iBankCoin
Joined Nov 11, 2007
1,458 Blog Posts

Buy the Close; Sell the Open?

I’ve been somewhat fascinated with a paper I found on the MarketSci blog. The paper is Returns in Trading vs. Non-Trading Hours: The Difference in Night and Day.

The paper is fascinating as it presents research showing a profitable trading strategy can be developed simply from buying the close of the QQQQ and selling the next open, day after day. The authors’ basic assertion is that the markets tend to be more bullish in the after-hours, at least on the QQQQ. Read the paper to understand why this strategy may not work as well on DIA or SPY.

Beyond presenting a profitable trading strategy, I believe the implications of the paper are that traditional strategies which trade open-to-open or close-to-close may be improved by switching to initiating positions at the close and liquidating them on the open. (This would certainly substantiate the aphorism that smart money trades the close while dumb money trades the open).

While it would be easy to take some of our existing strategies and apply a close-to-open scenario, I had hoped that I would first be able to get close to replicating the results presented in the paper. As of right now, I’m not close.

So all this build-up is not much more than an initial introduction to the idea of buying the close and selling the open, and an excuse for why I don’t have a great post tonight. Once I get closer to replicating the author’s results I will begin testing this idea across a few strategies and ETFs. Of course, I’ll share the results here.

If you enjoy the content at iBankCoin, please follow us on Twitter

8 comments

  1. cebu sun

    Wood:
    Thanks for the insightful reference. As I read itthe kelly-clark article reflects:
    40% of the daily Market volume is due to DayTraders. They have what is termed (LOL) the “Illusion of Control”. They erroneously feel that by selling EOD they decrease risk when in actuality they have the effect of decreasing the market prices EOD by their sales.

    The authors pulled data from 96 to 06. I wonder if the increased volatility of this 07 to 09 market would reflect different results? Selling EOD in 08 09 may have decreased a load of loss.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  2. Cuervos Laugh

    I did a quick run on UUP, GLD, USO (I had data for a post I’m working on) and the buy close->sell open results were fairly miserable. Expectancy ranges from 0 to $.04/share

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  3. Woodshedder

    Cebu, yes, I’m trying to see what has been happening from 07-Present. Part of my problem is that the authors used a volume-weighted average price of a five minute bar to determine their entry and exit prices. That was something I did not have enough time to code in last night.

    My suspicion is that from 07-present this buy close sell open has not been working, but right now I am not close enough to replicating their study to say that with any degree of certainty.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  4. Woodshedder

    Cuervo, per the research, we would not expect this buy close/sell open to work on UUP, GLD, and USO. I don’t have time to explain right now, so I’ll just say to read the research.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  5. Bob Barnes

    The results for this system over the last 16 months are dismal. And running it backwards doesn’t improve the equity curve much (Sell the close, cover at open). Running this on Monday’s is the only day that makes any money. The results improve significantly if you add a liitle filter, either RSI(2)or MA threshold cross based. Michael’s system may have worked back 1996-2006, but a few things have changed since then that warrant further consideration.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  6. cebu sun

    Wood and Bob:
    Yes intuitively we would expect the kelly-clark Long Strategy (buy at close, sell at open) would work in bull markets but not in this nasty bear.

    I would like to see some 07 to 09 data on the 3pm effect on the market(Bond traders quit the bond pit and start on equities). They seem to be a whale in a 50 gallon aquarium. They magically are gifted? an extreme market (hitting upper resistance or lower support lines)at approx 3pm and then go for the simple reversion trade to EOD.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  7. cebu sun

    Bob
    I am surprised the Reverse i.e., shorting the close and covering the open didnt perform well in the 07 to 09 bear months.

    • 0
    • 0
    • 0 Deem this to be "Fake News"
  8. Benn Fine

    Hey all:

    really like these discussions; i get addicted to ameritrade’s strategy desk from time to time and can spend endless hours studying these things and coming up with nothing.

    Now, if anyone has time, can they comment on the very simple formulas given here:
    http://www.hotstockmarket.com/forums/showthread.php?t=67526

    I am new to that website but a thoughtful person has nicely given the entry and exit signals for how he trades FAZ and FAS. The results are wonderful, but as far as I can tell, the code cheats by looking ahead-initially i thought it didn’t but i am almost certainly convinced now.

    He presents an interesting idea though, since I had never thought of using multiple time frames for my exit, though due to the way strategydesk implements things i don’t think its correct.

    For those who have issues reading strategydesk code, the bu rule is simply,

    buy on candle close if this 15 min candle closes 1% higher than the last 15min candle (ignore his strange time of day effect code):
    Bar[Close,15] greaterthan (Bar[Close,15,1]*1.01)

    the sell signal is a bit strange-here it is in strategydesk code form:
    Bar[Close,1] lessthan (Bar[Open,D]*.98) OR Bar[Close,1] greaterthan (Bar[Open,D]*1.08)

    we are now looking at 1minute bars compared to daily bars (????)

    but what is strange is he checks the “override interval” command in strategydesk, so that everything is run at the same interval, in this case 15 minutes. this rule then reads if the close of a 15 minute bar is SOMEthing compared to the open of the same 15 minute bar, then sell at the open of that bar. that’s peeking ahead!

    anyway, its interesting, but unfortnately doesn’t seem to work; any comments from this crazy but brilliant group would be most appreciated.

    Thanks!

    • 0
    • 0
    • 0 Deem this to be "Fake News"