Trade Performance by Weekday - What's your data?

L

Laissez Faire

Hi all,

Curious if you ever checked your results/statistics sorted by weekday. Any patterns...?

Not long ago I read one of the Wizards books by Schwager. There's an interview with Mark D. Cook. Interesting fellow, by the way.

He had long term data which clearly showed that Friday was his worst day of the week - while Tuesday was his best. He reasoned that after a weekend, he needed a day to warm up and by Friday he was getting tired or reckless.

Based on this findings, he stopped trading Fridays and traded more aggressively on Tuesdays than the rest of the week. Allegedly, his performance improved strongly.

If we add more detail - one could look at hour per day or split the day in two for intraday traders.

I know from my own research that there are statistical differences based on day of week, so I'm not surprised by this.

Anyway, no reason to share any details. Just curious if anyone ever paid attention to this and found any interesting results.
 
Lately, I've had my worst days on Thursdays for some reason. Even though I had a strong long bias prior to the Open today - I still managed to take a few hits on longs (too tights or too wide stops) and had to fight myself back to the green barely breaking even on the day.
 
I ran an analysis of my past performance broken down by hours of day. I trade macro and generally performed the worst during Asian market hours. Generally the moves are very subtle and I often looked for entries when there really was not much going on. Most moves were quickly faded. Still could never really adopt to it.

In the past few years I generally start trading from when traders come into the office in London. I am often done 2 hours after US market open. Those 10 hours are my bread and butter.
 
Both Performance by Day of Week and by Hour (or half hour) are useful.

Performance by Day is a bit of a novelty (to me) however, in that
many shortened sessions (or no trading) occur on Mondays and Fridays.
And then you have personal situations that occur on random days. Just note, that
it is easy to get lopsided data points. Plus, there is no "market environment" criteria... maybe you nail "narrow movement" days, but on "average movement" days you aren't so hot. None of that appears in default PbD analysis.

I prefer Performance by Hour (or half hour)
Only through PbH analysis did I quantify that the last hour of the trading day is my worst... clearly the manifestation of fatigue. And the first 2.5 hours I shine! Also of note, for me, ETH hours AFTER Europe is open are just fine for me... this was VERY surprising since volume is an important input into my trading. Product selection does matter however.

Performance by Day+Hour requires work. Sierrachart has PbD and PbH as builtins, but not in combination.
 
You must log in or register to reply here.
Back
Top