Trading the Odds

A statistical approach to profit in the US equity markets, trading the markets like professional card counters are playing Blackjack or expert poker players are playing Poker.

Trading the Odds on Tuesday – May 19, 2009

WE031672-klein

It seems that the smaller the sample size and the more I emphasise that therefore one should be careful reading anything significant into it (like the bullish forecast for Monday’s session), the better the accuracy of forecast.

At any rate on Monday’s session we couldn’t have asked for more. The market perfectly complied to the bullish forecast (see my post Trading the Odds on Monday – May 18, 2009), the S&P 500 closed +3.04% on an exceptionally strong day, opening up +0.36% and continuously trading higher, and never looking back again. Breadth was exceptionally strong as well with NYSE Advancing Issues / Declining Issues at 7.54 and NYSE Advancing Volume / Declining Volume at 14.32.

One of the tradable (short-term, and on the short side) pattern triggered on Monday’s close (see my Twitter Updates) is based on the fact that the SPX triggered the ‘6 highs‘ setup, which means,

  1. the SPX opened higher (than the previous session’s close),
  2. the SPX posted a higher high (than the previous session’s high),
  3. the SPX posted a higher low (than the previous session’s low),
  4. the SPX closed higher (than the previous session’s close),
  5. the SPX posted a low ABOVE the previous session’s close and
  6. the SPX closed above the open.

(but it didn’t left an unfilled gap on the upside due to the fact that the low on Monday didn’t exceed the high on last Friday’s session)

The ‘6 highs’ setup shows -historically and statistically- a highly indicative tendency that the SPX will see a short-term set-back on the following session and will probably post an intraday low on the then next session on average -1.0% and more below the trigger day’s close (Monday’s close). That means selling the SPX short on close and covering the short sell during the next session (Tuesday) with a profit target of at least -0.75% below the trigger day’s close provides a significant short-term edge.

Table I below shows the SPX‘ intraday performance (since 01/02/1990, but the last 50 occurrences only) concerning the open, high, low, close (compared to the previous’s session close) and close verus open on the next session (in this event Tuesday, May 19) immediately following those sessions where the ‘6 highs’ setup had been triggered.

20090518-SPX-2

It is especially remarkable that in only 1 out of the last 25 occurrences did the SPX post an intraday low on the next session which was not at least -0.75% below the previous session’s close, and in 21 out the last 25 occurrences did the SPX post an intraday low at least -1.0% below the previous session’s close. Additionally ALL intraday stats (see Table II below) concerning the open, high, low, close and close versus open show a profit factor partly significantly below the respective at-any-time profit factor, which means going long the SPX on close of a session where the ‘6 highs’ setup had been triggered would have been an unfavorable short-term setup concerning the next session, especially due to the fact the any upside potential on the close is limited as well (below-average expectancy).

Table II below shows the SPX‘ performance summary (since 01/02/1990) on the next session (in this event Tuesday, May 19) immediately following those sessions where the ‘6 Highs‘ setup had been triggered.

20090518-SPX-1

________________________________

Bottom line:

  1. The outlook for the remainder of the week is still positive (see my post Trading the Odds on Monday – May 18, 2009), although a short-term set-back is probable on Tuesday’s session (with at least an intraday low significantly below Monday’s close).
  2. Monday’s close provided a favorable short-term setup by selling short the SPX on the close with a profit target of at least +0.75% (an intraday low of at least -0.75% below Monday’s close) during Tuesday’s session, see my respective Twitter update.

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

Advertisements

Filed under: Daily Update, , , , , , ,

6 Responses

  1. manatrader says:

    Never stop blogging, ever. Please :)

  2. Sam says:

    Frank,

    I enjoy your blog. It is thought provoking and I appreciate that. I noticed many of your studies are based on recent history, but when looked at over a longer period of time they do not hold up as well. Please realize this is not a critique of your models or anything of the sort, as there is no right and wrong way to make money from the markets. My question is how do you decide to stop trading a particular model?

    I ask because as a developer and trader it is a question I enjoy hearing different opinions on and truthfully it is a question that in my mind is still unanswered. Some people throw things out after a systems breaches a certain level of draw down beyond historical, some throw them out when it simply appears the edge has been worn away, some never throw them out and simply put them on the shelf for later when they under perform and some even say it is intellectually dishonest to ever throw out a system if it is truly a viable system in the first place.

    I would enjoy reading your response.

    • Frank says:

      Sam,

      thanks a lot for your kind feedback.

      Most of my studies go back to 01/02/1990, and from my perspective a 20-year history should be more than sufficient in order to determine probabilities and odds -as long as I come out with a hopefully statistically significant sample size- concerning a specific setup, but as I already mentiond even a small sample size which could as well be purely random should be at least taken into account (for probably the lack of any better decision basis) . And if in doubt, I’d always give the most recent history more weigth than any time period several decades in the past.

      First of all although I trade a particular model on one of my (smaller) accounts, I like to capitalize on whatever provides an edge in the market (positions are disclosed in my daily posts). The daily setups do not reflect any kind of ‘model’, they are nothing else than statistical abnormalities in the markets regularly solely based on price and breadth data with historically above or below probabilities and odds concerning the next session(s) performance.

      But to be honest trading the markets (e.g. SPX) based on daily setups, probabilities and odds is the smaller part of what I mainly do in the markets. The bigger part is based on statistical arbitrage, so I’m not the system developer and mechanical trader you’re probably looking for.

      Thanks again.

      Best,
      Frank

  3. Sam says:

    I would say 1990 would be more than enough data. I was just looking at this post and quickly ran the study back to 2001 and noticed that since some time in 2008 it looked great but prior to that did not look so great. Now of course I was running it on SPX, but having it execute on ES so there may be some differences there. Also I just had it exiting close of next day so that may be a difference. I am not sure. Again, not questioning just researching.

    As far as throwing models out, even as a stat arb trader there must be times when your models deviate far from the historic norm, right?

    Appreciate your previous response.

    • Frank says:

      Sam,

      from my point of view utilizing on one asset for potential setups, but then executing any potential tradable edge on another asset (if this is not the real purpose of the system), will give you some headaches to say the least (if not a receipt for desaster, even if both assets should be highly correlated). I’m going for an intraday low today, not for the close of today’s session (as stated in my post).

      You’re correct, there are times when the models deviate from historic norms, but at least during the last 4 years (when I begun with stat arb) it was always a question of time and patience until mean reversion set in. Questioning the model or getting impatient was always the biggest mistake I could have made. Fortunately up to now I was never forced to modify the model or to completey throw it out.

      Best,
      Frank

  4. Sam says:

    Frank,

    Thank you for your response. It makes sense that if a model is truly viable then throwing it out is a mistake. However, I have known “traders” to trade what they consider a temporary edge and then dump the system when it stops performing. They don’t stick around for too long for obvious reasons.

    I know what you are saying regarding one asset executed on another asset. When I first ran it I ran it directly on the ES and then directly on the SPX then settled on deriving the signals from SPX but executing on ES as SPX represents the cash index. The results were all pretty similar taking multipliers into account.

    Anyhow, the big difference is looking at intra-day low. Obviously I went to close, as it is impossible to know the low until the day is over.

    That’s neither here nor there. Thanks for the chat.

Leave a Reply

Fill in your details below or click an icon to log in:

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

Twitter Updates

  • w/ 18.30 at time of writing, the gap between $VIX and $VSTOXX is close to its all time closing low of 19.90, posted on 10/16/2008. 1 year ago
  • The $VIX gained 48.33% over the course of the last week. Since 1/2/1990 there were 38 other occurences w/ $VIX gained > 48% over 5 sessions. 1 year ago
  • On Friday iShares MSCI Brazil Capped ( $EWZ ) closed at 19.09 , a 11+ year low and its lowest level since 11/08/2004. 1 year ago
  • On Friday the Russell 2000 Index ( $RUT ) closed at 1,046.20 , a 2+ year low and its lowest level since 10/09/2013. 1 year ago
  • @QuantStratTradR Sorry, I didn't follow the entire communication. Link to what ? // @easyvolatility 2 years ago

DISCLAIMER

The information on this site is provided for statistical and informational purposes only. Nothing herein should be interpreted or regarded as personalized investment advice or to state or imply that past results are an indication of future performance. Under no circumstances does this information represent an advice or recommendation to buy, sell or hold any security.

The author of this website is not a licensed financial advisor and will not accept liability for any loss or damage, including without limitation to, any loss of profit, which may arise directly or indirectly from use of or reliance on the content of this website, including the information that others post here.

While every effort will be made to provide complete, the most accurate and current information, none of the information on this site is guaranteed to be correct, and anything written here should be subject to independent verification. I make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability or availability with respect to this blog or the information, analysis, statistics, or related graphics contained on the blog for any purpose.

I may or may not hold positions for myself, my family and/or clients in the securities mentioned here. Actions may have been taken before or after information is presented, and any opinions expressed in this site are subject to change without notice.

Please read the full ... DISCLAIMER

Calendar

May 2009
M T W T F S S
« Apr   Jun »
 123
45678910
11121314151617
18192021222324
25262728293031
%d bloggers like this: