Baruch's Forex System Journal

[TRADERguy]

Quote from TRADERguy:

OK- First of all trading the 10 year note on numbers is better than the 30 year bonds. Since they stopped issuing the 30 year the liquidity has dropped. The first thing you need to do before trading the 10 year notes on a number is to look at a daily and/ or 60 minute chart to see if they are overbought or oversold. If they are at either extreme be very cautious about going towards the extreme (selling an oversold market or buying an overbought one). I only put on a position immediately if the number(s) is out of range. Bloomberg surveys economists prior to each number; I throw out most outliers especially if they are not a major and well respected institution. I alter this range to be on the safe side if the market is overbought, oversold, or I feel that traders have a different bias than the economists. To illustrate this last point (and this is from memory so please don't correct me if I'm off a couple of $k), take the last non-farms payroll number. The Bloomberg survey had the estimate at 170K. The derivatives future/betting auctions (run by Goldman I think) had it at 186K (US the day before) and 193K (EU) a half hour before. The whisper number was in the low 200's; I don't know how to get this number, our analyst gets hold of it--I think its the rumor from the floor--was in the low 200's. So given all that I figured the safe # to sell at was 225K+. It came out at 287K so it was a no brainer. By safe number to sell, I mean see it as it comes out on Bloomberg and sell it instantly before I look at anything else (including the price). If you do not have access to Bloomberg/Reuters then I would say to look at the price first except for sure fire out of range numbers like the above example. On really important numbers (nonfarms and FOMC statements lately/ inflation #'s increasingly important) there will be at least 2 waves in the move. The first is when all the futures speculators click on the number. The second is when the paper (cash market) starts to move. Because of this you should be able to trade important #'s that are out of range even if you get the number from a source that is a little slow like Bloomberg TV.

P.S. Never, ever, ever trade the unemployment #, always wait for non-farms.

I'm catching up on reading Baruch's journal this morning and I see I missed something here.

You also have to watch out for the revision to last months number. This is especially important if it was out of range. If so people will be wondering if the number was wrong. For the above example, the previous month was 308K and it was revised up to 340K; so it was confirmation that it was a sell.

 

[TRADERguy]

Quote from Baruch:

Another thing: the whisper-numbers. From where can we - who don't have our own analyst - get those numbers, so we know the real consensus among traders?

I don't know of a source where you can get this number. For the US numbers, at least, it never seems to be right so I wouldn't worry too much about it. I use it as another piece of info to gauge what traders are looking for--but that it pretty much its only use.

 

[TRADERguy]

Quote from Baruch:

TRADERguy,

Thank you. I think we all learn a ton reading your comments. But you must get a lot of small losses, when you go flat if the price moves 1-10 pips against you? Do you use any indicators? How did you trade the news Friday?

Yes, trading short term is not for those who need to be right most of the time and not for anyone that lacks discipline to cut their losses. You have to treat each trade independently and stay in the moment (except if you start having a string of losers; them you have to ask yourself if you are doing something wrong).

Friday, I stuck to trading the Euro. Probably because I had a feeling that the numbers would be in range (boy was I wrong there) and also because it was a short trading day for the notes because of the holiday weekend and I wasn't sure how many big players were already on vacation. After the 8:30AM Eastern numbers I scratched 12250s then bought 12246 an took it up to 65 over several trades over the next half hour. On the Michigan consumer confidence I bought 37 and took a loss 33, bought 34 out at 33, bought 31 out at 36. On the Chicago # I sold 19 and took 14, then sold 21 and took a loss at 24, then sold 28s and took it down to 03 over about 10 trades. I should have quit there but I thought that they would continue down and I took a few losers at the end.

 

[LoosenUp]

Quote from TRADERguy:

My transaction costs are $.60 to $.80 per round turn depending on if my firm qualifies for the volume discount at CME that month: CME fees for members are $.20 per side for less that 150,000 contracts & $.10 per side for 150,000 contracts or more. My costs will come down in the future as I trade more; my firm offers volume discounts on their commissions at different levels up to 100,000+ RTs.

Most of the time I try not to pay the spread by using limit orders. During big moves especially after a number has just come out I have to pay the spread.

Yes most of the time I am looking to take 5 tics or less of heat. After a number that gets the market moving I usually have to take a little more heat or I will get shaken out. And yes I only trade the futures. One likes to think that at some point I will be trading big enough size that I will be forced to switch to the cash, but I am not there yet.

Traderguy,

Do you trade other futures instrument? E-minis, Bonds, Bunds, Eurostoxx etc? If so, I would like to know your comm. per RT for these instruments. How many RTs you do per month to get that kind of rates?

Thanks a lot.

 

[TRADERguy]

Quote from chinook:

\

Traderguy,

Thanks for the interesting suggestions. I also trade the ES. I might just have to give up the pre-US EUR session. Are you trading EUR from the US?

Chinook

Yes, I'm a couple time zones east of you but the hours still suck. I think that eventually I will trade from Europe.

 

[TRADERguy]

Quote from LoosenUp:

Traderguy,

Do you trade other futures instrument? E-minis, Bonds, Bunds, Eurostoxx etc? If so, I would like to know your comm. per RT for these instruments. How many RTs you do per month to get that kind of rates?

Thanks a lot.

I trade 10 year notes on some of the #s.

Transaction costs (fees & comm) per rt:

# of RT per month <5K 5-10K 10-40K 40-100K 100K+
CBOT $.62 $.58 $.54 $.50 $.42
Bunds E.88 E.84 E.76 E.72 E.64
Eurostoxx E1.04 E1.00 E.96 E.92 E.84

# of RT per month <2.5K 2.5-5K 5-10K 10-40K 10-100 100+
e-mini 1st 100 free $.63 $.59 $.55 $.51 $.47 $.35

 

[LoosenUp]

Quote from TRADERguy:

I trade 10 year notes on some of the #s.

Transaction costs (fees & comm) per rt:

# of RT per month <5K 5-10K 10-40K 40-100K 100K+
CBOT $.62 $.58 $.54 $.50 $.42
Bunds E.88 E.84 E.76 E.72 E.64
Eurostoxx E1.04 E1.00 E.96 E.92 E.84

# of RT per month <2.5K 2.5-5K 5-10K 10-40K 10-100 100+
e-mini 1st 100 free $.63 $.59 $.55 $.51 $.47 $.35

Thanks.

Pls let me know who your broker is and what are their minimal platform fee, if any.

And, why is the e-mini free for the 1st 100? what is the catch here?

 

[TRADERguy]

Quote from LoosenUp:

Thanks.

Pls let me know who your broker is and what are their minimal platform fee, if any.

And, why is the e-mini free for the 1st 100? what is the catch here?

I work for Refco Trading Services (Macfutures re-branded after Refco purchased them). They are a prop shop, not a retail broker, and do not offer remote trading.

No clue about the 1st 100 free for the e-mini. If I had to guess I would think that this is an exchange thing that was implemented when the product was new to get people to try it.

 

[Baruch]

Quote from TRADERguy:

Yes, trading short term is not for those who need to be right most of the time and not for anyone that lacks discipline to cut their losses. You have to treat each trade independently and stay in the moment (except if you start having a string of losers; them you have to ask yourself if you are doing something wrong).

Friday, I stuck to trading the Euro. Probably because I had a feeling that the numbers would be in range (boy was I wrong there) and also because it was a short trading day for the notes because of the holiday weekend and I wasn't sure how many big players were already on vacation. After the 8:30AM Eastern numbers I scratched 12250s then bought 12246 an took it up to 65 over several trades over the next half hour. On the Michigan consumer confidence I bought 37 and took a loss 33, bought 34 out at 33, bought 31 out at 36. On the Chicago # I sold 19 and took 14, then sold 21 and took a loss at 24, then sold 28s and took it down to 03 over about 10 trades. I should have quit there but I thought that they would continue down and I took a few losers at the end.

TRADERguy,

Thank you. Do you trade the number (sell the good number and buy the bad number) or just follow the flow? How do you know when to take your loss - and your profit?

 

[FX_Cowboy]

Quote from Baruch:

Where do you place s/l? I know you are a kind of computer genius: Why do you not construct your own system?

I place my support and resistance lines/trendlines the same way everyone else does, I suppose, by observing how price behaves at certain levels. In addition, in forex futures you have the advantage of having volume on the screen, which can be quite a good indication of significant price points/areas.

While I am in no way a computer genius, I am a programmer, and I have built systems in the past using IB's programming interface, and also eSignal's interface. So far, I haven't built one that is consistently profitable -- or to put it bluntly, they've all lost money in the long run. I haven't given up looking, but I do find that I trade better using some discretion.

I mean a automated/mechanical system - like we have a robot trading for us. Can this be done?

It certainly can -- and is -- done. The key elements are having a profitable system in the first place, and being able to break the system down into rules that a computer can execute. I'm thinking about putting together a "Baruch MA Breakout System", but I need to define the entry and exit criteria a little better. Maybe we can collaborate, and then I'll give you the program once we've defined the rules.