It has been almost a year since iBC opened PDS for subscription. You learn a lot about yourself when trading a system, and many of the subscribers have shared with me what they have learned. These discussions have been very helpful as I prepare for the system’s second year in subscription.
Without a doubt, the hardest part of system trading is trading the system. Because of this fact, it is important to make the trading part of it as simple as possible. All rules and methodology should be very clear and easily implementable. The psychology part of system trading is hard enough. We don’t want complex rules or order entry requirements to make things even harder.
To that end, I am making some changes in how I recommend subscribers trade the system. These changes will be reflected in PDS site literature over the next week or so, but I am going to outline them here.
- All models will use all available cash, any given day, assuming there are enough picks to use the cash. Previous rules limited new buys to 5 per day in the 1% risk models and 3 per day in the 2% risk models. Limiting new buys per day was done to decrease the drawdowns. This decrease was slight, and I’m inclined to believe that the inconsistency that may result from being unclear exactly how to trade the system may hurt performance more than the expected reduction in drawdowns can improve it.
- With the exception of the 1% risk models, the system will no longer be allowed to take multiple positions in the same security. While I have always advocated not taking multiple positions in the same security, I continued to allow the system to do so, for a variety of reasons (high win % being the primary reason). Then came ALKS. Having 2 positions in ALKS using a 2% risk model would have been devastating. I will continue to allow the 1% risk models to take multiple positions in the same stock, as two positions at 1% risk equal one position at 2% risk.
These changes should simplify the mechanical aspect of trading the system: Buy as many stocks as you have cash to purchase, and don’t take multiple positions in the same stock unless you are trading at 1% risk.
Year-To-Date Results:
Summary of Results:
The top part of the graph with the gray background reflects the results as I’ve been tabulating them since inception. The ALKS trade knocked the system backwards so that it is not currently beating the benchmark. The silver lining for me is that the 2% Risk, 3ATR model is less than 1% shy of its benchmark, even after suffering a large loss with ALKS.
To be clear, the YTD results start calculating at the open of the first trading day of the year.
The lower part of the graph with the white background reflects the historical results with some added “what ifs,” which include the changes that I discussed earlier.
My favorite “what if” is what if the system wouldn’t have traded ALKS? Had it not bought ALKS, PDS would be near doubling the performance of the benchmark, at 15.81% YTD. Even better, what if we figured no ALKS and .005/share for commission (consistent with Interactive Broker’s commission costs)? Now we are at 21.15% YTD.
The bottom three rows demonstrate YTD performance with the simplified rules of no multiple entries and no limit on new buys per day. While these changes improved performance so far this year, during previous years these factors have slightly reduced performance. Again, I think the added simplicity, leading to greater consistency, is what is important.
Looking Back and Looking Ahead…
Despite the recent setback, I am satisfied that real-time PDS performance is consistent with what I expected from historical backtesting. November 18th marks the year anniversary, and I am very excited to see this system moving into its 2nd year of subscription.
Woodshedder,
Congrats on making it past the first year of real time trading with the Power Dip System. Most traders, and definitely most stock market advice sites, don’t make it past the one year mark, so I think that it is a real testament to your system. Moreover, the fact that system was profitable over that period I think shows that it wasn’t curve fit and is realistic (of course, we already knew this). Honestly, if it had return like 70% or something in the first year, I would probably be MORE alarmed than the realistic result.
Of course, the historical return for some of the Power Dip Models is 70%, so for all you lurkers out there, subscribe to the Power Dip System now – before and not after it happens!
I wholeheartedly agree with the move to limit the positions in the same stocks. I was actually holding 3 positions of HDY, and for a moment there thought I was going to be finished – thankfully it bounced up 12% in one day and my killer loss was limited to just a normal loss. I think I have a pretty high risk tolerance, but sad to say I did cave in and sold one of positions early, missing some of that bounce.
So I would go one step further and limit the 1% models to only 2 max positions in the same stock (that’s what I’ll be doing at least).
As for limiting the number buys/day, it seems like that would be easy enough a rule to follow to me, as all the buys are MOO (but I’ll admit I may not be typical). You could keep the old models up on the website so that if subscribers wanted to do that, it would still be an option.
Anyway, thanks for all the hard work this year. I look forward to 2011, which I have a feeling will be the Year of the Dip!
Data, thanks for your kind words!
I am going to make some changes to the models, likely by getting rid of some of them altogether, but I will definitely highlight the effect that limiting number of new buys a day can have.
I believe that it is important to simplify, simplify, simplify, and I want the monthly reports to represent the most simple ways to trade the system.
Dude, you have some serious cojones to hold three positions in a stock like HDY. That is some serious discipline. My prediction is that this discipline will lead to great success for you in system trading.
Heh, honestly I didn’t think about HDY when I was placing the order(s) – just on autopilot. So maybe it’s just blissful ignorance on my part :-).
Plus, having zillions of backtests to look at seems to make it easier for some reason. You know what you’re doing is not unprecedented and more than likely you’ll be all right.
Good luck with the site – glad your looking out for the subscribers with the updates. I would probably succumb to the temptation of just publishing the best performing models with the lowest commissions – so good on you for making it realistic/conservative.
Well done!
Great job Wood!
I’ll be a continuous subscriber.
With regard to the models, I’ve been going back and forth between the 1and 2% risk with the 3 ATR stop.
What do you use now?
Hawaii, I’m using 2% risk, 10% stop, right now. But I vacillate between that and 2% risk, 3ATR position-sizing. Lately, volatility has dropped so low that 2% risk, 3ATR sizing has us taking some very large positions with tight stops. During a raging bull (like now) this works, very, very well. These large positions with tight stops is what allows the 2% risk 3 ATR model to beat out the 2% risk 10% stop model. I still don’t have the stomach for some of the positions, so I go back and forth between the two models. Remember, it is somewhat counter-intuitive. When the market is bearish, trading beneath the 50 or 200, trade the 2% risk, 10% stop. When it is bullish, trade the ATR position-sizing model.
Wood,
I’m not getting how 3ATR is a tight stop. Isn’t usually about 10 to 12%?
HA, usually they are around 10-12%. But when volatility drops, like now, they get lower. Look at FRX and SMG, two recent picks. Look at the ATR%. Multiply that by 3 and you see that the stops on these would be really tight and therefore the positions large.
Got it!
Yeah, I think I’d rather err on the side that gets me a smaller position, so I’ll probably go with the 10% model if the position gets too big. Regardless of the model, I’ve been keeping my positions at 1% of my entire port, which includes trades other than PD’s just to prevent a big ALKS like wipe out.
Investing in Biotech is like fighting Bruce Lee in a rattlesnake pit with Tigers circling around…you never know when and who will deal you the death blow….the folks at CDER and CBER are more inclined to protect their pensions by returning a verdict of “need more data” than to put their stamp of acceptance on a drug, even if it is proven safe and effective. And heaven forbid a monkey dies or someone keels over from a heart attack, related or not…. I do however have fond memories of Genentech’s solid returns…
I am confused about the systems ability to have multiple positions. And it looks like it has multiple entries on the same day at different prices.
Hi Red, can you direct me to where you are seeing it have multiple entries on the same day at different prices?
Here’s how it works. ABC stock is selected for purchase on Monday’s open. It is purchased.
After Monday’s close, ABC stock still meets the entry criteria, and it is again selected for purchase on Tuesday’s open. At this point in time, it is up to the trader to decide whether they want a double position in ABC stock.
There are not a lot of stocks that meet the entry criteria on back to back days. It happens from time to time. Even more rare is the same stock is selected for purchase 3 days in a row. Again, it would be up to the trader to decide whether he wants 3 positions in the same stock.
In terms of tracking the system, the 2% risk models will not be allowed to have multiple concurrent positions in the same stock. 1% risk models will.
I hope this helps. Let me know.
Sorry for the confusion Wood. I was looking at -> http://st.ibankcoin.com/membership_benefits/automated_trading_strategy.php and saw the same symbol listed and thought the sell date was the buy date.
Thank you for the clarification.
One thing that might be worth your time is to do a study on the frequency of large gaps down in stocks whose price (unadjusted of course) is below $10 and $5 vs stocks whose price is above that. I did a similar study a while back that I could probably dust off and send you. It is interesting to see the differences. This is in reference to the recent gap down you had. As someone who has traded a similar system, I know how tough it can be to have hard earned gains wiped out by a single loss. And of course, how hard it is to account for that in a simulated trading system vs an actual trading system, but I am confident you will get it all worked out.
Cheers.