size, and your personal appetite for risk.

My 1+2+4+8+16+....... cents,
David

TL;DR!

So...

Are you in or out?

Actually i did, i know a strategy that makes about 50%/month and its guaranteed profit, i traded just that a couple weeks ago, but i became greedy and tried to imporove it to make 200-300%, and then it backfired on me.

I just cannot accept the fact that strategies are not scalable, if it can make consistently 1%/month then it can make 1000% in the same fashion by either increasing the risk or the trade count.

I just gave up full technical systems and gambling ones, i dont think the long term bias can be understood without fundamentals and long term sentiment.

And as for gambling, well, i havent seen a roulette player getting billionaire, so probably you wont be either with your shitty martingale strategy...

Actually meant to poke Proximus' hive a little bit with that comment ... (well, actually, with the entire posting). Seems like he's got a chip on the old shoulder ... (sorry, Prox, just calling it as I see it.)

In any event, always like good analysis to back up a point of view. At least you're not talking out of your butt, which is appreciated. (Unless you want to do that, which would also be an accomplishment.)

In the meantime, I intend to enjoy the fireworks in the head-to-head throw down that is going to ensue!!! (I would partake, but my excuse is that I'm using a Mac; got MT5 set up using PlayOnMac and Quartz, but it's glitchy ...).

I am personally looking for blood, gore, decapitations, and amputated limbs ... !!!

My bet, though, is on the most well-coded EA. I can't imagine using this method without 24-hour eyes on the trades, since TP's will have to be adjusted as positions are added to.

Sure you don't want to charge to watch?

It's weird, I almost don't care about the merits of the system anymore ... . I don't envision using it myself for a number of reasons ... .

To me, it is now all about just seeing how long a person can get the system to go, modified or otherwise ... .

It's like a solar panel car race ... . There are likely to never be any solar powered cars on the streets of our cities ... . Yet, every year, various college and university engineering programs design these cars and race them just to see how far and fast they can get them to go ... .

(Insert soaringly inspirational orchestral music here).

If a person is able to withdraw a certain amount of money everytime the equity reach a certain level, sooner or later the account will be free funded. account started with $100, reaches $200, withdraw $50. reach $300, withdraw another $100. even if after that the account is blowed or margincall, this guy is still profit for $50.

To be able to martingale successfully, one must know when to withdraw money consistently..

100 is not enough to trade martingale there big shot. 10k is the minimum given the 0.01 minimum trade size allowed from brokers. However I will accept your donation that I will not lose 10k on any account, real demo , 10k, 100k, 1mil with martingale strategy. But will see steady increase like typical martingale of +/- 1000/month.