my journal 3

Yeah, this thing is very confusing. Everyone keeps talking about JPY in terms of dollar and yen during the same sentence sometimes. It doesn't happen on the other currencies.

Well, right now I am down about 2000 dollars. If I were home, I wouldn't resist and I'd close it immediately. But I am at work.

In the meanwhile I've spotted a tremendous long opportunity on CAD.
 
  • Like
Reactions: D70
Yeah, this thing is very confusing. Everyone keeps talking about JPY in terms of dollar and yen during the same sentence sometimes. It doesn't happen on the other currencies.

Yeah, 6J and any other cross with USD as base is the inverse of spot.
 
Back home.

Usual routine. I checked first if there's emails from IB talking about margin violation or liquidation and there's none. Great.

Other than this, now comes the moment of truth.

When I last checked, this morning at 9 am CET, JPY had gone just 15 ticks from my first contract exit. Did it fall after that, in the ensuing 6 hours?

I wonder.

I think at least 2 contracts got executed, but I dream of all 4 contracts being executed.

And after this, if I get away with this, I must really quit my martingale crap.

I'll check in a while, after getting changed so I can be ready in my pajamas in case of a heart attack.

...

Didn't check yet. Anyway, in my opinion it went up because remember what I said about the JPY? It either goes up all day long or down all day long, and so given that it decided to go up, it will go up all day long, and big time hopefully.

So I am definitely hoping to see all my 4 contracts closed. Of course, whatever happens, given that I spread them out over 40 ticks, I'll have something to regret. But let's remember that it was what I chose to do, and that when I did choose, I did not know the future.

...

Wow, I've been home for half an hour and I am still postponing checking how my trades turned out. I am looking for some magical propitiatory ritual to do before checking, pretty crazy... but I can't come up with anything really powerful to skew JPY in my favor.

I'll click another poke to my facebook girlfriend. By the way, we're back together I think. She's been poking me a few times per day as of lately.

Nope.

She hasn't poked me back, so I can't poke her again.

What do I do?

I feel like going out and buying some sushi. It's japanese food, so it might help my trade, literally. If everyone started buying sushi, JPY would somehow be affected.

Yeah, I should definitely go out and buy sushi before checking the markets.

Besides, as i said, it goes all day in one direction, so the longer I wait, the longer it has to go up, and you know what happens if I check. I will be so happy to get out with a small loss, that I'll cut profits short even on the last contract, or whatever is still open. Hopefully just the last one, which didn't have an exit.

...

I guess it's like when someone dies and I get very mad at my aunt for telling me that "x died". Stupid bitch. Or my father does it, too. Mother ****ing asshole. I tell them to break the news to me little by little and not all at once.

In the same way, I'd like to do for JPY. Someone to meet me and tell me: it is going pretty well, but not as well as you expected...

Or someone to tell me: sit down, unfortunately...

I think I am going to get drunk, which is the equivalent of having the news a little at a time. Right, because when you're drunk, everything is softer.

I'll go buy sushi and japanese beer, so I can actually help the JPY to rise.

Then, once I am drunk, I will check how my trades are doing.

...

OK, back.

It's been an hour.

I ate my japanese food, and drank one sapporo beer already. About to start the second one, at the end of which, I will check my JPY trades. Already a bit dizzy.

The child next door is talking, he can't be nuked, so I need some heavy duty music.

Actually I am in a good mood, because I presume my trades went well and got executed.

If it's not the case... I better be drunk when I find out.

...

Ok, the child is still noisy. I need red hot chili peppers. Norah Jones won't do.

The Zephyr Song - Red Hot Chili Peppers + lyrics - YouTube

You know, one sapporo beer can is one huge mother ****er...
You can walk into a movie theater in Amsterdam and buy a beer. And I don't mean just like no paper cup. I'm talkin' about a glass of beer.
And I don't mean no paper cup. I'm talkin' about a 650 ml can.

sapporo650.jpg

...

Ok, I am drunk enough to check the JPY.

I'll drink the rest of the can in case it didn't go as well as expected.

Let's go for it.

I'll start from esignal, because it breaks the news slowly and it prevents a heart attack. I don't want to see a -x thousands on TWS without expecting it. Yeah, that's how sensitive I am. That's why I engaged in martingale to begin with, because a 200 dollars bothered me that much.

Ok, but first of all, let's define my hopes. What am I expecting?

There was no margin call last time I checked so I am expecting to have a loss less than 4000. That's for sure.

1) loss > 4000 will make me very unhappy
2) ongoing loss = 2000 will not change my mood
3) 1 contract exited at 60 will make me slightly happy
4) a second contract exited at 70 will make me pretty happy
5) a third contract exited at 80 will make me abundantly happy
6) a fourth contract exited at 90 will make me extremely happy
7) a fifth contract with ongoing profit in the thousands will make me exceedingly happy

Red Hot Chili Peppers - Universally Speaking Lyrics - YouTube

A bit drunker.

Ready to go to esignal.

...

Wow, not looking good at all... approaching ti very slowly. I first looked at AUD, CHF, and GBP now. And wow, it's at record lows. That doesn't bode well.

Dollar is strong relative to GBP, so how weak can it be relative to JPY?

I wonder.

Little by little.

Let's check the other currencies, and last JPY. I am not ready. I am fearful. I am cold.

Holy ****!

AUD, CAD, CHF (so so), EUR, GBP are all weak to the dollar!

How on earth am I going to be making money on JPY?

I need to drink another class of sapporo.

Ok, **** it. I finished my last can of sapporo. It is one mother ****ing heavy can. It weighs about a kilo.

Let's go for it.

Let's check JPY.

No, wait. Let's check my email for a margin violation warning, to be prepared for the worse.

...

None.

So, where the **** is price?

Let's do it. Let's check JPY. On esignal.

wELL, I've written down my hopes anyway.

So let's check it. One more song before checking it.

Red Hot Chili Peppers - Can't Stop Lyrics - YouTube

My hands are cold. My beer is finished. I need to check JPY.

...

****!

It didn't get anywhere?!

In all these hours?!

The high is the same as this morning? 1.1045?!

How can it stay within that range for 24 hours?

I'm going to check the details now and then, in a minute, assess my happiness.

...

esigchartspon.png

wow...

No trades must have been triggered.

Yup.

Yep.

Right, because... JPY didn't exceed that high, from this morning... well, esignal is delayed by 10 minutes... but what's the chance?

So. How happy am I?

Let me check my ongoing loss on TWS.

I guess my prayers didn't work.

Well, at any rate it looks like it's bouncing big time. I mean, the chart looks real good.

Let's check the ongoing loss and assess my happiness.

Snap1.jpg

Mood lower than expected. A bit pissed off to unchanged.

I am going to keep this going overnight.

...

The CAD is flying now, upwards. Didn't I tell you it would?

I wonder how all these currency reversals will affect/interact with JPY.

I hope it will follow suit.

JPY seems to be stuck.

But remember what I said. If it goes up, it goes up big time. If it goes down, viceversa.

And today it was an up-day.

I cannot betray my appraisal of the situation, at a time when I was unbiased.

Its rise is meager. But i am not going to leave it to chase other currencies.

On the other hand, if I quit now, I would have equity at 20k, and my dream would still be alive.

Instead, I am monitoring the hourly chart, and praying:

Snap2.jpg

But don't forget that monitoring is bad for me.

20k.

3k lost relative to before starting this JPY trade.

3k/23k= 13% loss... acceptable.

Should I close at a loss of 3k?

Maybe I will, if I stare at it long enough.

I might regret it dearly if I don't.

I was expecting a 200 ticks rise today.

Ending the day with a capital of 20k would be a success at this point.

Ok, **** it.

I am going to set an exit at 40 fo everyoen.

0.011040 for everyone.

if it happens, so be it.

i'll lose 2700 from this evil trade - acceptable, totally acceptable

thats it

lets do it

17 ticks to go

fcourse i might cancel it in case it starts flying

16 ticks away from the end of the nightmare

Dream A Little Dream (Laura Fygi) - YouTube

14 ticks from the end of my nightmare

...

17 ticks...

Harry Nilsson - Over the Rainbow (explicit) - YouTube

...

Fine, JPY, you want a promise?

Fine, I swear I won't change my LMT order.

5 contracts at 40.

Just get there and I'll get the **** out.

I sweaar I won't change it.

0.011040 for all 5 contracts.

Not higher, not lower.

Oh, god, 20 ticks to go.

...

24 ticks to go

I should have thought abou texiting before.

now my 2700 of loss don't look so easy anoy more

25 ticks to go

27 tiocs tp gp

I shpouid have thpougth bout it before

28

29

screw you all

i am goin to turn it ooff

and set the exits much higher

...

no, wait there might be a way out right now...

3.jpg

... a bounce...?

the overleveraged trader hangs on the market's lips (can you say this? it sounds right in italian)

Anyway...

the move, what seemed a reversal is evolving:

4.jpg

As I said, I promise I won't move my exit LMT order.

This is it. If you let me off the hook, I will get the **** out.

The Godfather - Tessio is taken away

JPY, can you let me off the hook, for old times' sake?

20 ticks to go

18 ticks to go

...

The Godfather - 07 - Love Theme From The Godfather - YouTube

20 ticks to go

in the meanwhile CAD keeps rising. I told you so.

I should just get out of this crap and go long on CAD with all the capital I have.

19 ticks to go.

...

my facebook girlfriend poked me again!

Now the markets certainly will go my way.

I could make 10k, but I'll be happy with exiting with a 2700 dolalrs loss.

...

very odd. She poked me, on facebook, and yet the markets aren't going my way, yet.

now 21 ticks to go, before my orders get executed.

In the meanwhile, while I am blowing out the account, my journals are kicking ass:

j.jpg

Yep, 380k view all combined. I wish that was my capital. One dollar per view.

Damn. I suck.

24 ticks to go.

I guess I'll start another post, in order to finish the day. Too much information on this one and I wouldn't want some technical error to delete the entire post, with all the images attached.
 
Last edited:
fears, hopes, juxtaposing each other, in my head, restlessly

20 ticks to go

i keep pissing in my milk bottles, like the aviator, the radioactive sapporo beer

The Aviator - Come In With The Milk - YouTube

overleveraged trader

20 ticks to go

Harry Nilsson..." One " - YouTube

...

That's it. Can't take any more of this pressure.

long 1 gbp
long 1 cad
long 2 jpy

Exiting the other 3 contracts on JPY, pronto

...

exiting 3 JPY contracts at 21.

entering GBP and CAD wherever they are

no stoploss

at least I'll be diversified

...

Done!

Mother ****ers... at least I am rid of 3 JPY contracts

All the sushi I ate didn't help for ****.

...

Usual crap. Right after I sold my 3 contracts, JPY gained 10 ticks. I should have known. They were waiting for me to sell.

Good thing I kept 2 contracts, so I only lost 375 dollars in 2 minutes.

...

**** it.

More diversification.

I sold another JPY at 31 and bought instead an AUD.

So now I am long 4 currencies against the USD.

Screw them all.

All I need is to make 2000 dollars on all of them and I'll be fine.

If I can get back to 22k of capital, I am fine.

So be it.

Time to take a break and let the guy in, with the milk. Provided he doesn't touch the paper and holds the bag out to me at a 45-degree angle.
 
Last edited:
It's going ok, because I am out of JPY, except for one contract, and I am not regretting getting out of it, because it still didn't rise (didn't fall either).

This is the situation:

Snap1.jpg


Actually JPY momentarily went higher, but it was after midnight and there was no way I would have known, could have known, would have risked keeping 5 contracts for that long. It would have been crazy.

Now capital is at 20k, a bit higher than yesterday, 3000 lower than when I started my JPY trade.

I officially suck.

But these trades are better than what I had yesterday. I am happy about it.

...

Came back from the shower.

You know, I went long on NG, for just a 600 dollars trade, hopefully.

After a... two days with overleveraged 5 JPY contracts overnight and overweekend... you lose all inhibitions.

I am going to work and leaving innumerable positions open. I feel ok about all of them:

2.jpg
 
Last edited:
The only trade losing money right now is the one the systems made, NQ.

Gee, what a disappointment, regarding the systems.

Regarding my trades, very happy, because, judging from esignal (I am at work), all of them are being profitable by at least a few hundreds. So, given it's six positions I have opened, I am up at least 600 dollars, but probably 1000.

So, right now I should be at about 20500. Which is still 2500 dollars down relative to before starting the JPY martingale, but... I am back!

I haven't blown out my account, and probably won't! At least this time.

Hopefully, by the time I get home, a couple of positions will have been closed by the LMT orders being triggered. Because otherwise, I'd be tempted to close all those green positions, that's just the way I am, like the scorpion. So I better not cross rivers using frogs as a means of transportation.

At the end of this trading binge, I hope to be at 25k. At which point, I will have to withdraw some money to pay for the server (monthly fee).

If it happens, and I do reach 25k, I'll post the usually equity line, without bragging about it.

...

I checked again and I am making hundreds on every single future I have bought. No, wait: except NG, where I am losing 200, but that is probably only temporary.

Capital is probably at 21k.

Not much yet, but I'll hold it until tomorrow.

I am extremely satisfied.

Of course, if I had kept those 5 JPY contracts, I'd have even more money right now, but there was no way to know - I would have run a risk of blowing out my account. So I should not forget that I do not know the future, and be aware that i didn't trade the 3000 dollars I would have made with JPY with the 1000 dollars I am making with what I have bought instead, but I traded

1) the 50% probability of blowing out the account and another 50% probability of making 3000,

with

2) the 5% probability of blowing out the account and a 95% probability of also making 3000, although maybe more slowly.

So it was worth, despite the fact that, as a consequence of this exchange, now I have 21k vs 23k I'd have had if I kept the 5 JPY contracts.
 
Last edited:
I will have to withdraw some money to pay for the server (monthly fee).

Your server sounds expensive. Which provider do you use and what is your autotrading setup?

I'm asking because I pay 20 EUR/month for my Amazon EC2 instance which autotrades my InteractiveBrokers account (it looks like you are using IB too).
 
Amazon EC2 instance? What is that? And with that setup who's being allowed to see your code?

I am renting a server at a friend's house, in his closet. He's also an ISP, so he's really helpful with both programming and networking, and I am paying 300 dollars per month. On top of his help, I also trust him, as I wouldn't want to allow strangers to see my systems.

I trade with IB TWS in connection with Excel. All that needs to be running is TWS and an excel workbook that I created.

Let us know some more details about your setup, as other readers are probably interested. For example, I was talking about this subject recently with D70, who needs advice and help and ideas for his setup. He also wants to do automated trading.
 
  • Like
Reactions: D70
Amazon EC2: Amazon Elastic Compute Cloud (Amazon EC2), Cloud Computing Servers

It is a virtual computer (they call it instance). You may use many different operating systems (linux, windows).

You connect to the instance with SSH using private/public key (for security) and a VNC client to get to the OS GUI. These servers are located in locked buildings with no human access while they are running.

In theory there could be somebody getting physical access to the server but would have to bypass your private key login which is stored only on your computer.

And even if somebody knows which phisycal server you are using and knows why are you using it for and that your trading systems are very profitable, he wouldn't trust them anymore after a first little drawdown.

I would say somebody at your friends place is more likely to hack into your servers since he has physical access and also knows you are running your systems on those servers.

My configuration:

  • Amazon EC2 with Ubuntu
  • InteractiveBrokers Gateway
  • customized JBookTrader Java application for autotrading
 
Hmm, thanks for taking the time to provide this useful information. To me - although I am not a professional programmer - it seems a little less safer than running things on my own computer. On the other hand, I use VNC to connect, so some of the risks apply to Amazon Elastic Compute Cloud could apply to VNC. Also, I don't think I need it.

Oh, regarding my friend's house, no risks. There's only a cat there, and he can't reach - at least my friend told me so. He said the cat won't be messing with my server.

Well, I can tell you're a programmer by your setup: something I cannot afford in terms of programming. Congratulations on being a programmer, by the way. I have always had admiration for people more intelligent than me. It's such a rare event.

I mean, anyone who's a programmer or a science major is smarter than me, but precisely... they're rare. Probably not even 2% of the population.

I hope D70 will read and find useful your information, too. Maybe he'll join the discussion, hopefully.
 
Hmm, thanks for taking the time to provide this useful information. To me - although I am not a professional programmer - it seems a little less safer than running things on my own computer. On the other hand, I use VNC to connect, so some of the risks apply to Amazon Elastic Compute Cloud could apply to VNC. Also, I don't think I need it.

Oh, regarding my friend's house, no risks. There's only a cat there, and he can't reach - at least my friend told me so. He said the cat won't be messing with my server.

Well, I can tell you're a programmer by your setup: something I cannot afford in terms of programming. Congratulations on being a programmer, by the way. I have always had admiration for people more intelligent than me. It's such a rare event.

I mean, anyone who's a programmer or a science major is smarter than me, but precisely... they're rare. Probably not even 2% of the population.

I hope D70 will read and find useful your information, too. Maybe he'll join the discussion, hopefully.

Yup, I'm here and following.
Spending time on IB - API (dde links). Lots of questions will undoubtedly follow.
 
Ok, great. We're a crowd then. There's also others who are automated traders and are following the journal (e.g. the hare).
 
Last edited:
Ok, great. We're a crowd then. There's also others who are automated traders and are following the journal.

I've familiarised myself with the API for IB. Now I'm back on Tradestation seeing if I can code the logic. I then hope to either run it in TS or swing it across into the API (DDE) for IB (as you did Travis).

Plus trade it all manually at the same time.

Standard day is 0630 - 1800. Break til 2030. Trade the US close then continue with the homework til my eyes hurt at usually about 2330.

Rock'n'roll..........
 
I don't want to offend anybody with this post. I just want to share my story and why I don't use Excel for IB.

Many years ago when I decided to use automated trading strategies I also started with Excel. I wanted to make it easy on myself even if I knew how to program in other languages. I also read forums regarding IB and their DDE API.

After a while I realized I need more flexibility and more robust platform for my strategies. I also wanted my backtesting code to be the same as the one I use for life trading. I decided to use Java API. The main reason was that TWS and IB Gateway are written in Java and I decided to use the same language as IB (I know how to program in C++, Visual Basic so I could chose any language I wanted).

I'm glad I did that and in the past years I was able to build strategies and test them before going life. Many times strategies were giving me good results only to find an error in the way my application used data during backtesting. This saved me a lot of money.

In the past I also used four other backtesting platforms. I can only say that all of them are giving better results than they should. Once I found out I cannot trust how a backtesting platform executes orders, I spent all my time looking for similar errors in my application. I "destroyed" many "good" strategies by testing them as close as possible to the realtime life trading ;)
 
Message for D70 (post before jazeonli's).

Your daily schedule sounds like one of my trading systems: "Standard day is 0630 - 1800. Break til 2030". If they're so easy to code, then it can be done on excel. But you also need vba, and that took me at least six months to learn.

For example, one of my systems does this. Computes the moving average of a given future, all the time (complex to do on excel, but I've figured it out). Then, if at a given time, price is above it, it goes long for the rest of the day, closing at 22.00 CET.
 
Last edited:
Here is one of my strategies that is similar to your approach: Increasing/Decreasing Peak/Trough Volume with Normalized Momentum

Idea:
  • We observe the peak/trough volume. When volume at this points is increasing/decreasing, the direction change is probably to hapen.
  • We observe 30 minutes bars to determine increasing peak/trough volume.
  • 30 minutes bars should have at least 10K in volume (aproximately 5% of a long term daily volume).
  • Track momentum of the new lowest low/highest high bars.
  • Direction changes when there is opposite direction momentum greater than the last momentum of the new lowest low/highest high bar.
  • Track current total daily volume in relation to the average daily volume.

Enter/Exit Rules (descriptions are for long/short positions):

  • After minimum volume has been traded for the day (30%), wait for an extreme volume bar:
  • - If the price direction reverses, enter in the direction opposite to the last price move.
  • - If there is no significant move or daily price change, enter in the direction of the previous day move.
  • After maximum volume has been traded for the day (100%), wait for an extreme volume bar after which the price direction reverses and close the position.
  • If position is still open at the end of the New York session (22:00 CEST), close it before rollover time (23:00 CEST).
  • Allow for three up waves and two down waves before entering/exiting a position.
  • Wait for a failed breakdown (breakout) through the well known resistance area where many retail stop orders are waiting.
  • Wait for the price to pierce through a well known trading area:
  • - previous peaks/troughs
  • - daily open price
  • - simple moving averages (50, 200 days)
  • - price doesn't continue in the current direction when the volume increases
  • - volume decreases while price continues in the same direction but is beginning to slow down
  • - significant peak in volume after a clear trend
  • - major market moving events (wait after the event is over and the price settles down)
  • Wait for the total number of trades to quiet down before executing entry/exit trade. Be patient.
  • If it is a large move down with three waves, enter long on the blow-off bottom of the third wave.
  • Wait for an opposite direction normalized momentum greater than the last normalized momentum of the new lowest low/highest high bar.


I believe you agree this is too much for Excel and not a good idea to test it on one platform and then execute it on another one (like Tradestation + Excel). It is complicated enough as it is and I don't want to search for bugs in two different languages.
 
Last edited:
I don't want to offend anybody with this post. I just want to share my story and why I don't use Excel for IB.

Many years ago when I decided to use automated trading strategies I also started with Excel. I wanted to make it easy on myself even if I knew how to program in other languages. I also read forums regarding IB and their DDE API.

After a while I realized I need more flexibility and more robust platform for my strategies. I also wanted my backtesting code to be the same as the one I use for life trading. I decided to use Java API. The main reason was that TWS and IB Gateway are written in Java and I decided to use the same language as IB (I know how to program in C++, Visual Basic so I could chose any language I wanted).

I'm glad I did that and in the past years I was able to build strategies and test them before going life. Many times strategies were giving me good results only to find an error in the way my application used data during backtesting. This saved me a lot of money.

In the past I also used four other backtesting platforms. I can only say that all of them are giving better results than they should. Once I found out I cannot trust how a backtesting platform executes orders, I spent all my time looking for similar errors in my application. I "destroyed" many "good" strategies by testing them as close as possible to the realtime life trading ;)

Interesting. I am replying now, because I was writing another post while you wrote this. Absolutely excel is the least robust, but that's all I can afford. Yes, good point about noticing that automating on excel while back-testing on tradestation causes more problems than doing both on the same platform. By the time I realized I could not / should not execute my trades on tradestation (I don't want to be prisoner of them as a broker), it was too late, so I moved on to excel.

I told you I could tell that you were a professional programmer - of course you didn't teach yourself java and c++ just for the hell of it in your free time, right? Otherwise you're a genius.

Yep. I know about that problem, but it's not my case. I double-checked data, too and my systems are so simple that they don't test one way and execute another, and my data is reliable (backtesting on disktrading.com data, and forward-testing and executing on IB's data). The reasons for the worse performance relative to the back-tested one are to find elsewhere (survivorship bias or something of that kind - I explained that problem of my systems several times before).

Congratulations on your hard work and bright mind.
 
Top