Santa Claus Rally Backtester for ThinkOrSwim (Free Download)

Overview

In this tutorial I build a Santa Claus rally backtester for ThinkOrSwim to put the idea to a test: has the market tended to rise in the week before Christmas? The strategy buys at the open of the first bar on or after December 18 and sells at the close of the first bar on or after December 24, every year on the chart.

Then I use the strategy report to look at the Decembers inside two bear markets, 2000 to 2002 and 2008. The backtester is 4 lines of thinkScript with no inputs, so it suits beginners, and you can change its two dates to test another window inside a single month.

Key Takeaways

What Is the Santa Claus Rally?

The Santa Claus rally is a seasonal tendency for stocks to rise around Christmas, and definitions of the window differ. The best-known one, from the Stock Trader's Almanac, covers the last 5 trading days of December and the first 2 trading days of January. In the video I work from a snippet from Investopedia and test the week before Christmas instead.

Two points stand out in that snippet. First, the rally takes place in the week prior to December 25, and a week before Christmas is December 18, which becomes the start date. Second, it describes the strategy as having had a small positive average return.

A positive average return would imply a positive expectancy, and that's the part worth testing. The Santa Claus rally backtester asks two questions. Does the rally happen more often than not? And what would the P/L have been if you had bought a week before Christmas and sold right before it, year in and year out?

The Santa Claus Rally Rule

The Santa Claus rally backtester has one entry date and one exit date, and both are written the same way: the first bar on or after a day in December. The table lists every part of the rule in the download.

The exit uses December 24 because the market is closed on the 25th. When the 24th has no bar, the exit moves to the next trading day. That happened in 2021, when the bar after December 23 was December 27.

How to Build the Backtester in ThinkOrSwim

The Santa Claus rally backtester has no inputs, so the whole build is a new strategy with 4 lines in it:

1. Click the Studies icon on your chart.
2. Click the Strategies tab, then click Create.
3. Give the strategy a name. I called mine TI_SantaClausRallyBacktester.
4. Delete the starter code and paste in the code from the Code tab on this page.
5. Click OK and apply it to a daily chart of SPY. Set the chart's time span to the history you want to test.

The download has the same 4 lines as a .ts strategy file. If you'd rather import it, our guide on how to import a thinkScript file into ThinkOrSwim shows each step.

How the Code Works

The Santa Claus rally backtester code has 2 date conditions and 2 orders. The logic follows the same framework as our Starbucks holiday drinks backtest: a start date, an end date, and care over which price each order uses.

Step 1: The Entry Condition

This step’s code is free with an account. Create one or sign in.

Three built-in thinkScript functions do the work here:

• getYYYYMMDD() returns the date of the current bar.
• getDayOfMonth() takes a date and returns its day of the month as a whole number, such as 17 or 18.
• getMonth() returns the month of the current bar, and December is 12.

Feeding the bar's date into getDayOfMonth means you don't have to type a date for every December of the past 20 years. A day greater than 17 means the 18th or later, so entry is true on every December bar from the 18th to the end of the month.

Step 2: The Exit Condition

This step’s code is free with an account. Create one or sign in.

The exit condition is the same line with 23 in place of 17. It is true on every December bar from the 24th onward.

Step 3: The Orders

This step’s code is free with an account. Create one or sign in.

AddOrder places the simulated trades. The buy order needs entry to be true on the current bar and false on the previous one, written !entry[1]. That picks out the first bar of the window, so the strategy buys once per year at that bar's open, for 1 share.

The sell order uses the same test on exit and sells to close at the close of the first bar dated the 24th or later. The strategy has no short orders, because the Santa Claus rally is a bullish idea.

Checking the Entry and Exit Prices

The Santa Claus rally backtester fills at the open of the entry bar, while ThinkOrSwim draws the pink strategy marker on the next bar. In the video I checked the December 2021 trade with Show report, which you open by right-clicking on the chart.

• Entry. The report listed the entry on December 21, 2021 at $454.48. That price is the open of the December 20 bar, the first bar after December 17, so the strategy used the right bar. The triangle on the chart marks the actual entry price.
• Exit. The report showed $477.26, the close of the December 27 bar. The bar before it was December 23, so the 27th was the first bar dated the 24th or later.

Run this check on any backtester you build. A strategy that uses the wrong bar's price can show a P/L that no one could have traded, which is the bug I fix in the SPY meltdown backtester video.

Results From the Video

The Santa Claus rally backtester produced these results on daily SPY charts in the video. I most likely recorded it in December 2022: the newest trade I check is December 2021, and I talk about the December ahead. They are historical results for 1 share per trade. The video reads the direction of the P/L for each time span but no dollar totals and no win rate.

The 20-year span is the same two decades I believe Investopedia was referring to, and there the small positive result lines up with its description. Over 5 years, the negative P/L told me not to count on a positive expectancy heading into that December, based on those 5 data points.

On the maximum data, the P/L stayed positive. Scrolling back, the strategy worked for most of the early years until it stopped working, which accounts for much of the P/L, and lately it has been climbing back up. The same test runs on QQQ or any other symbol you load.

The Bear Market Decembers

The bear market Decembers on SPY split 2 and 2 in the video. I opened the report on the maximum data and looked at the two bear markets that have data available. I only looked at whether each December's P/L was positive or negative.

In the 2000 to 2002 bear market, the rally held in 2 of 3 Decembers. Counting 2008, the 4 bear market Decembers split 2 positive and 2 negative, and I said in the video that this didn't make me optimistic about a Santa Claus rally that year. Four Decembers is a small sample, so read it as context.

How I Read the Results

The Santa Claus rally backtester gave me one input for my bias that December, alongside the chart. SPY was near the top end of a longer-term downtrending channel that I had been following for some time. There was a recent uptrend, but the larger trend was still down, and up toward a previous high I was watching for sellers to come in and push price lower.

With the bearish read on the chart, the negative 5-year result and the split bear market Decembers, I was looking at that December through bearish goggles. The backtest describes past Decembers. It's a read on the season and not an entry signal by itself. Decide where your stop goes before you enter, and size the position so a stop-out costs a small, fixed share of your account.

Changing the Dates or the Market

The Santa Claus rally backtester's dates sit in the two conditions, so you change them in the code. The table shows what to edit.

Keep the entry day smaller than the exit day and both conditions in the same month, or the strategy can't close each trade inside the window.

For a wider view of the calendar, our Seasonal Analysis indicator shows how often each month has closed higher and its average change. Our other free event tools use different triggers:

• The SPY meltdown backtester triggers on price instead of the calendar. It buys the open after a day that closes more than 3% below its open.
• The election backtester swaps the day-of-month test for a list of exact dates, 6 election days, and exits a set number of bars later.
• The Super Bowl indicator is a study, not a strategy. It labels each year with the Super Bowl winner's conference and that year's change.

Our research on S&P 500 November seasonality patterns in bear markets covers the month before December, and the full list of strategies is on our ThinkOrSwim backtesters page.

Download

Questions

Related tools

Super Bowl IndicatorIndicator · FreePost-Election BacktesterBacktester · FreeSeasonal AnalysisIndicator · FreeSPY Meltdown BacktesterBacktester · FreeTTM Squeeze Histogram BacktesterBacktester · FreeMoving Average Pullback BacktesterBacktester · Free