the “comments” text of the EA status uses a default property in MT4, literally called Comments(), and it has no properties that can be set regarding size/color/location.
I can change it from Comments() to a text label, with more properties. But I don’t think making it orange or putting a background around it is going to make it any easier to read.
Simple fact is I’m trying to put too much info on the screen. That’s the way it is with all charts, the minute you start adding indicators and EA’s which have status text. You start running out of room.
I can remove the comments entirely, but I for one like to see why something is or isn’t working.
I can also add a switch True/False to show comments, so you can optionally have them on or off.
And it is only the top left corner. Might I suggest you just look at a different location on your chart??
Make a list of US holidays while you’re at it,.
Now you’re talking if you live in the UK don’t trade UK holidays, and if you live in US, don’t trade US holidays, or both. Where you live is not part of MT4, but it can tell the time of your local machine, and make a guess where you live.
I suggest you just turn off he EA button on those days.
Well, leave it to me to create a false sense of success. I had made a mistake in the MA Channel indi, and instead of measuring trades above the High, and trades below the Low as OK, I found I did a copy/paste of code and forgot to change the High to Low on the lower band.
What that means is the filter was saying, if the price is above the 5ma High, ok to trade. If the price is below the 5ma High, also ok to trade. So all it really did is prevent trades being made when the price was exactly ON the 5ma High.
Who knows, maybe we’re on to something new. Anyway, I’m making corrections and retesting to see what the real story is.
But hey, if a mistake produces better profit, so be it. Add it to the bag of tricks.
Here’s the bottom line. UPC/GMT eliminates all the headaches about Broker Time and Daylight Savings Time.
That’s why I like it.
You mentioned that trading starts at 5PM Sunday and stops at 5PM Friday (10 am and 10 pm GMT). Is that true for Summer AND Winter. Because 5PM EST isn’t always 10PM GMT. Depends on DST. In fact, right now I think that 5PM EST is actaully 9PM GMT, but it is still 10 PM UK time. Because in summer, UK is GMT+1, but in winter, UK=GMT
So, I guess the real deal is that in the Winter, forex opens at 10PM GMT Sunday, and closes at 10PM GMT Friday,
but in the summer, forex opens at 9PM GMT Sunday, and closes at 9PM GMT Friday. Which means it is probably based on UK time. In which case, I have to code in the UK DST start and end.
The software can already determine DST on/off in your local location, so it would be different depending on where you live.
One other thing I discovered after putting in the new code: during backtest, it does continue with correct UPC time, even during backtest, but that’s ACTUAL REALTIME UPC time, not the UPC time the backtester should be seeing. So, I actually had to hard code the time as being 23:45 server time for closing trades. Because, if I"m not mistaken, most brokers’ servers are GMT+3 and the day ends at midnight, and new day starts at 00:00.
Guess there’s still a llittle work to be done on the friday close thing,
No don’t get rid of the comments or waste time adding a true/false etc it works as it is, just thought some people might prefer things that way.
I do move the chart to a different location, its all good, i’d rather have a fast EA than a lagging one that looks pretty lol.
Are you sure… im sure you code the EA to use a DLL to send a 1 or 0 command which could be used to switch on a usb heater attached to the PC… Presto heated coffee lol
Don’t forget Penicillin was found by accident
Yes you got it right about the whole DST and BST, it can get confusing, i’ve lived with it my whole life i think it as odd not to have it.
I just found this article one of the babypips members wrote about it all a few years ago:
Still relevent, not sure if this makes it a pain to automate ?
I’ll try get a list together tomorrow. The thing i do remember is most trading is done in London, then New York so if we have a major British or American holiday and most of the Brits or American’s go on holiday due to one of our Bank Holidays or American Independence Day then its advised not to trade as most people are not trading and the market will be thin resulting in fake trades.
I do agree with you that on these days i would just simply turn off the EA, i only suggest this as i doubt you can filter trading days on a back test so if something like this could be easily entered it might be useful for back testing even if we never actually used it for forward testing or live trading.
I’ll try get a list together of major days that we should not trade. For me since we are trading those 5 pairs we used i would advise not to trade a US or UK bank holiday, perhaps you guys who have been trading this strategy for alot longer and live could elaborate if you trade the bank holidays or if you have traded them and what results you have or if there are any days you would avoid completely such as the obvious ones of Christmas and New Years ?
Thanks for the additional info. Very helpful. From reading this I realize it’s worse than I thought. Each country that has a DST change, moves it’s trading business to the new time, so it still looks like trading opens at the same time, eg: 08:00, but the trading times for all the other countries does not change, so relatively speaking, it looks like the trade open changes in all those countries by +/- one hour (depending on going into/out of DST).
Gee, that’s great. What a bunch of ego-centric/nationalistic business mentality. As has been suggested, and always ignored, DST should be dumped altogether. Stupid is as stupid does.
Anyway, long way to say, SB, no need to put together the list of holildays, because they change calendar dates year to year, such as ones defined by day-of-the-week, and not a calendar date.
I’m not saying it can’t be done, but is it really worth the effort to make tables of dates and/or days of the week related to months of the year, and tracking which country it is and where the computer is you’re working on, which day to turn off/on the EA.
I don’t know about you, but I do not have a burning need to figure this out.
You could just call the electricity company and give them a schedule when to cut the power to your home, and that would cover it, I think.
P.S., working on v14, with some StopLoss changes, and found out through backtest that the channel filter works better with period 3, on EURUSD jan-sep anyway.
There are plenty of look up resources on the net for those who don’t know what time of day it is. Surely leave it to the trader. It shouldn’t be an issue.
Maybe your suggestion of coffee was a good one though.
Tell me about it… Too many ego’s and no one wants to play ball. Im not sure what the real loss is if everyone just picked a random day of the year and did it then. Would make everyone’s live’s easier !!
Year i know the holiday thing is a pain, not really needed for the EA, as we should use our brains for trading a little, just thought it might help the back testing a little to make it a tad more accurate
P.S., working on v14, with some StopLoss changes, and found out through backtest that the channel filter works better with period 3, on EURUSD jan-sep anyway.
Thanks for the work on the EA, looking forward to v14, i wonder if there are different settings which work better for each pair ?
Hi SB,
you’re probably right about the backtest thing, and holidays, however, over the long run, a few botched trades on the holidays, as infrequent as they are, probably won’t skew the results enought to worry about. Not to mention, the platfrom is actually down (i.e. no candles to test) for the majors like Chirstmas and New Years, if my memory serves me correctly.
re: the different settings per pair, I’m afarid that will be the case as it usually is. The Holy Grail is an EA that can turn any pile of rubble into $$, but as yet, haven’t found it. Although, I have to say, one of the first impressive about Windsor’s system is that it DID show a profit with default settings on most pairs. That was a first for me.
Still grinding away at v14. Chasing my tail a bit, fix one thing which screws up another, and round and round.
Trying to get a clear mind to see it all at once.
Take your time in coding the EA, were in no rush, plus if your worried about the v14 EA doing odd things you can compare its trade to your v13 which you can view on my fxbook account (its set to public viewing)
Also to make life easier for those who want to view my fxbook account just go here:
As far as I can figure, trading the opposite on a pairs that run opposite, is same as doubling up on the single pair. If you win, you win on both, if you lose, you lose on both. Not really hedging. Anyway, true hedging, taking opposite positions on essentially the same commodity is a “net win or loss (losing one side of the hedge and winning the other)” always less than the win or loss without the hedge. (maybe that’s what you meant anyway). I think it all comes out in the wash.
But it’s too much for my brain to keep track of. That’s why I don’t trade options, buying puts/calls and selling covered calls/puts to “hedge” your trades. Kirapp. Can’t do it.
But hey, whatever you can get from this thing, go for it.
Off to bed.
Talk to ya later.
Opps sorry for that, thanks for correcting me. I think this EA works quite nicely on the 5 pairs we have so far
Have a good sleep, will be here when you wake
Just looked at the charts this week, if i wasnt mucking around the the fxbook account and the EA yesterday so far it would be in a profit on the daily chart quite nicely. Nice work on the EA serisouly !
Plus i would get + 4.5 pips from my rebate for these trades, so i would be almost breakeven for today using the EA.
I would normally close that 36.9 and call it day for today but since i want to see how this EA does on true auto-pilot i am resisting the urge to close any trades it makes. If i had i would of close a 40 pip trade which would of made it to 70 pips from fridays.
Only difference is i increase the max spread to 8 to take into account openning pending orders at night when the spread might be higher, i think we only need to keep it limited low if were scalping which were not.
Im trading across GBPUSD, USDCAD, AUDUSD, EURUSD and AUDCAD.
Glad were getting simiilar results Mitchell, by the way in a week a two could you compare your results to what im getting on my fxbook account just to ensure were getting similar results ?
Makes sense, but testing other pairs is useful though, there are likely to be other pairs which we could trade with successfully or some pairs which are better in the summer, others in the winter etc who knows.
I think for now for the EA at least its best we test using the same pairs to compare results, if its not too much trouble to ask if you could test your new filter theory with the pairs were using then you can compare your results with ours.
If you don’t want to run James EA just look at my fxbook url for the stats on the daily EA so you can see how your manual trading is comparing to the EA. It updates every 5 mins, so should be quite reliable
Plus every little price action filter which makes the EA more accurate is better for us all, who knows this may be our dirty little grail secret
to a different location, its all good, i’d rather have a fast EA than a lagging one that looks pretty lol.
im sure you code the EA to use a DLL to send a 1 or 0 command which could be used to switch on a usb heater attached to the PC… Presto heated coffee lol