Studies - Posted on Friday, May 15, 2009, 9:52 AM GMT +1


May Friday 15

S&P 500' Historical Intraday Performance on Option Expiration


For those who love the numbers, the table below shows the S&P 500′ intraday performance stats on option expiration Fridays since 01/02/1990 (open, high, low and close in comparison to the session’s close immediately preceding option expiration, almost always Thursday).


Is is notable that option expiration’s session itself is not an especially favorable time frame for the market, at least not in comparison to an average session and the respective at-any-time probabilities and odds.

On option expiration the SPX historically did not post a high above the previous session’s close as often as it -on average- did on an randomly chosen session (82.40% versus an at-any-time probability of 86.80% for a high above the previous session’s close), and the average winning trade on the close shows a profit of 0.63% compared to an at-any-time profit of 0.77% on a higher close. The profit factor as the sum of all profits divided by the sum of all losses with 0.87 on option expiration (that means one only would have won $0.87 for every dollar lost on option expiration) compared to an at-any-time profit factor of 1.07 clearly shows that -although option expiration week (!, not the session itself) represents a favorable time frame for the market- it regularly (in the long run) does not pay going long the SPX on close of the session immediately preceding option expiration’s Friday.

That DOES NOT mean that we won’t see a positive close on Friday’s session, but if you’d have always bought the SPX on close of the session immediately preceding option expiration and closed the trade on close of option expiration, you’d have lost money and fared worse compared to a ‘buy and hold’ approach or investing randomly with a one day investment period.

The following figure shows the SPX‘ (bell curve) distribution of profits and losses on option expiration (end-of-day gains/losses), and listed below are the intraday performance stats concerning option expiration’s session since 2003. The bell cure is slightly right-skewed, which means a bigger part of the distibution is concentrated on the left side (losses) of the distribution, which is also reflected by a profit factor below 1.



Successful trading,


P.s.: WordPress recently implemented a Twitter widget, so I’ll regularly make some intraday updates as well using Twitter. If you’re interested in, please have a look at the blog during the trading session as well or subscribe directly to Twitter (recommended).

Disclaimer: Long BGZ (Daily Large Cap Bear 3x Shares) at time of writing.

Comments (6)


  1. moneyfriend says:

    Thank you Frank, I think this would be more relevant if you split the study up in to times when the opex is below the 200dma and times when it is above, or maybe in relation to rsi14, for example.

  2. moneyfriend says:

    I have found in my own backtesting that separating the results in to 2 groups group 1: > 200 dma, group 2 < 200dma, often reveals very different results between the 2 groups. I like seeing the results for both groups separate and also together. I agree it is good to see the pure results as well.

  3. moneyfriend says:

    I think it is relevant as well because opex relates to the max pain effect, and max pain price often correlates with a 40 day moving average for instance. Thanks for the great work you share here.

  4. Peter says:


    Interesting that there is an 86.8% chance of a high above the previous days close. Actually, the average since October 2007 is 12.2 S&P500 points higher. How about buying at close and setting a stop-limit order for the next day to sell at closing price + say 8 points?

    The problem of course is limiting the downside when the limit order fails to trigger. Like 26th September 2008, the limit never triggers and the close to close is down 106 points. Still, food for thought!


  5. Clark Sims says:

    That is a great piece of work. Thanks for posting it.

Leave a Reply

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