IB or Oanda or ????

[jimrockford]

Quote from jessieblue:

How do you calculate expectancy? Should I make a note of all my trades and calculate in percentages my losses and wins than fill in the terms of that ecuation? or is it more like develloping an instinct.

No, jessieblue, you can't calculate an expectancy based on past performance of a trading strategy. This is because markets are statistically "nonstationary". The average behaviour of the market, or of any trading strategy, changes over time. The average behaviour you observe this week will be different from the behaviour you observe next week. The average behaviour you observe this year will be different from that you observe next year. It is not valid to calculate average behaviour over a time period, and then expect to encounter the same average in future time periods.

The market is different from a statistically stationary process, for example, flipping a fair coin or rolling dice. These stationary processes are the kind where you can calculate average behaviour and extrapolate to the future. The technique of averaging and extrapolating works, because you are always collecting data from the same unchanging process. Every time you flip the coin or roll the dice, it is the same coin or the same dice. Not so with the market. The market is always changing.

The illusion that markets are stationary is very dangerous, because it misleads inexperienced traders into thinking that they can make vast fortunes, when in fact, almost all of them lose. You will find lots of people who are trying to sell you something, by trying to convince you the market is stationary. You will also find lots of people who truly believe the market is stationary, because they don't know any better, and some of them will probably be very irritated by the suggestion that the market is not stationary.

Anyway, remember what the CFTC requires trading advisors to warn the public: "Past performance is no guarantee of future results."

 

[jessieblue]

Quote from late apex:

Yes, exactly. Based on your trades, preferably real, not simulated. Van Tharp's first book is a highly recommended read for basic expectancy stuff, and much more.
20 on those majors is far too tight in my experience. Cable sneezes, and it's moved up or down more than that, many times every day. Why try to be a hero, a neuro-surgeon with ordinary kitchen tools (TA), when you can do very well, playing just a doctor on TV?
Didn't have a chance to look at that book yet. But don't worry, I will. Is that where you got 20-pip stop loss from?

Never heard of van Tharp, I'll do a search

Quote from late apex:
20 on those majors is far too tight in my experience. Cable sneezes, and it's moved up or down more than that, many times every day. Why try to be a hero, a neuro-surgeon with ordinary kitchen tools (TA), when you can do very well, playing just a doctor on TV?

Sorry, didn't catch that analogy. If 20 is too tight, what do you reccomand? 50, 100? Remember, I'm a daytrader (for now anyway)

Quote from late apex:
Didn't have a chance to look at that book yet. But don't worry, I will. Is that where you got 20-pip stop loss from?

Making fun of noobies dude? That's just mean.
Yes, that's where I got the ideea. Come on man, don't knock it till you tried it.

 

[TriPack]

Quote from late apex:

I enter with stop or limit orders (which are, of course, one and the same on Oanda) and exit with stop, limit and sometimes market orders. Don't shy away from volatility at all. Unless you mean something like "do I deliberately straddle the news?" -- no, that's not a part of my trading strategy.

What do you mean by "actual fills vs. system fills"? Do you use the API?

All my orders are entered at market by an automated trading system. My ATS simply records the current price at the time the trade was placed, and I assume that any deviation is "slippage". Of course some (or all?) of the slippage I get is because of internet lag (the price moves before I get filled). If I used your strategy of placing limit orders I would probably need to use the API so I could cancel orders that were close but the system signal changed. As it is I have automated the FXTrade interface. I lack the java, c++ expertise at the moment. When the VB.NET API comes out of beta maybe I'll take a look at that, or just hire someone once I figure out exactly what I want. The API would give me a lot more flexibility and might reduce the slippage I receive or perceive (due to faster? quotes and faster? fills).

 

[phoenix555]

so if i read it correctly, Tripack, u're not using the API now, and in ur own words, u lack the java expertise now as well. If that is the case, may i ask then, how are u automating ur system right now?

 

[TriPack]

Quote from phoenix555:

so if i read it correctly, Tripack, u're not using the API now, and in ur own words, u lack the java expertise now as well. If that is the case, may i ask then, how are u automating ur system right now?

GUI automation.

 

[late apex]

Quote from jessieblue:

Never heard of van Tharp, I'll do a search

Quote from late apex:
20 on those majors is far too tight in my experience. Cable sneezes, and it's moved up or down more than that, many times every day. Why try to be a hero, a neuro-surgeon with ordinary kitchen tools (TA), when you can do very well, playing just a doctor on TV?

Sorry, didn't catch that analogy. If 20 is too tight, what do you reccomand? 50, 100? Remember, I'm a daytrader (for now anyway)

Quote from late apex:
Didn't have a chance to look at that book yet. But don't worry, I will. Is that where you got 20-pip stop loss from?

Making fun of noobies dude? That's just mean.
Yes, that's where I got the ideea. Come on man, don't knock it till you tried it.

Making fun of noobies? What would be the point of that? No, I never knowingly make fun of newbies. If I don't believe that what I have to say might be of some interest or help to someone here, I usually don't bother saying it. That ambiguous smiley technically stands for "cool", you know.

Van Tharp's first book is called "Trade Your Way to Financial Freedom". Packed with solid, essential contents for beginners.

"Why try to be a hero, a neuro-surgeon with ordinary kitchen tools (TA), when you can do very well, playing just a doctor on TV?" Kind of mixing metaphors here, I guess. IOW, why attempt to pick precision entry points with generally non-precision tools, such as Technical Analysis, when you can do very well, identifying and executing trades within high-probability, low-risk entry areas? "Playing just a doctor on TV" is merely a reference to a worn-out phrase from TV commercials where some usually has-been actor is earnestly hawking the latest pharmaceutical miracle: "I am not a doctor, but I play (played) one on TV" . Might be more of an American idiom... what's more American than mom, apple pie and celeb marketing / advertising? But I digress...

 

[late apex]

Quote from jimrockford:

No, jessieblue, you can't calculate an expectancy based on past performance of a trading strategy. This is because markets are statistically "nonstationary". The average behaviour of the market, or of any trading strategy, changes over time. The average behaviour you observe this week will be different from the behaviour you observe next week. The average behaviour you observe this year will be different from that you observe next year. It is not valid to calculate average behaviour over a time period, and then expect to encounter the same average in future time periods.

The market is different from a statistically stationary process, for example, flipping a fair coin or rolling dice. These stationary processes are the kind where you can calculate average behaviour and extrapolate to the future. The technique of averaging and extrapolating works, because you are always collecting data from the same unchanging process. Every time you flip the coin or roll the dice, it is the same coin or the same dice. Not so with the market. The market is always changing.

The illusion that markets are stationary is very dangerous, because it misleads inexperienced traders into thinking that they can make vast fortunes, when in fact, almost all of them lose. You will find lots of people who are trying to sell you something, by trying to convince you the market is stationary. You will also find lots of people who truly believe the market is stationary, because they don't know any better, and some of them will probably be very irritated by the suggestion that the market is not stationary.

Anyway, remember what the CFTC requires trading advisors to warn the public: "Past performance is no guarantee of future results."

If I didn't have a trading business to run, I'd gladly point out some of the critical flaws in your reasoning, your thinking process, over the next few days, weeks or however long that discussion might take. However, even if I were able to set aside the necessary time, unless you happen to have a PhD in statistics, an MBA in analytical finance and a couple of decades of experience playing with the quantitative nature of markets, such a discussion might not be fair to you. If nothing else, you might want to peruse acrary's journal, just for fun.

If, based on your beliefs, you are consistently profitable in your trading, more power to you. Then there's no real reason for you to alter those beliefs, is there?

 

[jessieblue]

Quote from late apex:

unless you happen to have a PhD in statistics, an MBA in analytical finance and a couple of decades of experience playing with the quantitative nature of markets, such a discussion might not be fair to you.

Oh, so that's why you were using those technical terms. So you are actually schooled in trading. I am specialised in psichology myself, but I also took economics courses.

You said you run a trading business. Does that involve coaching? I might be interested in that.

By the way, I knew you weren't making fun of a noob there. I was just kidding. jeez, you americans are uptight

 

[traderob]

Quote from late apex:

If I didn't have a trading business to run, I'd gladly point out some of the critical flaws in your reasoning, your thinking process, over the next few days, weeks or however long that discussion might take. However, even if I were able to set aside the necessary time, unless you happen to have a PhD in statistics, an MBA in analytical finance and a couple of decades of experience playing with the quantitative nature of markets, such a discussion might not be fair to you. If nothing else, you might want to peruse acrary's journal, just for fun.

I thought Jim's post was rather insightful, so I'd be interested in reading a discussion on this.. But maybe it should be moved to the trading forum.

 

[jimrockford]

Quote from roberk:

I thought Jim's post was rather insightful, so I'd be interested in reading a discussion on this.. But maybe it should be moved to the trading forum.

Thank you, roberk, but I would, for administrative purposes, like to highlight the fact that I was not trying to initiate a discussion between experienced traders regarding the statistics of price changes and trading strategy performance. I was, instead, trying to address the problem of how newbie traders are being defrauded by the FX industry. My posting was an amplification of the CFTC's official boilerplate warning that "Past performance is no guarantee of future results." I hope you will leave this discussion where it is, in the FX forums, because newbies desperately need some kind of warning to counterbalance all the FX industry propaganda designed to prey upon their gambling fevers.