Rol's Trading Journal

Quote from d08:

Not exactly. When there is high correlation selling, the positions would be initiated during the first phase, for example, between 9:30 and 11:00. The second phase, from 14:00 to 16:00 would also trigger signals. Randomizing entries loses time priority. So if you could take 20 trades, you would already have full exposure before the second phase could start, you are essentially betting that the trades from the first sell-off outperform those from the second sell-off in the following days - but in my experience, this often is not the case.
Monte carlo is better than nothing but the only bulletproof way to know is with intraday data.
More...
d08,
You are right about the time priority during the day. I often notice I get maxed out on positions during a market drop near the open. If the back tester takes a random sampling based on EOD data, then it would take some positions near the close on a down trending day that I would be unable to take. I always suspected paper trading was cherry picking the trades somehow, so this makes sense.

I don’t think this spells the death of my strategy. Sensing this shortfall in real trading, I have tried to compensate for this discrepancy between real life and paper trading. I keep tight reigns on the max number of positions I can take. Then each day, I gradually increase the number of allowed positions, until the market reverses. In effect, trading the account as a cash account at first, and then slowly adding margin buying power to the mix.

Another feature is that I have a max position size defined before I enter a trade. The first trade is 1/3 position size, and if triggered in subsequent days, the next trade is 2/3 position size. I know many will look at this as averaging down on losers, but profit factor and percent profitable goes up when I do this. It is different, I think, than just buying more every time a stock drops 25 cents or whatever. Also, since the subsequent trade has separate entry criteria, I view this as a separate strategy that just happens to be on the same symbol. Institutions do this all the time when entering positions, with a resulting average price paid. Being well capitalized is important, to be able to take advantage of trades when they present themselves.
 
I don't think it will kill profitability for your strategy either but I do think the equity curve won't be as smooth for the more difficult market settings (for long RTM).
I used to average down before and it also increased profit factor for me, I didn't really care what the rest said about that approach.
Your "follow-up position" approach is interesting, gave me a couple of ideas to try out as well.
I still suggest you test it all on intraday data, you will get a better overview of performance under different conditions. Plus you get peace of mind or "knowing where the bottom is".
 
Total Net Profit $11,106.67
(Per Share) $0.15
Gross Profit $25,294.54
Gross Loss ($14,187.86)
Profit Factor 1.78
Total Number of Trades 683
Percent Profitable 66.33%
Winning Trades 453
Losing Trades 230
Avg. Trade Net Profit $16.26
Avg. Winning Trade $55.84
Avg. Losing Trade ($61.69)
Ratio Avg. Win:Avg. Loss 0.91
Expectancy 0.27
Largest Winning Trade $429.64
Largest Losing Trade ($796.50)
Max. Consecutive Winning Trades 15
Max. Consecutive Losing Trades 9
Total Shares/Contracts Held 73815
Total Commission $1,716.01
Return on Initial Capital 23.21%
Annual Rate of Return 71.75%
Buy & Hold Return -1.34%
Return Retracement Ratio 3.47
Trading Period 3 Mths, 14 Dys, 5 Hrs, 50 Mins
Max. Equity Run-up(Daily) $11,987.32
Date of Max. Equity Run-up 4/15/2011 15:00
Max. Drawdown(Daily)
Value ($3,831.77)
Date 2/23/2011 15:00
as % of Initial Capital 8.01%
Max. Trade Drawdown ($1,511.50)

The CYH loss this week was the largest losing trade for the year. With the 10% max DD I now have in place, it would have been $300 instead of $800. Small position size is a form of stop loss, because it can also keep your losses small relative to account size. It is a reality that with rtm trading, there are going to be occasional "big" losses. It helps to have an action plan in place ahead of time. With auto trading it really is an "if, then" process. If the stock price does this, then my strategy will do that, letting the chips fall where they may. I think the problem some have is that they over leverage and scale in too quickly. In addition to that, I have not found it to work well on the short side, but that has just been my experience. Another edge I believe auto trading has is that markets can wear you down with long periods of relative calm, followed with sudden price moves. A computer algorithm has no loss of attention space.

I have heard that a reversion strategy is easier to code than a trend following strategy. I am not even sure how to approach a trend following strategy other than open > high of one bar ago, with increase volume. I am fairly certain there would be a lot of false signals.

Putting on enough different trades in a basket of stocks ensures that your equity still works for you. By scaling gradually into stocks over the past week, I had the BP to take a few profitable overnight and day trades today as the overall market reversed higher.

The profit factor so far for the year is comparable to paper trading, and the percent profitable is eerily at 66%. I would be pleased to maintain an annual rate of return over 70%. It would be hard to dispute that the strategy has no edge in its stock selection when the buy and hold return for the same period is
-1.34%.

My unrealized loss currently is ($398.85). Subtracting this from $11,106.67 gives a current net profit of $10,707.82 for the year.

attachment.php
 

Attachments

Quote from elitetradesman:

Looks great.

One question - when you hold overnight, do you reduce your position size compared to intraday trading?
More...

All positions currently start out at 100 shares, with the option to add 200 shares later. There is no distinction between intraday and overnight holdings. It has been my experience that holding overnight works more to one's advantage than against (in stocks that is), provided they are trading with an advantage. Stocks often begin their reversals the day after an exhaustion of sellers in the prior session, such as displayed with the recent gap up HOGS trade I posted. I view the game time from entry to exit as a continuum, with "time outs" between sessions.
 
Quote from Rol:

All positions currently start out at 100 shares, with the option to add 200 shares later. There is no distinction between intraday and overnight holdings. It has been my experience that holding overnight works more to one's advantage than against (in stocks that is), provided they are trading with an advantage. Stocks often begin their reversals the day after an exhaustion of sellers in the prior session, such as displayed with the recent gap up HOGS trade I posted. I view the game time from entry to exit as a continuum, with "time outs" between sessions.
More...

That's true, it's probably because some traders sell at close fearing overnight gaps; guess the short side is more risk tolerating when it comes to that.
Actually there is small edge in just holding overnight positions without any other strategy, naturally that edge isn't good enough to be tradable.
What is your average exposure as percentage of equity?
 
Quote from d08:

What is your average exposure as percentage of equity?
More...

TS does not give that number directly that I am aware of. The % time in market (Time in market/Trading Period) since going live 9/16/10 is about 63%. I am trying to come up with better ways of using margin, but as my equity increases, it seems I am becoming more risk adverse. I start becoming concerned as soon as I see my system begin using margin BP. I really feel like a reluctant buyer most of the time (until the trade is going my way). :D

My current exposure (Cost of Positions/Equity) is 80% after today's close. I would say 50-75% is typical at the close of a day.
It just depends on what the market is doing. Intraday I may be using more margin, as there are frequently trades opening and closing during the day.
 
I thought I would give an update on what I have been doing lately. I made some changes to the auto strategy entries. I made it more scaling in nature than it had been and more automated. At the end of the day the strat looks at how many open positions there are. Then for the next trading day, allows 5 addition positions to be taken. So for example if I had 10 position, I could take on a max of 15 total positions the next trading session. I anticipate this would prevent me from maxing out on available positions in the first signs of a broad market downturn, which is a shortfall of my current strategy. Today for example, I had 5 new entries, but there were 6 exits at the close with a realized P/L of $261.63. Unrealized P/L is ($806.45). There is a constant rotation of stocks in my portfolio. Tomorrow I can add 5 more positions if entry signals are given.

This should work well on typical daily market moves, and also allow me to ratchet up margin BP in a controlled manner during extreme market sell offs. I expect this would make more use of 4X daily buying power, during a black-swan type event, so I will have to monitor it closely, possibly even selling some stocks to meet margin. I don’t know what other’s experience has been with their brokers but I have noticed that TS does not hassle me about using 4X overnight BP once or twice a year if I am diversified, and I bring it back under 2X by the end of the next trading day.

I took an 183 share entry in Wells Fargo Monday in an IRA account. It came close to triggering an entry from my strategy, but did not. It had been down 8 days in a row, which on a decent company like Wells Fargo, to me qualifies as crazy oversold. My entry was at the second arrow, where I felt a double bottom was forming, creating a support level. I also knew earning would be released before market tomorrow, so I felt the selling was simply manipulation to get the stock ready for an earnings bounce. Banks somehow manage to spin their earnings in an upbeat way, and the market often reacts positively. We will see how it does tomorrow.

attachment.php
 

Attachments

Total Net Profit $11,852.06
(Per Share) $0.16
Gross Profit $26,537.04
Gross Loss ($14,684.98)
Profit Factor 1.81
Total Number of Trades 703
Percent Profitable 66.15%
Winning Trades 465
Losing Trades 238
Avg. Trade Net Profit $16.86
Avg. Winning Trade $57.07
Avg. Losing Trade ($61.70)
Ratio Avg. Win:Avg. Loss 0.92
Expectancy 0.27
Largest Winning Trade $429.64
Largest Losing Trade ($796.50)
Max. Consecutive Winning Trades 15
Max. Consecutive Losing Trades 9
Total Shares/Contracts Held 76215
Total Commission $1,764.01
Return on Initial Capital 24.77%
Annual Rate of Return 72.65%
Buy & Hold Return -1.36%
Return Retracement Ratio 3.52
Trading Period 3 Mths, 19 Dys, 5 Hrs, 50 Mins
Max. Equity Run-up(Daily) $12,732.71
Date of Max. Equity Run-up 4/20/2011 15:00
Max. Drawdown(Daily)
Value ($3,831.77)
Date 2/23/2011 15:00
as % of Initial Capital 8.01%
Max. Trade Drawdown ($1,511.50)

It was a surprisingly slow week for me, other than the gap up on Wednesday, where I then exited several positions for profit. I actually expected the market to head lower, so I was not using much margin BP at all this week, wanting to save it if needed in a bigger sell off. That is somewhat how my strategy behaves. It holds back because market direction is never certain, and so tries to capitalize whichever way the market turns. As a result, I will most likely be missing huge short-term gains, while missing monstrous losses as well. I am shooting for a "low risk" strategy that I can observe hands off.

At some point, I think I would like to trade remote through a prop firm to leverage up when I have more entry signals than I can currently take. However, I want to give my strategy time to operate live to be more certain it can survive a wide variety of market conditions. I am still looking into using Tradelink to bridge Tradestation to a prop firm.

Unrealized P/L ($186.90)

attachment.php
 

Attachments

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