Turnaround Tuesday – Testing a simple Weekday Pattern

There are many patterns out there – some of them are real, others are just a myth. By back testing them, we can find out which one can give us an edge. In this article, I would like to show how to code and test the famous Turnaround Tuesday pattern for the stock market.

What is the Turnaround Tuesday?

You might have heard of the “Turnaround Tuesday” in the equity indexes. The idea behind it is simple: If the equity index has a negative performance on Monday, then Tuesday`s market action often shows a bullish turnaround. Now let us code this. It is really easy using the dayofweek(date) function:

Equilla code for Turnaround Tuesday

  • In line 2 we check whether today is Tuesday
  • in line 3 we check whether the previous day`s close was smaller than the previous day`s open
  • If both conditions are met, the strategy goes long at the open
  • In line 6 I added a condition which exits the position on the following day`s close (Wednesday close)

Backtesting the S&P 500 ETF – Does the Pattern work?

Here are the results for SPY (without commissions) when each signal was traded with 100k USD:

  • 614 trades
  • +0.168% on average

Turnaround Tuesday Backtest for SPY Can we improve the performance when we only take a signal when there was a significant down move on Monday? For example at least 1%? Let us find out by adding this rule into the existing code.

  • In line 5 we check whether the close was x per cent lower than the open on Monday
  • In order to have a flexible parameter, we use an input called MinDown (line 2)

Modified Turnaround Tuesday Equilla Code

Take a look at the results now:

  • the number of trades came down to 129
  • the average profit per trade increased from +0.168% to 0.445%

Modified Turnaround Tuesday Backtest for SPY


Tradesignal is a registered trademark of Tradesignal GmbH. Unauthorized use or misuse is strictly prohibited. Data by Thomson Reuters Eikon.

4 thoughts on “Turnaround Tuesday – Testing a simple Weekday Pattern

    1. Hi Michael, I made a backtest including your suggestion. If you add a filter which requires the previous week`s performance to be lower than -1%, then the number of trades falls from 134 to 54, but at the same time the average trade profit climbs from 0.71% to 1.25% – which is great I think.

Leave a Reply

Your email address will not be published. Required fields are marked *

Subscribe to this Blog