The Cost of Second Guessing Trading Strategies (Why You Need A Written Trading Plan and Log)
Last Updated on January 17, 2023
During my twenty years of trading, I have learned that the best recipe for success is to follow and execute my trading plan. The cost of second-guessing my written trading strategies shows why you need a trading plan.
In this article, I provide hard facts about how my second-guessing and discretionary trading reduce my profits.
Why I keep a trading journal
I keep a detailed log of all my trading activities, both day trading, swing trading, and long-term positions. I believe this is the only way I can improve my trading. Please read the trading journal example.
However, doing all this “extra” work is not precisely the most exciting job on this planet, but I believe by far the most important I can do to improve my results. The more parameters I put in, the more exciting testing I can do some months later!
Doing the records might be boring, but I enjoy drinking coffee and eating cheesecake while studying my trade records. I can see where I go wrong.
One of the things I do is to record P/L throughout the day, even if I have already realized the position, to improve my exits. Specifically, I record P/L at six different time intervals before the close. I later check the results against some other parameters.
Based on all the parameters I use, I can track every trade pretty well and make all my trading as mechanical as possible. I’m not very good at discretionary trading. Hence I try to do it entirely mechanically.
Second-guessing my trading plan backfires – always
But far too often I second-guess my strategies. Unfortunately, all my second-guessing deteriorates my P/L. My decisions to overrun my strategies do not create any value whatsoever.
Why do I second guess (my quite good strategies)? Because it’s tough to detach from money when having a terrible day or a good day. On bad days I start hoping for the better, and on good days I’m afraid of giving away money. Completely irrational, and I know it. But still very hard to put emotions entirely on the sidelines.
Because I keep records, I can calculate quite precisely how much my second-guessing is costing me. Below is a chart showing my accumulated profits since October 2015:
The pink line shows the completely mechanical results, while the blue line is my actual trading. The difference is enormous, both relatively and money-wise.
So to refrain from more second-guessing, I have made a big poster on my wall showing all my five rules. Yes, five simple rules. Also, instead of sitting in front of the computer, I’m better off walking the dog, going to the pub, or chasing girls. The more I stare at the screen, the more stupid mistakes I make!
On a side note: My day trading has not been good over the last two years. One reason is that I’m being forced to trade less because of new non-display fees from NYSE (I only trade Nasdaq stocks now), and the second reason is that I’m focusing more on long-term positions. Long-term positions give me more peace of mind because it involves fewer decisions.
The actual (blue) equity curve is fascinating. What’s the Sharpe ratio? (>4?) Has it been this smooth for many years or is it market-cycle specific? How many strategies, instruments, and over which time-frames?
Why not just compound this curve with leverage into 100s of $millions in less than 10 years?
Thanks from a fellow trader (systematic global futures).
Hi,
I believe the Shrpe ratio is very high, I have never calculated it. I might have a look at it later.
My equity curve is pretty much like this since 2003. However, I’m nowwhere near as profitable as 2004-2010. I only trade stocks and have about 6 strategies. Unfortunately, it’s impossible to compound it.
Is it hard to compund because of capacity constraints (illiquid stocks) or something else? Here’s what I had in mind:
Take 1/3 of your net worth as trading capital (TC) and trade this equity curve geared (levered) for max drawdown of 30% of TC. Stop loss (unplug everything) at drawdown of 50% of TC (or currently 1/6 of net worth). Assuming a sharpe of 4, you could compound your TC at >100%/yr (4 x 30%vol). Regardless of starting TC, it would go into $millions in just a few years.
If you have my address and are interested, pls shoot me an email to discuss.
Hi,
I appreciate your help, but this is liquidity constraints. I am already trading most of my stocks at what I believe is max size. Any more size and I will have problems executing and/or moving market. Even 2 mill avg stocks are sometimes very hard to get a fill in right now.
Got it. May be worth considering to exchange some sharpe (from 4 down to, say 2) for higher capacity (more liquid universe, maybe add futures & fx, longer time-frames), diversification and robustness (less broken strats). Volatility is life itself — something to embrace, not fear. 🙂
Do you mind describing your backtest and trading platform (software, data, execution, etc.)?
Either way, your results are fascinating. Good luck!
I’m using Sterlingtrader (daytrading) for execution and Excel for backtesting. Nothing fancy. Quotes from Yahoo.
Good luck to you as well!
Hi Oddmund,
Do you trade at the closing auction on Nasdaq? Do you know if Yahoo data consider the closing auction as the daily close?
Im very concerned about market data quality, Im looking for daytrade strategies. Im using IQFeed for intraday data but it is really messy with some absurd spikes.
What are those non-display fees from NYSE, could you elaborate a little more?
Regards
No, I’m not using any auction. But unless you are trading very liquid stocks I assume it’s hard to trade any size. You must exit before close.
The non-display fees applies to NYSE and AMEX, but it seems it’s not all firms being charged at current time. If I remember correctly, it is 3000 for AMEX and 6000 USD for NYSE (per month!). Horrible expensive. Non-display is if you have for example Sterlingtrader enabled for API, and at the same time you have market data. That automatically makes you liable to non-display fees. You can go arounf by using market data from eSignal but still send orders through Sterling, but not sure how long this will last.
I had not heard about it. You are right, its horrible and expensive. I dont know whats the point with it.
Now Im puzzled, how are you able to backtest daytrade strategies with yahoo quotes and without trading the closing price?
I download my own data while trading at regular intervals.