Stock Market Follow-Through on Small Days

10Jun09

I’ve written a lot on this blog about the concept of follow-through (i.e. the likelihood of up days following up days, and vice-versa), how it has evolved over time, and how it is becoming stronger.

A quick primer for the uninitiated – the graph below shows the results of two strategies: one going long the S&P 500 index at today’s close if the index closed up today (green), and one if it closed down (red), from 1950. Geek note: these results are frictionless.

20090608.02

For most of the market’s history, it exhibited positive follow-through – up days tended to follow up days, and vice-versa. But around the turn of the millennium (grey dotted line), the market flipped contrarian – up days now tend to be followed by down days, and vice-versa.

I would never recommend anyone trade follow-through in and of itself (it’s way too volatile and subject to painful stints of ineffectiveness), but the concept is interesting because it’s a nice barometer for other more sophisticated short-term indicators like RSI(2).

Stock Market Follow-Through on Small Days

One weakness of follow-through is that it doesn’t discriminate between big and small up/down days, leading a number of readers to question whether all of the findings hold true for days that close just a bit up, or just a bit down.

So in this post, I want to look at follow-through on small days versus all others.

20090609.01

The table above shows the results of, from 2000, trading long at today’s close when the market closed down, but this time, short when it closed up. I’ve divided the results by days when the market only closed up/down within +/- 0.10 standard deviations (these are our small days and equal about +/- 0.3% in today’s market) versus all other days.

The statistics represent (a) the average daily return, (b) percentage of days that were winning “trades”, (c) the average return divided by the standard deviation of returns (a simple return vs risk metric), and (d) the number of times that condition has occurred since 2000.

The results clearly show that follow-through has been equally as strong following small days as any other.

In the next table, I’ve relaxed the criteria to qualify as a small day, and added +/- 0.15 SD and 0.20 SD days as well.

20090609.02

Again, same conclusion:

Daily contrarian follow-through has been equally strong (or perhaps even a bit stronger) following small up/down days as any other day.

One last note. Our own YK Strategy, while not a follow-through strategy, does use the market’s closing direction (up or down) as the determining factor to decide which of two possible positions to take each day. I’m often asked whether it matters if the market closes big or small. I hope this post shows why my standing opinion is that it doesn’t. A small close doesn’t, in and of itself, change expected next-day impact.

Happy Trading,
ms

 

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.



4 Responses to “Stock Market Follow-Through on Small Days”

  1. 1 Damian

    Interesting that the return vs. risk is increased though, no?

    • 2 marketsci

      RE to D: actually strange I thought…the results surprised me, but I saw them previously in testing for YK. The market appears to be binary to some degree. michael

  2. 3 William T

    Hi Michael,

    Have you already done a study of “BIG” days… for example, days with > 1 or 2 SD moves? I guess I’m curious because a prudent, young man like Scotty seems to often take a stronger stance (higher percent investment) on larger moves.
    Thanks, Bill

    • 4 marketsci

      RE to William T: very little on this blog because I don’t think I could do a better job than Hanna over at Quantifiable Edges (quantifiableedges.blogspot.com/).

      Having said that, I did run the same study above on big days and average follow-through return tended to be much higher on average, but much less consistent and much more volatile. I think big days are so much more complicated in terms of what drives them, hence the more disparate results. Also hence the reason that Scotty is such a prudent man =)

      michael


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s