Daily Commentary - Posted on Friday, August 21, 2009, 6:43 AM GMT +1

5 Comments


Aug Friday 21

Trading the Odds on Friday – August 21, 2009

Despite the fact that on Thursday’s session the market rejected to go down although – from a statistical and historical perspective – it was supposed to do (see my posting Trading the Odds on Thursday – August 20, 2009), at least a red flag (for going or holding on to a short sell) was provided early in the session due to an exceptional strong BKX Banking Index which never showed any kind of weakness as a prerequisite for a sustainable move on the downside.

So I decided to cover my shorts entered into during the first hour of the session (see my respective Twitter Updates) and did not think about going short again (until the close) due to the fact that with a BKX Banking Index trading higher more than +2.0% while the ES (E-MINI S&P 500) was NOT up more than +1.0% going into the final hour of the session, chances were lopsided in favor of a positive tendency during the final hour of the session (ES E-MINI S&P 500 trading higher on 68 and lower on 23 occurrences only, see my Twitter Update).

For the records: It was the second occurrence only since 01/01/1990 only that the ES (E-MINI S&P 500) went into option expiration Friday on the heels of three consecutive sessions with a gain of at least +0.75% on the close, but luckily is was the 11th occurrence (which provided a favorable overnight opportunity on the short side, see my Twitter Update on the close) where the ES (E-MINI S&P 500) went into option expiration Friday after two consecutive sessions with a gain of at least +0.75% on the close (thereof 8 with a lower open on option expiration Friday, only 2 with a higher open, +0.20% and +0.68% respectively).

The ES (E-MINI S&P 500) opened lower -0.08% (which almost marked the low for the day), traded more or less flat during the first hour of the session, but contrary to what probabilties and odds supposed the market the market went then straight up and posted an intraday high of +1.08% above Wednesday’s close, and finally closed higher +0.78% on the day, while the Nasdaq 100 closed up +1.10% (S&P 500 +1.09%, DJ Ind. +0.76%, Russel 2000 +1.25%, SOX Philadelphia Semiconductor Index +1.02%, BKX Philadelphia Bank Index +2.87%).

Market breadth on the NYSE and NASDAQ was relatively strong (but not pertinent to the market’s positive performance, see setup S5 below), with NYSE Advancing Issues/Declining Issues at 2.75 and Advancing Volume/Declining Volume at 5.39 (NYSE TRIN / Arms Index at 0.51), and NASDAQ Advancing Issues/Declining Issues at 1.85 and NASDAQ Advancing Volume/Declining Volume at 2.45 (NASDAQ TRIN at 0.76). NYSE Advancing Volume accounted for 83.66% of NYSE volume.

____________

Notably on Thursday’s session were the facts that

  • Setup S1: the ES (E-MINI S&P 500) posted the 3rd consecutive session with a gain of at least +0.75% on the close,
  • Setup S2: the ES (E-MINI S&P 500) posted the 3rd consecutive session with a higher high (than the previous session’s high), a close above the open and a higher close as well,
  • Setup S3: the next session (Friday’s session) will be option expiration,
  • Setup S4: the ES (E-MINI S&P 500) posted the 2nd consecutive session with a gain of at least +0.75% on the close on those sessions immediately preceding option expiration,
  • Setup S5: more than 80% of the stocks included in the S&P 500 penetrated their previous session’s high (which marked a 10-day high).

I therefore checked -as always from a historical and statistical perspective- for the market’s performance on the then following session(s) after those setups listed above had been triggered in the past. Table I shows the ES (E-MINI S&P 500) performance (since 01/01/1990) on the next session (in this event Friday, August 21) immediately following those sessions where setups S1 to S5 listed above had been triggered in the past.

2009-08-20-ES-S1-5

Almost all setups (except setup S4 which shows a tendency for a higher close, but this is the setup which provided the favorable opportunity on the short side with respect to a lower open) are agreeing concerning their negative bias on the then following session, showing a (partly significantly) below-average probability for a higher close, a profit factor which (at least slightly) undercuts the at-any-time profit factor (sum of all winning trades divided by the sum of all losing trades), and an associated negative t-score close to or below (setup S1) the -1.645 mark for statistical significance (means there is a low probability only that the respective ES’ under-performance on a session after those setups had been triggered on close of the previous trading day in comparison to the at-any-time/average market’s performance and the negative performance at all occured by chance only).

Table II shows the ES (E-MINI S&P 500) intraday performance (since 01/01/1990) concerning the open, the first hour compared to the previous session’s close, the last hour compared to the first hour, the last hour of the session and the close (in chronological order) on those sessions (in this event Friday, August 21) where any of setups S1 to S5 (S1 or S2 or S3 or S4 or S5) had been triggered on close of the previous trading day, indicating the ‘overall‘ (including best case / worst case) bullish or bearish tendency on the then following session.

2009-08-20-ES-S6i

With respect to intraday stats (open/first hour vs. previous session’s close/last hour vs. first hour/last hour/close/, in chronological order) on those sessions (in this event Friday, August 21) where any of setups S1 to S5 had been triggered on close of the previous trading day, it is especially remarkable that the ES (E-MINI S&P 500)

  • shows an above-average tendency for a lower open,
  • shows a significantly above-average tendency (all with an associated t-score exceeding the -1.645 mark for statistical significance) for trading lower after the first hour of the session (in comparison to the previous session’s close), for trading lower between the first and the last hour of the session, and for a lower close as well (due to the significantly below-average profit factor, not with respect to the probability for a higher/lower close which is almost even)
  • shows slightly above-average odds (and a t-score close to the -1.645 mark for statistical significance) for a lower close.

There is a small but: The negative tendency is regularly not based on significantly below-average probabilities for a lower quote at the open, …, close, but mainly based on the below-average odds, means the expected negative payoff (probabilities in combination with the average -regularly negative- trade).

A favorable overnight opportunity on the short side of the market was provided at the close of Thursday’s session due to the fact that when the ES (E-MINI S&P 500) went into option expiration Friday after two consecutive sessions with a gain of at least +0.75% on the close (10 occurrences since 01/01/1990), the ES (E-MINI S&P 500) opened lower on 8 occurrences on option expiration Friday, and up on 2 occurrences only.

In addition, when the ES (E-MINI S&P 500) had posted three consecutive sessions with a gain of at least +0.75% on the close in the past, it opened lower on 10 out of the last 11 occurrences the next day. So with respect to historical probabilities and odds, this setup presented a highly favorable opportunity on the short side (Disclaimer: I already covered my shorts overnight at 997.75, with ES E-MINI S&P 500 trading lower -7 from yesterday’s close, but I’ll probably re-enter shorts if the ES E-MINI S&P 500 would be able to recoup a major part of it’s pre-opening losses before the open).

________________________________

Bottom line:

Based on historical probabilities and odds and the market’s regular performance on those sessions after any of setups S1 to S5 had been triggered on close of the previous trading day, the outlook concerning the ES’ (E-MINI S&P 500) performance on Friday, August 21 is negative right from the start of the session.

Therefore a favorable opportunity on the short side of the market may be provided in the event of any significant strength before, on or shortly after the open (from my current perspective at time of writing I don’t see any favorable opportunity on the long side of the market), or if the market would be able to recoup some or all of it’s overnight losses (trading -7.45 at time of writing, 00:40am CET).

Frank

P.s.: 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 mentioned in this post at time of writing (shorts already covered overnight).

Add to Technorati Favorites

Comments (5)

 

  1. Brian says:

    Fantastic stuff I read you every night and all the Tweets!

  2. JJ says:

    Market… just… won’t… comply. Ouch. :-)

  3. TheSmallFish says:

    hi frank,
    Are you seeing divergence of new 52 week highs between today and early august?
    Wondering if u have next day odds on this. If i recall it may be an st negative, but not sure about next day performance. If you are not able to, ill probaly do this weekend.

    Anyway, great work and days like this will happen(hey you never know, the day ain’t over yet!). We are in a rare environment here – stats/odds are more proned to failure in such environmen, but soon, we will be back to normal.
    Would be great if we could come up with an abnormal environment filter, eh? probably wouldn’t matter, since i think any such filter would only flag the 30s, possibly 99′, and now.

    • TheSmallFish,

      thanks for your kind words, but today’s session is disillusioning to say the least.

      Concerning the new 52-week highs, I regularly do not check for longer-term divergences. Additionally (and fortunately) I’ll be on vacation from tomorrow onward, I think it is the right time to not look at the markets 14 hours a day (minimum), and I regularly won’t have internet access during my vacation as well (and this time I promised to spend the time with my family, not with the markets).

      Best and thanks again,
      Frank

  4. Douglas says:

    Frank

    Good luck and thanks. I have been disillusioned and more and it happened to me earlier. I look at different criteria to yours and while I did not have a copper bottomed signal I had something that looked overwhelmingly bearish.

    In the past if you, Cobra and my own stuff were all ‘sell’ it seemed to me to be pretty strong odds.

    Perhaps what we need is some sort of filter with short and medium term moving averages where if beyond a certain strength then only the most extreme bearish signal should be acted upon.

    I do agree with the view that what is going on now cannot be compared easily with recent years. The data I work with goes back a certain amount of time…but it does not go back to the 1930’s.

    Best regards

    Douglas

Leave a Reply

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