I had the feeling that my cryptocurrency portfolio (which is less than 10% Bitcoin) was swinging in weekly cycles and that there are certain days of the week better for selling and certain days better for buying.
In this post I am going to investigate if there is evidence of weekly market cycles in Bitcoin prices as of 2018. This is no trading advice, just an analysis based on historical data.
It depends on the general market trend. In an uptrend: prices are low on mondays, high on saturdays. In a downtrend: prices are low on fridays, high on mondays.
I started with the daily High/Low/Volume data from coinmarketcap in the timeframe from the first week of January 2018 until the last week of April 2018. I was using excel for the whole analysis. The time is in UTC which is approximately UK time.
Let's take a look at the daily traded volume first. In theory, days with a higher trading volume should have a higher price fluctuation as there is increased demand to buy or sell Bitcoin.
To get an overview over the weekday volume, I calculated the weekly average volume and compared this average to the daily traded volume to see which days have more or less trading activity.
If we average this out, we'll get the following diagram:
As you can see, thursdays have the highest trading activity in the given time period. The trading volume drops significantly on the weekends with it's negative peak on sundays.
But now, how does this translate into price movement?
To analyze the price movement, I've calculated the weekly average price and then compared the daily average prices to the weekly average. The deviation of the daily average to the weekly average is then shown as a percentage (show as "Delta Avg" in the table"). A percentage greater than zero means a higher than average price and vice versa. I've also calculated the spread from the High - Low prices in respect to the daily average.
As you can see, the price is genereally lower during the mid of the week and higher on weekends.I would've expected to see more drastic price movement directions, but the data does not show this clearly. The data timeframe covers a really bumpy market with strong down- and uptrends. I assumed that weekly swing trends are dependent of the overall direction the market is heading. Let's see how the data looks like if we add trend information to it.
Adding Trend Dependency
To get the general trend direction, I took the weekly average of each week as an indicator. If the weekly average of the current week is below last week, the trend is downwards. If the weekly average of the current week is higher than last week, we're in an upward trend. This can of course be advanced by using a moving average to determine the trend direction.
This results in the following table containing all downward trend weeks:
And in this table for upward trend weeks:
The averages on the bottom now show a clearer picture and a weekly cycle can be seen more dominantly. The following chart visualizes the average daily deviation from the weekly average during downtrends.
And this is the average daily deviation from the weekly average during uptrends:
Based on the trend-dependent price movement, you can see that prices are usually the lowest on fridays and the highest on mondays during downward trends. In an uptrend, the prices are the highest on saturdays and the lowest on mondays. This makes perfect sense given the general trend direction.
So in order to buy Bitcoin, the data would recommend going in at the beginning of the week if you feel confident in an uptrend of the market. If you want to sell your Bitcoin during a downtrend, chances are high that you'll be getting the highest price on a monday.
Subscribe to insidr
Get the latest posts delivered right to your inbox.