Statistics of a anti-grid like system
You should first find a way to plot the floating equity, the strategy tester report is meaningless for these types of strategies that always close all positions at once (it only plots a datapoint when a trade is closed, not when its opened).
This little tool will help you: https://www.mql5.com/en/code/9343
or here latest version: http://sites.google.com/site/prof7bit/offline_charts-mqh
(include the include file and then just call recordEquity() at the very beginning of the start function on every tick). The resulting equity chart in the offline chart folder after a test run will then clearly show you how dangerous your EA really is and how strong your nerves need to be to trade it.
> are the results useful for a anti-grid like system and what are the key factors of such systems
I guess this is a 'reinforce success' system?
So I imagine the equity line in real-time trading would be somewhat... alarming?
IMHO this type of system will need a moderate to good degree of deliberate market behaviour, i.e. sustained trends &/or wide enough ranges to recover losses on a retrace
This has been just possible within the dates you test above - I'm fairly certain that years like the prior year (2009) would be much less successful
Great topic - keep it going!
-BB-
Hi, not sure what a reinforce success system is.
IMHO this type of system will need a moderate to good degree of deliberate market behaviour, i.e. sustained trends &/or wide enough ranges to recover losses on a retrace
The big danger in this system are ranges larger then the gridsize but not twice as large. In such phases your are buying high and selling low. Fortunately after a range a breakout or trend should happen. In this case the system is regaining very fast. I think this will be save in every trending markets. It does not matter how fast the trending goes. (Of course as faster the trend as more sessions gets opend/closed. But in general retracements are not a problem. (Maybe by introducing a more dynamic grid also this issue can be minimized, of course it will still be present.)
For example if i divide the gridsize by 2 the EA passes the current 'danger zones' very nicely. (Of course it will create new danger zones)
This has been just possible within the dates you test above - I'm fairly certain that years like the prior year (2009) would be much less successful
I have only tickdate starting from 2010 since (and you are probably right with that) the market conditions are quite different the last two years and i currently would like to develop strategys based on current market conditions. This will be a later step.
So I imagine the equity line in real-time trading would be somewhat... alarming?
I have included 7Bit's equity logger and the maximal drawdown was still 'only' 28.218%. So it might be worth to continue the research on such systems
Equity graph:
Addon:
System stays profitable if i close all trades once 10% drawdown is reached.
Hmm, looks quite good actually. As I suspected, the mt4-report relative drawdown is fairly accurate. I wounder if 7bit still uses his anti-grid system. You've spark my interest in anti-grids again. I'll see if i can duplicate these results. I like the curve of the 10% lets just call it stop-loss to butter-up those who don't like no-stoploss systems ;)
Ps: if you're gonna test this on another year, I'll recommend 2008. That damm year murdered one of my almost-exciting system today. I nick name it the tsunami range market.
> would like to develop strategys based on current market conditions...
Thats fine, thats all you can do - with a bit of projection perhaps on, e.g. long-term ATR values...?
Just be prepared for when the current conditions change... What will you decide on? Other than mounting losses :(
> 2008 - That damn year murdered one of my almost-exciting system today. I nick name it the tsunami range market.
Yes, four quite different quarters for behaviour... I'd guess this anti-grid would lose in Q1 & Q2 but recover in Q3 and Q4
-BB-
@ubzen
2008 on Metaquotes data:
beside two dramatic spikes the system behaved nicely. "Stoploss mode" is also running fine on 2008
even 2007 (one year which my EA's in general do not like) was fine. 2007 till now:
Stoploss mode was quite a good idea. It keeps the strategy from hitting the MaxTradelimit and generates a restart.
@BB:
Yeah, when to quit such an EA is a different task. Long term ATR values could be quite intersting. I definately will continue the work on this one.
But to keep the topic a bit On-Topic.
Since such an EA in theory can win in every trending market as long as there is enough capital (Way less then a simple martingale), especially if different symbols/stocks are used to diversify the risk of falling into a ranging trap why are such/similar systems are not that popular? My assumtion is due the amount of capital needed.
Very interesting analysis! I've been thinking a lot about the changing character of the market. The discussion reminds me of what I have been thinking about a lot - I think a very useful thing to achieve would be an algorithm that predicted with reasonable accuracy the character of the market for the near future (eg trending or not) and the selecting methods appropriate to that character (or simply just ignoring the market if it was predicted to be non-trending).
@zzuegg, the graphs look very impressive, but I am not sure you are exploiting your edge in the best way. First a question:
How does the margin you need escalate when you run into an unfavourable time? (you say it is not as bad as so-called martingale systems where, in their simplest form, margin doubles every time you lose a trade, but you do not specify how it does increase. Or how much capital you need)
I am not sure how familiar you are with the statistical theory of trading, so excuse me for repeating some theory.
It is a mathematical fact that the way to achieve the best long term performance (with a reasonable definition of this including your risk-aversion) from a series of trading opportunities with similar statistical characteristics is to use a similar amount of leverage for each of them. A reason so-called martingale money management is basically bad is because it diverges wildly from this principle. Crucially, in "martingale" systems, you are generally either using tiny leverage for some trades (and failing to exploit them effectively) or way more than the Kelly criterion would give you as an absolute maximum position size on other trades, and typically both on different trades. The problem with this is that those rare trades have a dramatic effect on the long term equity statistics (i.e. the probability distribution of equity at some chosen time in the future) and typically would cause complete loss with probability 1 in the long term. Admittedly, there are probably quite a few people who are willing to accept (or ignore) a small probability of losing very heavily in order to have a high probability of healthy profits, but this is best classified as pure gambling.
A trivial observation is that no trading system can ever be improved by including a trading opportunity which has a negative or zero expectation. This might sound ridiculously obvious, but is ignored by a lot of martingale enthusiasts. Another observation is that if two trades have independent statistical characteristics, there is no theoretical reason to allow what happens to the first trade to influence the size of the position for the second trade, (except that it may be changed in proportion to the equity in the account). There is one circumstance (which does happen in some situations) which justifies martingale like staking. This is where you have a series of opportunities where as more of them fail, the later ones have higher and higher expectation (strictly speaking higher and higher ratio of expectation to standard deviation).
Your system uses a genuine profitable characteristic of the market - that it tends to trend. Do you agree that if the opportunities you are taking have positive expectation, you should in principle be treating all of the opportunities independently, trading each of them independently with a sensible position size (proportional to the equity in the account and using leverage depending on the estimated probabilities of different outcomes for the trade) and as a result getting a better probability distibution of the equity in your account at some point in the future?
I should say that if I had looked at your graphs and not, as a mathematician, studied the statistical side of trading, I would not have ever thought like this - the graphs look very appealing, except for occasional craziness (which unfortunately may be very important to the long term expectations). But mathematical truth cannot really be ignored.
@zzuegg: everything looks good. I'll strongly recommend learning mt5 to test you multi-currency approach. I like monte-carlo type statistical analysis and recommend this tool. You'll need to sign_up with them and request a free license code. My license code just expired, hopefully they'll renew it.
@Elroch: nice to have a mathematician/statistician in the house. I don't wanna take the thread off topic. But i agree and dis-agree with some of your findings.
I Disagree: It is a mathematical fact that the way to achieve the best long term performance (with a reasonable definition of this including your risk-aversion) from a series of trading opportunities with similar statistical characteristics is to use a similar amount of leverage for each of them:
I Disagree: Admittedly, there are probably quite a few people who are willing to accept (or ignore) a small probability of losing very heavily in order to have a high probability of healthy profits, but this is best classified as pure gambling.
I Agree: A trivial observation is that no trading system can ever be improved by including a trading opportunity which has a negative or zero expectation.
I Disagree: there is no theoretical reason to allow what happens to the first trade to influence the size of the position for the second trade, (except that it may be changed in proportion to the equity in the account).
Not directed to me, but I'll attempt to answer.
How does the margin you need escalate when you run into an unfavourable time? The used margin stays the same, it's the free margin which fluctuate. Here are some resources on Anti-Grid.
About Anti-Grid: 7bit website explaining what it is.
Calculations: Here 7bit and Gordon tries to prove/disprove mathematically. Perhaps you can give your 2-cents.
MoreInfo: Discussion with allot of inputs from different people.
Oh and as far as the reasons why I Disagree. Mainly because I've concluded that Price Charts are random 90% of the time and contains Uncontrollable Risks. Those who don't want to accept this risk would be better off talking theory instead of trading. Here a quote from a book I just picked up. Something like: If you can only die by being struck by lighting, you'll eventually die by being struck by lightning.
I think the key to all the statistics is measuring if you're being struck by Lightning OR Rain. If someone don't even like a remote chance of losing then better not trade.
Hi,
Elroch, it is nice to have some mathementician on board, even i passed the advanced statistics course, i am not that good at.
Crucially, in "martingale" systems, you are generally either using tiny leverage for some trades (and failing to exploit them effectively) or way more than the Kelly criterion would give you as an absolute maximum position size on other trades, and typically both on different trades.
The main problem for me in margingale systems is that your used leverages (and the risked amount of money for each trades) is rapidly increasing while your profit target is constant low. This of course give you a negative expectation in the long run. (If you are trading randomly)-
The problem i have in analysing my current system is that if you only look at the net leverage it is always nicely low and gets only increased if previous orders are already in profit. In theory my system will automatically adjust the used leverage for each direction in a way that it will be higher in the current direction. No matter what direction. This is the main point of this strategy because i don't care where the market is going, all indicators are lagging or giving fail signals, price does not. From my side summing up indicators with negative expectations does never give you an edge in the long run. Purely statistical systems usually look very good on the history, since they are by definition optimized on that one.
Your system uses a genuine profitable characteristic of the market - that it tends to trend. Do you agree that if the opportunities you are taking have positive expectation, you should in principle be treating all of the opportunities independently, trading each of them independently with a sensible position size (proportional to the equity in the account and using leverage depending on the estimated probabilities of different outcomes for the trade) and as a result getting a better probability distibution of the equity in your account at some point in the future?
Why should i, i don't know when a trend starts/ends and i have no way to find out. If price moves up, i increase the long leverage, if price goes down short leverage is increased. Net leverage/position size will be adjusted in my direction.
How does the margin you need escalate when you run into an unfavourable time? (you say it is not as bad as so-called martingale systems where, in their simplest form, margin doubles every time you lose a trade, but you do not specify how it does increase. Or how much capital you need)
Currently backtests show that with a 20k standart or 2k micro account you will be fine. Also if my floating loss gets higher then 10% i also close everything and give it a fresh restart. The crazy balance chops happen in ranging phase with very specific size. (This is a point which needs to be improoved). The margin goes up, but the current risk i put in one direction is not. (which separates this system from the simple betting strategys i think)
Looking forward to hear you comment on this.
Best regards (time to work now :( )
//z
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hi fellow traders,
One of my current projects has reached a state where i don't know how to progress.
I simply don't know how to categorize an anti-grid like system. The system uses no indicator, and is based on the assumption that the market will not be sideways forever. The overall target is reached by adjusting the 'net lotsize' trough adding orders. Net lotsize will be adjusted in small steps in the current market directions. Current direction is based on price action. Profit faktor will be low of course but in this case i assume this does not matter. Also the drawdown is most likely to be shown wrong (the position gets closed one by one and the balance curve goes zigzag).
Since the strategy can be used on any pair my plan is to diversify risk troughout the majors. This will allow me to reduce the lotsize by 4/5th while keeping the global target (which is currently a workaround which simulates manual closing) to the same level.
Now my question:
are the results usefull for a anti-grid like system and what are the keyfactors of such systems?
Best regards
//z