Daytrading! Part Deux

Quote from Pivotas:

JTK.... this is what I've learned from looking at your charts/trades.

1. Most of your profitable trades occur when you enter after a minimum of 2 counter trend candles have completed.

2. Fewer losing trades occur when you respect the relationship of the Cyan to the Magenta moving averages. ie: only entering long when cyan is above, only entering short when below.

3. Your stopsloss is unrealistically tight. I suggest you go over all the instances in which you were stopped out of a trade that otherwise would have been profitable and determine what stoploss setting would have kept you in those trades. If you have confidance in your entry decision, you need to demonstrate that in the stop you set. If you are not confidant in the entry then you should not make the trade. I assure you the combination of un-necessary losses from a 1 point stop loss added to the profits you then denied yourself are greater than any loss you would incur with a stop of @2.5 pts. At least you need to allow the entry bar to complete then set a stop below it's low.

If you follow the above you should see a dramatic reduction in the number of losing trades and as a consequance get your account moving in the right direction.

Best of luck,

Pivotas
More...


Pivotas,

Thank you for your comments and suggestions. You advice on watching the relationship between the cyan and magenta moving average lines is a very good one, as it is easy to "anticipate" a long or short and make an entry before an actual trigger.

My stop loss per my trading plan is 1.25 points (5 ticks on the ES), though lately I have been using 1 point (4 ticks). I don't think there is an unreasonable amount of early stops hit, as I try to enter only one or two ticks above the magenta line (18MA). This leaves 2 or 3 ticks below the MA before getting stopped out. The times I have been stopped out have been mostly due to entering too early, before a proper signal has been given.

Thanks again!
 
JTK... To illustrate my point the below chart shows your 510 tick es chart with an 18 period ma. The indicator shows the high and low of each bar in relation to the ma. The white circles indicate those occasions in which the bar extreme corrected at least 1 es point over the 18 period ma before the trend continued. On these occasions, with your stops, you would have been taken out and missed the next impulse. When using discretion to enter a trade one cannot expect perfection on the timing and your stop system needs to take that into consideration. If you defined your stops in relation to the 18 period ma , ie: 18 period ma +/- 1.25 , rather than your entryprice, you would imho do better.
 

Attachments

Pivotas, thanks for the example chart and I understand what you're saying with regards to entry points.

Part of the reasoning for my tight stops is money management. a 1.25 point stop-loss is a -$125 loss on two contracts, which is what my trading plan calls for. On a winning trade, I pare out one contract on a 5-point gain and increase my stop loss to entry +1 on the remaining contract, for a total gain of $300 (assuming the price moves no further). This gain would cover 2.4 losing trades.

The other reason for tight stops is psychological... I find it very tough to "pull the trigger" on the next trade signal after having lost big amounts, and sometimes revenge trading comes into play. The tight stops allow me to risk less with each trade, which helps alleviate these flaws.

At some point in the future, after I have gained some confidence back (note the title of my journal is "Part Deux", this is the second time around for me), I can see myself opening up the stops more.

Thanks again for your advice, it makes sense but maybe not right for me at this time.
 
JTK.... last point. With discretionary entry, which is what I do, you will always "sometimes be early". Because of this there is an element of uncertainty with each entry and each entryprice based stoploss, you will not know until after your stop is hit if you were just early or if the trend is turning and your stop was correctly placed. If the correctness of the entry point is uncertain with each trade, then a stoploss of 1.25 is also uncertain, and may or may not be correct for that trade. In the chart I posted you will find certainty. With only one exception, when the market was in trend mode, the corrections never went to 1.25 points on the other side of the 18 period ma. By basing your stops on this ma, you will be offsetting the uncertainty of the timing of your entry with the certainty that if you are correct on your decision to enter the trade, you can be certain you will not be stopped out incorrectly and will have a profitable trade. It's all about probability and how do you tilt it in you favor.

Exposing your self to un-necessary stoplosses can only lead to frustration and effect the objectivity of your entry decisions. By basing the stop loss on the 18 period+/- ma you remove objectivity and emotion from your exit decisions and have a fixed and definable aspect in your trading that you can adjust with the conditions.

Most if not all of the important lessons come at great expense.

As you trade during the day, run a parallel system on paper using the stops as I suggest and you'll have a point of comparison.

Good luck,

Pivotas
 
Thanks again Pivotas, I'll paper trade both my system and one with the wider stops over 2-3 days and post the results in this journal.
 
Thursday, April 3, 2008

I'm late getting this entry posted. I had a miserable trading day on Thursday, with a loss of around -11 points on the ES. Needless to say, this wiped out several days of profit and left me in a pretty depressed mood. The only "upside" is that my paper-trading simulation that I do after market hours every day also had it's worst "loss", with -9 points lost. So of all days to closely match my trading strategy, I picked that one....

I didn't trade today (Friday, 4/4) to review my strategy and see what went wrong, if anything. Sometimes these things happen. I am trying to compile a spreadsheet of all my paper trades, and the underlying indicator conditions at the time of entry (rising/falling MA, stochastics, etc.) and then sort the list and see under what conditions the most profitable trades have been made. I may or may not trade on Monday depending on how far along I am in the process.

One thing I did note is that over time I seem to have deviated a bit from my original strategy of only making trades in the same direction of the 5 minute 18MA (or trades on the opposite direction in the case of a >7 point difference between price and the 5min 18MA). Over time this seems to have morphed into a "go with 510T 18MA direction no matter what". This warrants study at least, since it's always best to trade with the wind at your back. I noticed that week by week my trading simulations were earning less, perhaps this is the reason.

I'm also going to look at perhaps trading off the 5 minute chart using the 510 tick chart as a trigger for entry. This would allow wider stops as Pivotas has suggested.

Anyway, looking forward to better days ahead.
 
JTK... I converted your strategy into EL code to better understand some of the problems you were having. I optimized for profit, loss,breakeven, and the 2 ma's you were using for trend definition and entry. Below are pix showing the equity curve using the optimized settings for 1 contract only and a chart for the trades that would have been made for Thursday. The code is pretty basic but might be helpful as you try to get on top of the problems you are experiencing. Send me a PM with your email and I'll send it to you.
 

Attachments

Pivotas, I sent you a PM with my email address.

Although it is difficult to read, I've attached a chart showing Thursday's trades.

One thing I noticed is the short entry made on your chart at 12:36pm... at the time the 510 tick 18MA was uptrending, so per my plan the trade would not have been made. Was the trade triggered by the EL code?

I appreciate you taking the time to write out the EL and charts, it helps a great deal.
 

Attachments

You must log in or register to reply here.
Back
Top