Someone tell me about the realities of intraday trading

C

Corso482

So I've been swingtrading off and on for years now with a fair degree of success, and have been researching some intraday strategies lately.

Specifically, I'm curious about the viablity of one strategy, and I was hoping somone with daytrading experience can chime in about the real life viability of it.

The strategy runs off 2 minute charts, and I'm only feeding it stocks that have at least 2mm shares daily volume. The average profit per trade ranges from only a few cents to maybe 20 cents. I'm curious whether slippage would kill something like this in real life. What other concerns might make this impractical in real life?

I'm thinking because of the very short term trading and the long list of stocks that need to be monitored, this would have to be automated.

Thanks.
 
Quote from Corso482:

So I've been swingtrading off and on for years now with a fair degree of success, and have been researching some intraday strategies lately.

Specifically, I'm curious about the viablity of one strategy, and I was hoping somone with daytrading experience can chime in about the real life viability of it.

The strategy runs off 2 minute charts, and I'm only feeding it stocks that have at least 2mm shares daily volume. The average profit per trade ranges from only a few cents to maybe 20 cents. I'm curious whether slippage would kill something like this in real life. What other concerns might make this impractical in real life?

I'm thinking because of the very short term trading and the long list of stocks that need to be monitored, this would have to be automated.

Thanks.
More...

If you backtest it, the most conservative approach is to use the most adverse exit and entry of your bars. This should account for slippage issues and if it's profitable thiat way, then it should be profitable in real trading.

Cash
 
Corso - I run automated systems on TradeStation. Assuming you are talking about entering via market orders and taking your stops at the market, I think you are better off by taking longer term daytrades, i.e., those with larger profit expectation. I use a volume filter on each bar to make sure that I don't get hit with unexpected slippage on entry. I keep records of my slippage vs. my system and found that I average slippage of .02-.03 each way. Best of luck. Thunderous.
 
corso482 -

For stocks you need to include about 5c of slippage per round trip. Obviously, the shorter the time duration you are analyzing the more important your slippage model becomes.
Personally, if you have found some success with your swing-trading I would try to implement other strategies using that same timeframe.

Good luck!
 
Quote from cashonly:

If you backtest it, the most conservative approach is to use the most adverse exit and entry of your bars.
More...

Not sure what you mean by this.
 
Quote from Thunderous:

Corso - I run automated systems on TradeStation. Assuming you are talking about entering via market orders and taking your stops at the market, I think you are better off by taking longer term daytrades, i.e., those with larger profit expectation. I use a volume filter on each bar to make sure that I don't get hit with unexpected slippage on entry. I keep records of my slippage vs. my system and found that I average slippage of .02-.03 each way. Best of luck. Thunderous.
More...

What are you trading?
 
This was sent to me via PM and I'm answering here to also help others who may have a similar question.


wrote on 08-01-06 01:50 PM:
Hi Cash I just saw your post about adverse entry and exit, I have a question if I may. I trade stock index futures and currency futures. Let's say I have a set up that works very well I want to back test, but the R to R is sometimes 4 risk for 1 reward. I want to see how accurate it is over time to see if it is worth risk so much to make so little a high percentage of times.

What is a good time period for a valid back test? 1 year, 5, etc?

To you point about adverse.......would I shave a tick off the top and bottom after doing the backtest to account for a possible no fill on a limit at the top of the candle or a stop loss being skidded on the execution?
More...

Time period is tough... it depends on your timeframe for holding trades... if it's an intraday thing, I'm usually doing a backtest every week for the past 90 days. If it's more of a swing trade, then you would backtest for a longer time. I also like to look for "similar" times in market history. Say for example, for the last 90 days, the market has dropped 3% in a relatively smooth fashion (ie determined by using linear regression), then I like to look for a similar time period in market history and see how things worked then.

But keep in mind not to go too far back for intraday testing, cause when we switched to pennies, the intraday activity and slippage/profit was different than it was for sixteenths, which was different than it was for eighths. When it was eighths, you could use a strategy that only looked for 1 tick and if your RT commis was 2c or 3c, that would work nicely, but now with ticks being a penny, it's a different story.

Regarding adverse, if you go long a stock, the most conservative backtest would have you buying the high of the bar you are going long in and when you cover, selling the low of the bar you are covering in and vice-versa for shorts.

HTH,

Cash
 
Corso - I trade stocks and futures. The slippage refers to stocks. I generally find little to no slippage in futures. Thunderous
 
It depends on whether you are following the crowd or not. If you're trying to buy a breakout signal that 1000 other people are buying, you'll get a lot of slippage. If you're selling a signal that 1000 others are buying, you might actually get improvement instead.
 
You must log in or register to reply here.
Back
Top