Daily Commentary - Posted on Saturday, May 16, 2009, 10:05 PM GMT +1

4 Comments


May Saturday 16

Trading the Odds on Monday – May 18, 2009

WE031672-klein

On Friday the S&P 500 perfectly complied -from a historical and statistical point of view- to the expected negative outlook for option expiration’s session with respect to those (negative) setups which were triggered on Wednesday’s (S&P down more than 4.50% on day during expiration week, see my post Trading the Odds on Thursday – May 14, 2009) and Thursday’s session (strong breadth on lower NYSE Total Volume, and SPX with a lower low and lower high on a higher close and on lower NYSE Total Volume, see my post Trading the Odds on Friday – May 15, 2009).

On Friday the S&P 500 closed lower1.14% on the day, and breadth was mixed with NYSE Advancing Issues / Declining Issues at 0.57 (only) and NYSE Advancing Volume / Declining Volume notably weak at 0.25. The Nasdaq 100 outperformed the S&P 500 again, with a lower close of -0.34% only, and NASDAQ Advancing Issues / Declining Issues at 0.66, but NASDAQ Advancing Volume / Declining Volume relatively strong at 0.73 (and the respective Arms Index/TRIN closing below 1 in regularly bullish territory).

But there were some other notable particularities and observations as well:

  1. Although option expiration week is regularly a favorable period for the markets, the S&P 500 closes lower -5.0% on the week, and lower on Friday (option expiration) as well.
  2. The breadth index, in this case the 5-day (or one week) SMA (Simple Moving Average) of [NYSE Advancing Issues / (Advancing Issues + Declining Issues)] closed at 37.95%, and the 5-day SMA of [NYSE Advancing Volume / (Advancing Volume + Declining Volume)] closed at 30.41%. That means, during the last 5 sessions on average NYSE Advancing Issues only accounted for 37.95% of the sum of Advancing and Declining Issues on every session, and on average only approximately 30% of NYSE Total Volume went into advancing stocks. This not only marks a 2-month low for both breadth figures, but fall -from a historical perspective since 01/02/1991- at least into the bottom 10th percentile as well.

First of all I checked for those occurrences since 01/02/1990 where one (or combinations) of the following setups were triggered:

  1. Option expiration in general (Setup S1),
  2. Option expiration, with the S&P 500 posting a lower close (Setup S2)
  3. Option expiration, with the S&P 500 posting a lower weekly close (S&P 500 closing lower than 5 sessions before) (Setup S3)
  4. Survey 1 AND Survey 2 AND Survey 3 combined (Option expiration, with the S&P 500 posting a lower daily and weekly close) (Setup S4)
  5. In comparison: Fridays, which are NOT option expiration, with the S&P 500 posting a lower daily and weekly close (Setup S5)

Table I below shows -side by side- the SPX‘ performance (since 01/02/1990) on the next session (in this event Monday, May 18) immediately following those sessions where setups S1 to S5 had been triggered.

It is especially remarkable that concerning both 1) the probability for a higher close (‘Winning Trades’) as well as 2) the expected pay-off (‘Profit factor’) on the next session (in this event Monday, May 18) immediately following option expiration (Setup S1) are significantly below the respective at-any-time probability and odds, and they are getting worse if one additionally takes into account that the S&P 500 closed lower on option expiration and lower on the week as well. Although the sample size is getting small with 64 occurrences only, the respective profit factor concerning setup S3 of 0.44 is way below the respective at-any-time profit factor of 1.07, and as well way below the respective profit factor of 1.03 on it’s counterparts, means those Friday’s which were NOT option expiration, but where otherwise all other conditions had been met and setups being triggered (Setup S5: Friday, with a lower daily and weekly close on the S&P 500).

20090516-SPX-11

Table II below now shows -side by side- the SPX‘ performance (since 01/02/1990) over the course of the next week (following 5 sessions) immediately following those sessions where setups S1 to S5 had been triggered.

20090516-SPX-21

Although we may note an improvement in the respective performance stats concerning setups S1 to S4 (‘Winning Trades’ and ‘Profit Factor’), they are still below the respective at-any-time stats, with stats concerning setup S4 still significantly below the respective stats regarding it’s counterpart S5.

So a lower SPX’ daily and weekly close on option expiration shows a remarkable tendency for some follow-through weakness over the next couple of session, not only in comparison to an at-any-time investment and/or buy-and-hold approach, but in comparison to all other Friday’s which are NOT option expiration, but where otherwise all other setup were triggered as well.

____________________________________________

The second part of today’s report addresses the remarkable weakness in NYSE’ weekly breadth figures.

Remember: The breadth index, in this case the 5-day (or one week) SMA (Simple Moving Average) of [NYSE Advancing Issues / (Advancing Issues + Declining Issues)] closed at 37.95%, and the 5-day SMA of [NYSE Advancing Volume / (Advancing Volume + Declining Volume)] closed at 30.41%.

First of all I checked for those occurrences since 01/02/1990 where one (or combinations) of the following setups were triggered:

  1. 5-day SMA of [NYSE Advancing Issues / (Advancing Issues + Declining Issues)] below 40% (Setup S1),
  2. 5-day SMA of [NYSE Advancing Volume / (Advancing Volume + Declining Volume)] below 35% (Setup S2)
  3. Survey 1 AND Survey 2 combined -means triggered on the same session- (Setup S3)
  4. Unchanged for comparison purposes: Option expiration, with the S&P 500 posting a lower daily and weekly close (Setup S4)
  5. Unchanged for comparison purposes: Fridays, which are NOT option expiration, with the S&P 500 posting a lower daily and weekly close (Setup S5)

Table III below shows -side by side- the SPX‘ performance (since 01/02/1990) on the next session (in this event Monday, May 18) immediately following those sessions where setups S1 to S5 had been triggered.

20090516-SPX-31

Table IV below now shows -side by side- the SPX‘ performance (since 01/02/1990) over the course of the next week (following 5 sessions) immediately following those sessions where setups S1 to S5 had been triggered.

20090516-SPX-41

The setups concerning NYSE breadth instead show a completely different picture: The respective performance stats concerning setups S1 to S3 -especially S3 where both setups were triggered on the same session- (‘Winning Trades’ and ‘Profit Factor’) are significantly above the respective at-any-time stats, and the profit factor concerning setup S3 (likewise triggered on Friday’s session) triples the profit factor concerning setup S4 (Option expiration, with the S&P 500 posting a lower daily and weekly).

And last but not least -you may assume what will be the final act of the night-, the combination of both, setups S3 and S4, represented in form of the new setup S5 in the stats below:

S5 is triggered when/where

  1. 5-day SMA of [NYSE Advancing Issues / (Advancing Issues + Declining Issues)] below 40% (Setup S1),
  2. 5-day SMA of [NYSE Advancing Volume / (Advancing Volume + Declining Volume)] below 35% (Setup S2), and
  3. Option expiration, with the S&P 500 posting a lower daily and weekly close

Table V below shows -side by side- the SPX‘ performance (since 01/02/1990) on the next session (in this event Monday, May 18) immediately following those sessions where setups S1 to S5 had been triggered. Setups S1 to S4 remained unchanged to table IV above:

20090516-SPX-51

Table VI below now shows -side by side- the SPX‘ performance (since 01/02/1990) over the course of the next week (following 5 sessions) immediately following those sessions where setups S1 to S5 had been triggered. Setups S1 to S4 remained unchanged to table IV above:

20090516-SPX-61

To my own surprise the combination of the two setups, one with a significant negative outlook concerning the next session and the following trading week (options expiration), and the other one with a significant positive outlook concerning the next session and the following trading week (notable weekly breadth index), came up with an even stronger outlook (Winning Trades at 62.50%, and Profit Factor of 2.90, both figures representing the highest figures among all single and combined setups) concerning at least the following trading week (the next session shows the same positive Winning Percentage, but due to a single major loss on the close an only average profit factor). It seems that the weak breadth index was able to trigger (or ‘pick up’) with a high accuracy the most profitable occurrences among the otherwise notably weak option expiration week setups.

But the pity -expectedly- is that the sample size with 7 occurrences only is way too low to read any statistically relevant into it, but nevertheless not only something to be taken into account, but it additionally negates the negative outlook with respect to the weak option expiration week and session.

________________________________

Bottom line:

  1. With respect to all the stats, observations and conclusions mentioned above, I think there is a good chance that although the last week went to the bears, the coming session on Monday and coming week as well will probably go to the bulls again, additionally taking into account that speculative interest (outperformance of the Nasdaq 100 and it’s respective relatively strong breadth figures) is still running high, and the ‘buy the dip’ mentality seems still to be the prevailing theme at least for the time being (means at least over the next couple of sessions).

Successful trading,

Frank

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: No position in the securities menitoned in this post (SPX, NDX) at time of writing (but nonetheless positioned long in the market).

Comments (4)

 

  1. Rod says:

    Super post. Was that done in Matlab or Excel?

  2. Jimmy says:

    Kudos on your work. Pretty much coincide with what I been seeing on the short term basis (though not on a day to day outlook.) While your blog was down for implementing MathLab the past weeks, have you out of curiosity did some back test see if the missed days did in fact suggested what the market would do?

    I do most of my research on medium and long term basis, so I’m in agreement with Doug that this is just a bear market rally (and not a new secular bull market.)

    Jimmy

Leave a Reply

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