I don't think I was being defensive at all. I mean honestly, we can discuss this forever. It's just easier to see the numbers play out honestly. I'm pretty sure people prefer evidence over theoretical discusssions.
Index futures automation
- Thread starter hilmy83
- Start date
I don't think I was being defensive at all. I mean honestly, we can discuss this forever. It's just easier to see the numbers play out honestly. I'm pretty sure people prefer evidence over theoretical discusssions.
Speaking of which my bot just crapped out. I decided to trade manually now stuck short at 15485. If it goes to 15538, it'll be major loss due to some malfunction that I still can't figure out and due to stupid action on my part today. We'll see.
The funny thing is I have two otehr accounts following the same bot that executed just right.
It's amazing how crappy bot can be when actually trade live. I know the strategy is legit. It just needs perfect execution.
The funny thing is I have two otehr accounts following the same bot that executed just right.
It's amazing how crappy bot can be when actually trade live. I know the strategy is legit. It just needs perfect execution.
Just wondering, if you see it's a malfunction why keep the trade open at all?
Wouldn't it be better to cut the loss and move on instead of hoping that the market will turn around?
Well, booked a loss today. The algo initially had a long position, then it took a loss and reversed. Instead of taking a b/e like it's supposed to do, the TP was cut short booking a loss. Completely unrelated to the strategy. Initial loss was -$900, then I decided to trade manually and made it worse. Total loss today is about -$3000
Meanwhile wife's account and another account did exactly the algo was supposed to do!
So it's the repeat I keep seeing back in QT. One account does well and the other somehow crapped out. It can't be the algo. I have a feeling it's data/latency related. I switched over back to MT5 back in mid september this year, and NO ISSUES till today.
I thought that I had something to do with AMP update of removing commissions MT5, but again, doesn't explain the discrepancies.
I dont know, I'm sick of debugging that I'm pretty much accepting this is the best I can have it. Annoying that a SIMPLE bot work all the time.
Anyways, I'm not going to include today's negative outlier in the results because it had NOTHING to do with the strategy. Just as I didnt include 2 positive outliers I had (i think totaling $9k profits altogether), so that I can keep the journal focused on the strategy itself, not the shitty execution I get from time to time....
Today's loss is about -8% overall. Could've been worst. I should've left it alone but part of me was angry the bot didnt execute like it should and I felt cheated.
Sigh...fuck. Back to grinding.
Meanwhile wife's account and another account did exactly the algo was supposed to do!
So it's the repeat I keep seeing back in QT. One account does well and the other somehow crapped out. It can't be the algo. I have a feeling it's data/latency related. I switched over back to MT5 back in mid september this year, and NO ISSUES till today.
I thought that I had something to do with AMP update of removing commissions MT5, but again, doesn't explain the discrepancies.
I dont know, I'm sick of debugging that I'm pretty much accepting this is the best I can have it. Annoying that a SIMPLE bot work all the time.
Anyways, I'm not going to include today's negative outlier in the results because it had NOTHING to do with the strategy. Just as I didnt include 2 positive outliers I had (i think totaling $9k profits altogether), so that I can keep the journal focused on the strategy itself, not the shitty execution I get from time to time....
Today's loss is about -8% overall. Could've been worst. I should've left it alone but part of me was angry the bot didnt execute like it should and I felt cheated.
Sigh...fuck. Back to grinding.
Last edited:
So I'm trying to figure any change I did to cause this. Now I did bring a new instance of MT5 last week onto the VPS (the one my account is running on). So I have 2 MT5 instances running on one, and my wife's running on a separate VPS.
I didn't bother to change anything about the algos running on those 2 instances. But I'm thinking now that since they both share magic number, I'm wondering if that caused my EA to mess up.
I thougt MT5 magic number doens't get shared across different MT5s runnign separately. I don't think that's the issue, but I'm going to make sure both instances have different magic numbers going forward.
It did seem that way, though, as you were inviting me to open a journal in order to prove you wrong. I didn't even say anything about win rate or other systems. Just my view on yours as you shared a comment from Pax which I tend to agree on and had earlier said the same thing.
I've considered a journal, but I'm not sure I will as I'm wondering if the negative effects outweigh the potential benefits. One being that there will be a lot of comments. I imagine that's easier to handle if one trades algorithmically, though. Maybe a journal will be positive for me. Need to think on it.
Journals is what gives any forum any integrity and also a way to prove a point. I mean you go to these threads and what do you see, just endless back and forth. Oh sure, there'll be few charts here and there, nicely marked in hindsight. Maybe even a few calls, but at the end of the day, people just talk past each other.
WIth journals you can prove your point with numbers. As nerdy Shikara would say, numbers don't lie. And also journals need to be a commitment. 1-2 months of good results is just pointless.
For your peace of mind, open a 1 lot of MNQ in one MT5 instance and then immediately try closing it using the same magic number in the second MT5 instance. If it is closed then you have your answer. To date, every "bug" that you've encountered in MT5 has had a logical explanation.
So open 1 lot on both instances? then close one manuallyusing same magic number? I think magic number wouldn't exist for manual trades, it's EA specific.
I'm not sure how to replicate that with the EAs, unless you're talking about closing with a script using the same magic number.
I'm just going to use unique magic number for each EA on the platforms that shares teh same VPS, and just go from there.
This happened literally 1 day after I added that second instance of MT5 on the same vps. 2 months prior to going back to MT5, no issues.
I've looked at journal and expert tabs to see what the issue is. All three platforms executed the trades within ms of each other, they all had the same exact EA alerts/messages/everything in same order.
Drives me nuts. Can't be this damn complicated lol
My account on the vps was at 37k and the other new account was at 6k. So assuming they communicated with each other due to same magic number (not sure how since it's on 2 separate platforms), my EA thought that it just needed small price move to recover to b/e? like could that be it? That is some ghetto ass algo function if it's true.