I am testing an intraday strategy that performs the best if all I set is a stop loss, then allow the strategy to only exit on market close (or some preset time). Obviously the strategy works by finding the right direction to get in, then ride the trend. However, most "trend following" exit strategies seem to degrade performance significantly. I have tried specifying different set times other than market close, such as a few minutes before.. 30 min before etc.. In an effort to make sure it was not something specific to the close itself (such as bad data). All tests regardless of which pre-set time appears to perform great. Any thoughts on ways to manage the trade better? Exit strategies to try?
Intraday trends can be very noisy, therefore, if you use tight stops you will degrade performance. I'm a big believer of letting time work in your favor yet if you are daytrading you are limited to the close of the day. Have you tried a moving average cross or perhaps a certain number of green or red bar below a moving average of choice. I don't think there is a perfect answer to your question, depends on the instrument and the price action at hand.
I agree unless you have a super awesome entry and a great exit strategy you will do much better with wider stops. Stops should be based on the volatility of the market + the time frame you are trading. I to also use red and green bars to weigh out a potential position. I like to trade in the line of least resistance as Jessie livermore would say. I have a 4 bar perpetual direction finder based on 3 of the last four bars being either red or green based on the perceived direction and the last bar breaking the support or resistance of the previous three. The trick is to know how much the setup is worth. The last thing you want is to be in the money and have the trade turn on you because you didn't do the most important aspect of trading (taking your profits).
you are sort of trading on the daily bars in a sense... like if you were trading manually you would think "if I can get in at this price, most likely it will close higher". You might just look at the daily bars and use that to guesstimate an exit point..
We discussed some of this, especially I did, in the crude oil thread, CL Redux, the last while. I don't have time right now to find all the posts, but you can look through my posts. Yesterday I had a short position in CL near the first hour high of the day, HOD, at that time. And I caught it within about 10 ticks of the top before a selloff. And I was up $700 at one point as I let the trade develop. And it was up $600-700 for quite a while. But I let it go too long and it retraced the whole move to b/e and I made nothing. And then it broke higher and I was busy doing something else and I missed that too. Oh, well. A lot of this has to do with what you're trading, when you're trading it, what the action is that day, whether you're trading with the trend or against it, your entry, and a lot of other factors. It's good to let a trade develop obviously to maximize profits, however it's also important to take significant profits once they are there before they disappear maybe. As they always say, "No one ever went broke taking a profit."
truthfully in my belief, most of the strategies' expected value are a function of time, where as time goes by the expected value of your edge increase. So basically as not putting in exiting criteria (except for MOC) you are increasing your trade time, which overall increases your expectancy (assuming it's a function of time)
The big problem with intraday strategies is your expectancy is already constrained by the short time frame. If you further require an event such as crossing below some moving average, or trailing stop etc.. You have given back a nice chunk of the move on already low "margins"...
dumbest statement ever.....People go broke all the time "taking a profit".... Why? because the vast majority of losing novice traders have losing trades that are much much larger than their winners...any real trader would already know that