If the average annual return for the S&P 500 is 8%, wouldn't it be far better if you only bought on days that it fell?

  • Better than what, and better how? – yoozer8 Jan 5 at 21:58
  • I guess to clarify what Im trying to say is lets say you purchase s&p stocks spread over the course of a year, are you better off only buying on the down days? – Daniel Jacobson Jan 5 at 22:00
  • 4
    The problem is you don't know whether a day is up or down until trading for the day has stopped. – Eric Jan 5 at 22:03
  • 2
    Down relative to open same day? Week? Month? Year? What does your money do while it's sitting around waiting for down days? – Hart CO Jan 5 at 22:12
  • 1
    @Fattie - You've missed the point. If the OP sets up some sort of buying criteria for buying (say the SPY is down 5 or more points), it's quite easy to recognize that moment when it occurs during trading hours and even easier to buy at that price. Just place a limit order to buy the price when the SPY is down 5 points. It doesn't matter how violently SPY fluctuates or what time it becomes down 5 points. When it gets there, you're filled. – Bob Baerker Jan 6 at 13:36

I ran the numbers in Excel. Here are the assumptions I made:

  • You started investing in SPY on the day that it launched (January 29, 1993).
  • Your broker allowed dollar-based investing.
  • On each trading day, you deposited $10, and invested it according to one of two strategies:
    • Strategy A: You bought $10 of SPY at the closing price, regardless of what that price was.
    • Strategy B: You instructed your broker to buy $10 (plus your saved cash) of SPY at the closing price only if the closing price is lower than the previous closing price. If the closing price was not lower than the previous closing price, you would instead save the cash to invest it later.
  • You reinvested all dividends at close on the in-dividend date. (How nice of your broker to lend you the cash to do this!) In other words, the performance of SPY is assumed to be the performance of the "Adjusted Close" column in Yahoo Finance.

At close on January 5, 2021, if you had used Strategy A, you would have $329,337.26, whereas if you had used Strategy B, you would have $329,273.05 (which includes $10 in cash, since the closing price on January 5 was higher than on January 4). So, if you had used Strategy B, you would have $64.22 less, or 0.019% less, than if you had used Strategy A.

In conclusion, it really won't make any difference at all.

  • Thanks, it sounded like a really good idea when I thought about, but I can see it doesnt play out as nicely as Id hope. There may be some better strategies here because you are comparing the investment of more money because both strategies invest 10 dollars. Strategy b would need to take the average number of down days and divide that by $3650 so that both strategies are investing roughly the same amount of money. Another strategy would be something like investing $5 on up days and investing $10 on down days – Daniel Jacobson Jan 6 at 18:16
  • There is an old investing adage: “Time in the market beats timing the market.” . The vast majority of the gains in either scenario are from the moneys invested in the early years. The specific timing of those investments becomes largely irrelevant as time goes on. – D Stanley Jan 6 at 18:45
  • I'm kind of impressed how closely those two numbers sit! I take it that if you looked at any point over the 28 years they're always about this close together? – Andrew Jan 6 at 18:57
  • @Andrew More or less. The difference starts at $0, obviously, and then goes up and down over time, eventually reaching a maximum of about $64. I charted the difference and found it looked nearly identical to a price chart of SPY, with a peak in 2000 at about $40, then a trough in 2003, then another peak in 2007 at about $40, then another trough in 2009, before finally increasing to $64. – Tanner Swett Jan 6 at 19:09
  • It's impressive that you went to all of that effort to provide the OP with an example with actual numbers. However, it does not prove that buying at some other amount of 'down day" does not outperform buy and hold. And for that matter, that it underperforms. Yours is just one strategy that "really won't make any difference at all." – Bob Baerker Jan 6 at 19:34

You'd have to better define some parameters to be able to assess returns via back-testing, but the primary issue with this strategy is that you could sit with idle money for long bull runs. You could also buy on a down day that turns out to be the start of a long bear run.

A popular notion is that time in the market beats timing the market. That's generally true because people can't reliably time the market, if they could they would have vastly superior returns. There are tools available for back-testing strategies if you're so inclined, Think or Swim includes back-testing capabilities (not sure if available with their free paper trading accounts).


The short answer is that if the long term trend is up after your initial buy and hold purchase then buy and hold will outperform your intermittent purchase strategy. If the market is down long term then the intermittent strategy will outperform. Note that this is an equal dollar comparison, eg. invest all initially versus investing the same amount of money at various intervals.

If it's an oscillating market then it's up for grabs which one will do better and the answer will depend on what your intermittent strategy is. Will you buy just before the close if the S&P 500 is down 3 points intraday? 5 points? 10 points? Once you determine that, set up a spreadsheet and download historical data from Yahoo Finance or similar.

An unrelated way to explain this is that if you play with a DRIP calculator, for a stock in a long term descent, not reinvesting dividends loses less money, and vice versa for an up trend. The price of subsequent investments determines the success of the strategy.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.