Hanover replies to PMs | Page 28 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Abokwaik replies to PMs 96 replies

Hanover - can you help me with MT4 qs please? 4 replies

Attachments: Hanover replies to PMs
Exit Attachments

Hanover replies to PMs

  • #541
  • Nov 4, 2016 11:53am Nov 4, 2016 11:53am
Quoting Toranaga
Question: Did you ever tried grid trading ??
No, I have never cared for random entries or potentially unlimited risk (assuming thats what you mean by a "grid").
There are more reliable, and safer, ways to approach trading: for example, start with the concepts in myth #22 here.
  • #542
  • Dec 8, 2016 2:30pm Dec 8, 2016 2:30pm
Hey David,

I see that you're no longer coding indis/EAs unless they're useful to you personally.
I have an indi that I'm wanting coded (will pay) and from what I've seen you're one of the best coders around here. Where would be the appropriate place to message you to see if it would be something you might personally find useful, and if you're not interested in coding it, any good coder(s) you would recommend. (I don't see a message button on your profile). Cheers!
  • #543
  • Dec 8, 2016 6:56pm Dec 8, 2016 6:56pm
Quoting acdfx
Hey David, I see that you're no longer coding indis/EAs unless they're useful to you personally. I have an indi that I'm wanting coded (will pay) and from what I've seen you're one of the best coders around here. Where would be the appropriate place to message you to see if it would be something you might personally find useful, and if you're not interested in coding it, any good coder(s) you would recommend. (I don't see a message button on your profile). Cheers!
Hi,

Thanks for your message.

Personally, I find the most useful indicators are those that provide info like news events, time zones, summary data like daily ranges, and some horizontal levels. I turned messaging off a long time ago because I was getting too many coding requests. I've been trying to focus more on trading and less on programming.

For some good programmers, see this post. There are other good coders who post at FF, but I'm not sure if they are willing to work for money.

I assume you're aware that MT4 will be discontinued soon (more info in this thread).

David
  • #544
  • Dec 9, 2016 2:42am Dec 9, 2016 2:42am
Hi Hanover,

I am new in forex, but have some experience with programming. I am not a programmer, it was/is my hobby. 3 years php/javascript, half year Csharp.

As mql4 is going to be the past, what do yourecommend, mql5 or Csharp with ctrader?
Only my scepticism keep me from being an atheist. Be sceptic ever.
  • #545
  • Dec 9, 2016 10:18am Dec 9, 2016 10:18am
Quote
Hi,

Thanks for your message.

Personally, I find the most useful indicators are those that provide info like news events, time zones, summary data like daily ranges, and some horizontal levels. I turned messaging off a long time ago because I was getting too many coding requests. I've been trying to focus more on trading and less on programming.

For some good programmers, see this post. There are other good coders who post at FF, but I'm not sure if they are willing...

No I had no idea about MT4 being discontinued, that's terrible
How much longer do you think we'll be able to use it before any changes?

The indi I'm wanting would list all pairs and be similar to your PSA (which I currently use), but instead of measuring movement it would list the pips from daily high/low, time since last daily high/low (in hours:minutes:seconds), as well as the pips from/time from other key horizontal levels.
If you're not interested (which is understandable due to the MT4 issue), I'll try contacting those coders about if they can do something forward-compatible. Thanks for your reply!
  • #546
  • Dec 9, 2016 5:20pm Dec 9, 2016 5:20pm
Quoting acdfx
If you're not interested (which is understandable due to the MT4 issue), I'll try contacting those coders about if they can do something forward-compatible.
I would definitely contact those coders, especially as they will be more familiar with the transition from MT4 to MT5 than I am.
Quoting sinusgamma
I am new in forex, but have some experience with programming. I am not a programmer, it was/is my hobby. 3 years php/javascript, half year Csharp. As mql4 is going to be the past, what do yourecommend, mql5 or Csharp with ctrader?
I'm sorry, but I have no experience with either MT5 or cTrader.

As a programmer, I would expect a transition from MQL4 to MQL5 to be smoother than MQL4 to cTrader, but if you are starting to write indicators/EAs from scratch and you have C# experience, then perhaps cTrader would be more logical. However, even more importantly, I would first try both platforms and see which one you prefer, with regard to features that are useful to you, operational ease-of-use, etc.

As a trader, my priority would be to find the most trustworthy br0ker with the lowest transaction costs (spread, commissions etc), and then select from the platforms/programming facilities they are offering. Most folk who are new to trading greatly underestimate the account bleed caused by repeated transaction costs.
  • #547
  • Dec 22, 2016 4:40am Dec 22, 2016 4:40am
Quoting LearnAlways
Questions about the 2 traders which u mentioned using TA and another 1 trader using FA......
Don't know if you're still around, but I just noticed that I hadn't answered your questions. Not sure how I failed to see your post. My apologies.

1) TA - What type of trader are they? - Scapler, Day, Swing, Long Term
All are predominantly intraday traders, as far as I know. One trades all TFs from M15 to weekly.

2) TA - How do they trade? - Naked PA or Indicators?
3) TA - If Indicators - What do they use? - S/R, S/D, Pivot, Volume, Fibonacci, Harmonics, Moving Average, Breakout?
One uses Bollinger bands, candle patterns, stochastic divergence and some S/R (including pivot levels and fibos). Trades all TFs from M15 to weekly.
The other uses range bars, Guppy MMAs and a proprietary signal generating indicator that he developed himself.

4) FA - Short term news spike or Long Term Fundamental?
Both. He (Jarratt Davis) trades into and out of news announcements. He also uses macroeconomics (predominantly central bank minutes/announcements and interest rates/CPI/QE data) in both currencies to establish pairs to trade and holds trades of anything between 20 to upward of 200 pips.

5) MM - How much risk do the traders take?
No more than 1% equity per trade, in all cases. Jarratt occasionally risks 1% equity per 100 pips, rather than 1% equity per SL pips.

6) TF - What TF are they looking at?
See above.

I've disabled PMs because I get too many programming requests.
  • #548
  • Jan 2, 2017 1:55am Jan 2, 2017 1:55am
Happy New Year everybody.

Attached is a ZIP containing a comma separated (CSV) file of all (more than 44,000) FF calendar news announcements from Jan 1, 2007 thru Dec 31, 2016, timestamped New York EST (or DST when applicable). The CSV file can be imported to other platforms (e.g. Excel) for further analysis.

Columns (from left to right) are:
Date, Time, Currency, Impact, News description, Actual, Forecast, Previous, Revised from, FF event ID
Attached File(s)
File Type: zip News events 2007-01-01 thru 2016-12-31 timestamped New York time.zip   729 KB | 354 downloads
  • #549
  • Edited 11:40pm Jan 2, 2017 2:55pm | Edited 11:40pm
Quoting javad.111
I use iCustom but it return 2147483647 (empty value) Can you please take a look ? For example its a simple supertrend in up trend it third index shows zero (wrong) ,,, down trend show 2147483647
1. Supertrend indicator compiles with errors. It has an unmatched right curly brace }
See screenshot below

2. Your iCustom() statement is wrongly constructed. Try something like:
iCustom (Symbol(), Period(), "xSuperTrend", "x", "x", 10, 1.7, 0, 0);

3. Supertrend indicator uses 5 buffers, but only 2 of them (buffer# 0, and one of the others) have values for any given datapoint. Hence you should use only buffer# 0 for extracting the indicator's value.

Sorry, but you are on your own after this. I am no longer coding or troubleshooting. You could try posting in one of the "I code for free" threads, or start a new thread. Hopefully somebody who is more generous than I am will help you.
Attached Image (click to enlarge)
Click to Enlarge Name: PPI 130.png Size: 29 KB
  • #550
  • Edited 11:49am Jan 13, 2017 12:50am | Edited 11:49am
Quoting Mingary
If one looks at the Forex Retail trading situation objectively (objective reality as you put it) in terms of the performance of all customer funds then the conclusion will be that a consistent and prolonged success is indeed a statistical aberration. I am not saying it can't happen. I am saying that it must happen at the extremes and the data shows this ! So, yes in the long run only brokers and system vendors make money in Retail Forex.
I'm replying here, so as not to further disrupt the other thread.

Let's be clear.

Everybody knows that (from a retail trader's perspective) costs make trading a negative sum game, and that the 'average' retail trader eventually loses. There have been lengthy (and heated) debates as to exactly what the failure rate is; supposedly it's somewhere between 90% and 99.9%

Hence a trader who achieves prolonged success is indeed a "statistical aberration". No argument there. The issue is whether those who do achieve prolonged success do so because they are genuinely proficient, or due to mere 'luck'. I contend that any fair minded person would see 26 consecutive profitable months as adequate proof of proficiency. That being so, it is possible for the retail trader to profit systematically (as opposed to fortuitously) from forex.

Quoting edge540
Whoooopitee doooo,what an incredible investment.You'd be better off working at Maca's and not drinking in the weekends.
You completely miss the point. You said that only br0kers and vendors make money from forex (and without any evidence to support your claim). I've provided evidence that it's possible for a retail trader to profit systematically, with 26 consecutive winning months. The gent who is managing my account is making upwards of seven figures annually. As are some of his bigger clients, and without having to work at all. QED.
  • #551
  • Jan 13, 2017 1:13am Jan 13, 2017 1:13am
Quoting hanover
{quote} Don't know if you're still around, but I just noticed that I hadn't answered your questions. Not sure how I failed to see your post. My apologies. He (Jarratt Davis) trades into and out of news announcements. He also uses macroeconomics (predominantly central bank minutes/announcements and interest rates/CPI/QE data) in both currencies to establish pairs to trade and holds trades of anything between 20 to upward of 200 pips.

Hanover,

The few times I followed Jarratt Davis his calls were unsuccessfully. Compared to David Pegler who was a trader from FXbootcamp and was deadly accurate (entry, stop and takeprofit levels), but he no longer teaches.

Never saw actual accts of either trader. The days I trained with David he was very accurate but at the time I was new to trading and did not understand how he made his calls.

But Jarrat has not impressed, maybe i need to observe him some more.

Did u see him make accurate calls ?
  • #552
  • Edited 11:48am Jan 13, 2017 1:48am | Edited 11:48am
Quoting fxworldwide
The few times I followed Jarratt Davis his calls were unsuccessfully. Compared to David Pegler who was a trader from FXbootcamp and was deadly accurate (entry, stop and takeprofit levels), but he no longer teaches. Never saw actual accts of either trader. The days I trained with David he was very accurate but at the time I was new to trading and did not understand how he made his calls. But Jarrat has not impressed, maybe i need to observe him some more. Did u see him make accurate calls ?
The fund manager I've been referring to is not Jarratt Davis. As far as I know, you need to be a very big fish if you want Jarratt's company to trade for you.

However, since you ask about Jarratt: During the time I've been with him, he's made some winning calls, some losing. I believe I've received some value for my membership because I've gained some new perspectives on FA. As I've said elsewhere, I don't believe in 'hand-held' mentorship, at least in the sense of mindlessly copying a more experienced trader's calls. I prefer to accumulate info from a wide variety of sources, and then devise my own trading ideas.

Jarratt's being ranked the world's second best forex trader between 2008-2013 by Barclay Hedge is genuine (I can send you a screenshot if you e-mail me privately).

More recently I signed up for free e-mails from BKforex (Boris Schlossberg and Kathy Lien), and I believe that their analysis is pretty good also. They are amongst the FA-based links I posted here.

I have no experience with David Pegler, hence I can't comment.

About trading calls: As we both know, you don't need to make perfect calls, a 60% win rate at 1:1 RR (for example) represents an edge. IMHO it's not so much about the size of the edge, it's about how consistently it can be maintained over several years' worth of changing markets. Consistency allows you to safely trade a higher VAR (or invest a larger sum of money with some confidence -- however, in terms of a balanced portfolio, I would still regard forex as one of the more risky types of investment).

David
  • #553
  • Jan 27, 2017 3:32pm Jan 27, 2017 3:32pm
Quoting mrhigurashi
Just noticed you write your from the shire. Hows it over that side of the ditch?
All good here in middle earth, many thanks! Although NZ's South Island has been ravaged by a number of nasty 'quakes in the last few years. Fortunately I live hundreds of kilometres away from the 'quake centers, haven't been hit yet.
  • #554
  • Feb 9, 2017 3:02am Feb 9, 2017 3:02am
Hi Hanover. Thanks for your great work. Sir, I have a problem with you swing indicator from this thread: http://www.forexfactory.com/showthread.php?t=411874 this indicator doesn't update in real time. So I find it very useful for me, but "her" little problem don't let me work. Can you fix it?

Thanks
Best regards
  • #555
  • Feb 9, 2017 4:53am Feb 9, 2017 4:53am
Quoting Joakim
I have a problem with you swing indicator from this thread.........Can you fix it? Thanks Best regards
Wow, this was 4 years ago, I don't remember why I wrote the indicator, or what it was supposed to do.

Sorry, but I'm trying to fix problems in my own EAs at the moment, don't have much time to help anybody else. I've provided the MQ4 file here if anybody wants to modify the indicator.
  • #556
  • Edited 3:05am Feb 23, 2017 2:52am | Edited 3:05am
Quoting alphaomega
Look at this chart very carefully.....
I've just encountered what IMHO is a 'must read' post by alphaomega. It's along the lines what I was trying to say here about statistical fluctuation, but he presents it so much better.

It illustrates why seeing a positive, even a highly positive, trade explorer or myfxbook is not enough. It is also important to know WHAT one's edge is and WHY it works, i.e. the strengths and weaknesses of one's strategy, under what market conditions (e.g. trending, sideways, volatile, flat, etc) it performs profitably, and under what conditions it struggles. Failing that type of knowledge, I'd rather see a strategy that's profitable after 200 trades made over 5 years -- i.e. across several months worth of potentially changing market types -- than 1,000 trades over 1 year. As alphaomega's post implies, anything less than statistical validity amounts to a crapshoot.

Having said all that, I'm still betting that my manager friend has a genuine edge. 26 consecutive profitable months defies odds of several million to one, it seems beyond the realm of 'luck'. But even if I'm wrong, I'm letting him play with money that I can afford to lose.

A coin toss game will generate 'trends' of heads or tails, but these trends have no predictive value: the outcome of the next toss is always 50/50. When markets trend, it's helpful to have an idea as to how sustainable the trend is, i.e. whether it has the likely backing of fundamental (or other) factors, or it merely 'coin toss' random (hence the 'saddle blanket' analogy). To whatever extent it's not possible for a retail trader to make this distinction, a prediction-based edge is likewise impossible.

Another important consideration is TooSlow's comment:
Quoting TooSlow
The market doesn't care about the reason you entered. When you win, it is because other people came into the market and made the market move in your direction. When you lose, it is because other people came into the market and made the market move against your direction.
The market is a mix of anonymous participants whose agendas are unknown to the retail trader. Markets don't move because of analysis, rather moves are caused by mismatches in volumes of executed orders, across whatever liquidity happens to be available. A bank or institution can suddenly place a huge order of several yards at any time, causing a reversal that no TA-based system can possibly predict. Likewise if a central bank makes an unscheduled high impact announcement, effectively causing a liquidity vacuum that results from all 'smart money' instantaneously rushing to exploit the announcement, creating a gap of maybe hundreds of pips.

Lest I be accused of excessive pessimism, and despite my own personal preference for FA over TA, I need look no further than my manager friend to know that, in suitably experienced hands, TA -- despite its limitations -- plus some careful risk management can nonetheless provide what has thus far proven to be a solid result. While I don't believe in a holy grail, I'm betting that the naysayers are wrong: that however difficult achieving systematic profit might be, it is nonetheless possible.
  • #557
  • Feb 23, 2017 6:58am Feb 23, 2017 6:58am
Quoting hanover
Another important consideration is TooSlow's comment: {quote}The market is a mix of anonymous participants whose agendas are unknown to the retail trader. Markets don't move because of analysis, rather moves are caused by mismatches in volumes of executed orders, across whatever liquidity happens to be available. A bank or institution can suddenly place a huge order of several yards at any time, causing a reversal that no TA-based system can possibly predict. Likewise if a central bank makes an unscheduled high impact announcement,[/highlight]...
Aren't we saying the same thing?
My Threads: Trading is as simple as 1-2-3, Highest Open / Lowest Open Trade
  • #558
  • Feb 23, 2017 9:53am Feb 23, 2017 9:53am
Quoting TooSlow
Aren't we saying the same thing?
Yes. I was attempting to elaborate on what you were saying. IMHO it's an important point, that only a few people seem to understand. Some folk apparently believe that the market should move in a certain way because they (for example) placed a trendline on their chart. Topherhk88 also explains it well here.
  • #559
  • Mar 3, 2017 9:20am Mar 3, 2017 9:20am
Quoting hanover
{quote} Yes. I was attempting to elaborate on what you were saying. IMHO it's an important point, that only a few people seem to understand. Some folk apparently believe that the market should move in a certain way because they (for example) placed a trendline on their chart. Topherhk88 also explains it well here.

Hi Hanover,

I am starting to code my trading robot EA. Though I am not a professional programmer and only learn things on the fly, my programming experience teaches me that exception handling is extremely important. I anticipate to be able to thoroughly test an EA and understand all the idiosyncrasy errors that one may encounter in reality would take tremendous amount of time and cost.

I have followed some of your posts and learned that you are a professional programmer and have participated in several MT4 project. I am wondering if you would, given your solid experience in developing EA, provide a list of possible errors that one may encounter in practice? This would be a huge saving in development/testing time and cost for me. Because otherwise I would have to wait for the errors to happen in realtime and only then start to work on how to handle it.

For example, I believe the most common one would be sudden jump in price. Or platform freeze. Or unexpected closing of position. And maybe adjustment in DST. Weekend gap. etc. Or market order failed to be filled. These are the things I need to think thoroughly and be prepared for.

Thank you and looking forward to your reply.
  • #560
  • Edited 2:41pm Mar 3, 2017 1:36pm | Edited 2:41pm
Quoting antonym
I am wondering if you would, given your solid experience in developing EA, provide a list of possible errors that one may encounter in practice? This would be a huge saving in development/testing time and cost for me. Because otherwise I would have to wait for the errors to happen in realtime and only then start to work on how to handle it.
First, here is a description of MT4 error message codes, as provided originally by MetaQuotes.

Below is a code fragment from one of my (unpublished) EAs, that deals with placing an order. It assumes that the entry price, SL and TP have already been calculated according to the trading strategy, and that they are correct (e.g. for a buystop order, the entry must be higher than the current market price, and the SL must be at least a few pips lower than the entry, etc). In other words, many errors that inexperienced programmers encounter never occur because I make sure that I have valid prices, trade size, and values for other OrderSend() parameters before OrderSend() is executed. Hence the only errors I'm concerned with are those returned by the br0ker, for unforeseen reasons, and there is nothing I can do to prevent them anyway, except instruct the EA to wait a while and try placing the order again.

Hence rather than have a specific handler for each different type of error, I'm a lazy programmer: I use a 'brute force' approach when handling these errors. In the parameters of my EA, I have:
Inserted Code
extern int    NumberOfAttempts
extern double SecondsBetweenAttempts
 
: : : : : : : : : :
 
  sym     = Symbol();
  ccy     = Symbol();
  tmf     = Period();
  bidp    = MarketInfo(ccy,MODE_BID);
  askp    = MarketInfo(ccy,MODE_ASK);
  pnt     = MarketInfo(ccy,MODE_POINT);
  dig     = MarketInfo(ccy,MODE_DIGITS);
  spr     = MarketInfo(ccy,MODE_SPREAD);
  tickval = MarketInfo(ccy,MODE_TICKVALUE);
  minlot  = MarketInfo(ccy,MODE_MINLOT);
  lotstep = MarketInfo(ccy,MODE_LOTSTEP);
  lswap   = MarketInfo(ccy,MODE_SWAPLONG);
  sswap   = MarketInfo(ccy,MODE_SWAPSHORT);
  if (dig == 3 || dig == 5) {
    pnt     *= 10;
    spr     /= 10;
    tickval *= 10;
  }

If an error is returned by the br0ker, the code goes through a re-try loop NumberOfAttempts times, performs a RefreshRates(), and then waits SecondsBetweenAttempts, before the next attempt. If the error persists beyond the number of attempts, details get logged to a diagnostic file in my /MQL4/Files folder, the order does not get placed, and nothing happens until the next setup occurs (depending on the strategy).

If you simply copy this code, it will compile with a zillion errors, because the rest of the EA is missing (and it's coded for the MQL4 build 509 compiler). I've posted it simply as an example as to how I've coded the 'brute force' loop, to give you ideas for a possible starting point. Then you can modify it depending on how you might want to handle specific types of error, according to your strategy/trading philosophy.

If you have specific questions, post again and I'll do my best to answer them.
Inserted Code
// Calculate position (lot) size.....    
    riskstep = CountRecentLosses(EAname + ReduceCcy(Symbol()));
    risk = RPT[riskstep];                                      // defaults to EA parameter setting.....  
    if (StringLen(desc[1]) > 0)    risk = desc[1];             //  ..... but can be overridden by the 2nd entry in the box's decsription
    RiskType    = StringSubstr(ExtractAlpha(risk,"#%$:&")+"#",0,1);
    RiskVal     = MathAbs(StrToNumber(risk));
    AcctFreeMgn = AccountBalance();
    if (StringFind(StringUpper(risk),"E") >= 0)    AcctFreeMgn = AccountEquity();
    if (StringFind(StringUpper(risk),"F") >= 0)    AcctFreeMgn = AccountFreeMargin();
    if (RiskType == "&")           ordvol = DivZero(RiskVal/100*AcctFreeMgn, CalcADR(ADRdays)     *tickval);    else     // ADR pips = approx RiskVal% of account
    if (RiskType == ":")           ordvol = DivZero(RiskVal/100*AcctFreeMgn, 100                  *tickval);    else     // 100 pips = approx RiskVal% of account (RiskVal:1 leverage)
    if (RiskType == "%" && SL>0)   ordvol = DivZero(RiskVal/100*AcctFreeMgn, MathAbs(entry-SL)/pnt*tickval);    else     // SL pips  = RiskVal% of account
    if (RiskType == "$" && SL>0)   ordvol = DivZero(RiskVal                , MathAbs(entry-SL)/pnt*tickval);    else     // SL pips  = $RiskVal
                                   ordvol =         RiskVal;                                                             // # lots   = RiskVal
    if (lotstep > 0)
      ordvol = lotstep * MathInt(ordvol/lotstep);
    if (lotstep == 0.01)   ordvol = NormalizeDouble(MathInt(ordvol,2),2);
    if (lotstep == 0.1)    ordvol = NormalizeDouble(MathInt(ordvol,1),1);
    if (lotstep == 1)      ordvol = NormalizeDouble(MathInt(ordvol,0),0);
 
// Finalize the entry/TP/SL levels, and place the order.....                          
    entry = NormalizeDouble(entry,dig);
    SL    = NormalizeDouble(SL,dig);
    TP    = NormalizeDouble(TP,dig);
    for (z=1; z<=NumberOfAttempts; z++)   {
      int tkt = OrderSend(Symbol(),order_type,ordvol,entry,99,SL,TP,order_comment,magic,0,arrow_color);     // attempt to place order; return ticket# in tkt
      errnum = GetLastError();
      if (errnum < 1 && tkt >= 0)   break;
      Sleep(1000*SecondsBetweenAttempts);
      RefreshRates();
      if (dir>0)              entry = NormalizeDouble(Ask+EntryConfPips*pnt,dig);
      if (dir<0)              entry = NormalizeDouble(Bid-EntryConfPips*pnt,dig);
      if (limit_entry>0)      entry = NormalizeDouble(limit_entry,dig);
    }
    if (errnum > 0)
      dump("Error creating order " + err_msg(errnum));
    else if (PostDiagnosticInfo >= 2)    // Post Diagnostic Info !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
      dump(NumberToStr(tkt,"'Order# 'T12' created successfully; '") + err_msg(errnum));
    if (StringLen(SnapshotFile) > 0)  {
      string fname = SnapshotFile + Symbol() + "," + TFToStr(Period()) + DateToStr(TimeCurrent(),"'['Y-M-D H-I-S'].gif'");
      WindowScreenShot(fname,1024,768,-1,-1,-1);
    }
  }
  return(0);
}

Quote
I have followed some of your posts and learned that you are a professional programmer and have participated in several MT4 project.
I'm not a "professional" MT4 programmer. I don't write code commercially for other traders' requests. It's too much hassle, the projects tend to go on forever as the trader keeps changing his rules....... and you never get paid enough, LOL.
6 traders viewing now