Are the best trading days of the week(TDW) really good? Larry Williams used TDW as an additional element improving the result of simple trading figures and never used it as a separate element. In particular, he used TDW in a volatility breakout system in order to reduce the number of transactions and increase winrate.
The main problem with TDW is their variability. On one time interval, purchases on Fridays will work well; on another time interval, they will lose money. In most cases, trading systems consist of several elements complementing each other and improving the overall result. Each element of the trading system should be self-sufficient and provide an advantage regardless of successful gluing with other elements. All elements should complement each other improving productivity, and not stretch one element at the expense of another.
At the beginning of Chapter 7, the book “Long-Term Secrets of Short-Term Trading,” Larry provides a simple trading system based on buying the SP500 index every day at the opening price and closing the transaction the next day at the closing price using a stop of $ 3250. The system was tested for the period 1984-1998. For each day of the week, statistics were provided. Where the number of trades column was of particular interest, in which the number equal to approximately 400 trades for each day of the week was indicated. But if you count how many were for example Mondays or other days of the week for 14 years, then you get a figure of the order of 730-770. Because no additional input was provided that could reduce the number of transactions by 2 times, I think the strategy tester used by Larry could not open several transactions at the same time. Most likely he opened a position on Monday and held it until the end of Tuesday afternoon, preventing him from opening another position on Tuesday and so on for all other days. I consider it more correct to open positions every day in order to incorporate into the statistics all cases of the appearance of each day of the week. I do not think it is advisable to compare the test results in the book with the ones I received. They are very different. Only the information regarding Mondays on the SP500 matches, as the strongest day of the week.
SP500
As can be seen from the summary statistics for the entire available time period 1984-2020, all days of the week have a winrate of more than 50% and make a profit when buying daily at the opening of the day and exit at the close of the same day. The most powerful days in terms of points earned are Tuesdays and Mondays. The weakest day is Friday.
I find it strange that Larry begins to consider TDW in terms of holding a position until the end of the next day, when closing the entry day is a more relevant option, for a more visual result of TDW. As was done, the results are divided into time periods presented below. The number of accumulated points and win rate are indicated.
1984-1998
Monday: 343 (57%)
Tuesday: 486 (51%)
Wednesday: 282 (54%)
Thursday: -108 (51%)
Friday: 121 (52%)
1998-2011
Monday: 356 (53%)
Tuesday: 227 (52%)
Wednesday: -9 (55%)
Thursday: 315 (55%)
Friday: -115 (51%)
2012-2020
Monday: 220 (54%)
Tuesday: 693 (56%)
Wednesday: 785 (56%)
Thursday: 441 (56%)
Friday: 183 (54%)
When broken down into time intervals, Mondays and Tuesdays invariably plus. Larry gives one of the variations of the strategy based on buying on Monday, if Friday closed down or buy on Monday when opening Monday with a gap down. All these variations have the right to life, because Monday is a strong day on the SP500.
Brent Oil
1991-2001
Monday: -26 (43%)
Tuesday: -9 (46%)
Wednesday: 0 (53%)
Thursday: 2 (50%)
Friday: 2 (50%)
2002-2011
Monday: -73 (46%)
Tuesday: 3 (49%)
Wednesday: 38 (53%)
Thursday: 15 (52%)
Friday: 16 (53%)
2012-2020
Monday: -61 (43%)
Tuesday: -18 (48%)
Wednesday: -3 (52%)
Thursday: 10 (52%)
Friday: 32 (54%)
For Brent, Mondays are stable for sales, Thursday and Friday are the best days to buy.
Conclusion
To use TDW as Larry does, I consider it incorrect to select the best days for each trading system. If you apply this approach, then taking into account the specifics of the tool, if on the SP 500 at all time intervals throughout history, Mondays are the strongest days, then you need to add Mondays to the trading systems for purchasing the index, but not Wednesday, which Larry uses in the system GSV. And sell index in every day except Mondays also not good idea.
The main problem with TDW is their variability. On one time interval, purchases on Fridays will work well; on another time interval, they will lose money. In most cases, trading systems consist of several elements complementing each other and improving the overall result. Each element of the trading system should be self-sufficient and provide an advantage regardless of successful gluing with other elements. All elements should complement each other improving productivity, and not stretch one element at the expense of another.
At the beginning of Chapter 7, the book “Long-Term Secrets of Short-Term Trading,” Larry provides a simple trading system based on buying the SP500 index every day at the opening price and closing the transaction the next day at the closing price using a stop of $ 3250. The system was tested for the period 1984-1998. For each day of the week, statistics were provided. Where the number of trades column was of particular interest, in which the number equal to approximately 400 trades for each day of the week was indicated. But if you count how many were for example Mondays or other days of the week for 14 years, then you get a figure of the order of 730-770. Because no additional input was provided that could reduce the number of transactions by 2 times, I think the strategy tester used by Larry could not open several transactions at the same time. Most likely he opened a position on Monday and held it until the end of Tuesday afternoon, preventing him from opening another position on Tuesday and so on for all other days. I consider it more correct to open positions every day in order to incorporate into the statistics all cases of the appearance of each day of the week. I do not think it is advisable to compare the test results in the book with the ones I received. They are very different. Only the information regarding Mondays on the SP500 matches, as the strongest day of the week.
SP500
As can be seen from the summary statistics for the entire available time period 1984-2020, all days of the week have a winrate of more than 50% and make a profit when buying daily at the opening of the day and exit at the close of the same day. The most powerful days in terms of points earned are Tuesdays and Mondays. The weakest day is Friday.
I find it strange that Larry begins to consider TDW in terms of holding a position until the end of the next day, when closing the entry day is a more relevant option, for a more visual result of TDW. As was done, the results are divided into time periods presented below. The number of accumulated points and win rate are indicated.
1984-1998
Monday: 343 (57%)
Tuesday: 486 (51%)
Wednesday: 282 (54%)
Thursday: -108 (51%)
Friday: 121 (52%)
1998-2011
Monday: 356 (53%)
Tuesday: 227 (52%)
Wednesday: -9 (55%)
Thursday: 315 (55%)
Friday: -115 (51%)
2012-2020
Monday: 220 (54%)
Tuesday: 693 (56%)
Wednesday: 785 (56%)
Thursday: 441 (56%)
Friday: 183 (54%)
When broken down into time intervals, Mondays and Tuesdays invariably plus. Larry gives one of the variations of the strategy based on buying on Monday, if Friday closed down or buy on Monday when opening Monday with a gap down. All these variations have the right to life, because Monday is a strong day on the SP500.
Brent Oil
1991-2001
Monday: -26 (43%)
Tuesday: -9 (46%)
Wednesday: 0 (53%)
Thursday: 2 (50%)
Friday: 2 (50%)
2002-2011
Monday: -73 (46%)
Tuesday: 3 (49%)
Wednesday: 38 (53%)
Thursday: 15 (52%)
Friday: 16 (53%)
2012-2020
Monday: -61 (43%)
Tuesday: -18 (48%)
Wednesday: -3 (52%)
Thursday: 10 (52%)
Friday: 32 (54%)
For Brent, Mondays are stable for sales, Thursday and Friday are the best days to buy.
Conclusion
To use TDW as Larry does, I consider it incorrect to select the best days for each trading system. If you apply this approach, then taking into account the specifics of the tool, if on the SP 500 at all time intervals throughout history, Mondays are the strongest days, then you need to add Mondays to the trading systems for purchasing the index, but not Wednesday, which Larry uses in the system GSV. And sell index in every day except Mondays also not good idea.