Expiry Fridays - Bullish or Bearish?

I wrote this Expiry Fridays article about weeks with Expiry Fridays which shows that if measured by the DJIA then those weeks are generally more bullish than your average week.

The average gain per week in the DJIA from 1986 to 2005 is 0.205% but if you only look at weeks with expiry Friday's then that gain is 0.326% and if you exclude weeks with expiry Fridays then the gain is 0.195%.

If you want a copy of the spreadsheet that I used for calculating this then just reply to this topic and I'll email it to you in a ZIP file. (You do NOT need to include your email address in the reply.)

we are in liquidating markets
All taken before the closes and in some cases there was more pain these are some examples of just todays action
gold -6%
silver -11%
plat -3%
pall -9%
Copper -6%
Alli -3%
lead -3%
Nickel -7%
Zinc -8%

Russia -9%
India -4%
Korea -3%
Japan -4%
Chile -4%
Colombia -8%
Europe down 2 to 3%
etc etc
They have to sell the good to pay for the bad.
Liquidating markets tend to create and instill panic. There is no way to tell when the last guy gets out for that is in effect the only way that a bottom can occur. In MP terms we have to drive the market such that we shut off the selling and boy is there still some way to go.

cc the Chart forum

I agree, it doesn't look like it's going to stop soon. I don't recall if my study gave the best and worst week or just averages...

Daytrader perhaps you could isolate your stat so that the bear markets are picked out and see what happens then you can compare the good with the bad . Just a thought.

I've just done a little bit of further analysis on that spreadsheet that's posted with the article and have the following info:

The worst 3 weeks during the study period were:
09/21/2001 at -14.0%
03/16/2001 at -6.6%
09/20/2002 at -3.9%

and the best 3 weeks were:
03/17/2000 at +8.8%
03/21/2003 at +8.5%
12/18/1987 at +4.3%

Remember that we are only looking at expiration Friday weeks in this study. Expiration Friday's are defined as the third Friday of each quarter month (Mar, Jun, Sep, Dec) and as such happen 4 times a year.

Now here's a crazy but fascinating piece of information.

This study covers 20 years of data in order to collect enough expiration Friday weeks to make the study significant because there are only 4 of these weeks each year.

Looking across the returns over all the weeks it turns out that the best 2 weeks and the worst 1 week were also expiration Friday weeks.

So the 3 worst weeks during this study period were:
09/21/2001 at -14.0%
10/16/1987 at -8.5%
10/13/1989 at -8.0%

and the 3 best weeks were:
03/17/2000 at +8.8%
03/21/2003 at +8.5%
10/30/1987 at +8.4%

We are now talking about all weeks in the study period and not just expiration Friday weeks.

Correct me if I'm wrong, but I've just looked at last week's change in the S&P500 cash index and it's a drop of 2.78% and this week, so far as I type this at 9:34am on Wednesday morning, is a drop of 2.07%. Neither of these drops appear to be very extreme when taken in context of the big drops we've seen above.

oh they are not . US stks are in fact benefitting from a flight to quality. ie out of emerging markets and bringing the money back home
PS russia lost a further 10% today.
So what one perhaps should look at is the 15% of Russel or the levels of NAZ which are below the dec 05 settlement and the dec 04 setllement and within 57 points of taking the dec 2003 settlement

This makes me think that perhaps there is a viable trading system out that can be developed that flips between markets depending on the volatility seen in those markets. At the beginning of the year we had those 3 or so weeks with very tight range (recording setting tight range) and now we have unusually high volatility (especially for summer). From one extreme to the other in the S&P500 market.

So what I'm thinking is that there is probably a strategy that would have worked on the ER2 or NQ during those low volatility weeks at the beginning of the year which would now work on the ES. In other words the ES is now behaving like the ER2 and NQ would behave under normal conditions.

If you add in the MC and YM you probably have enough of a collection of liquid futures contracts that you could skip between contracts depending on which is most profitable for your tested strategy. In this case you would have a volatility setting under which your strategy was best suited and then you would trade the index which most closely (currently) matched that volatility profile.

Comments...?

If you had told me 2 days ago that this market had the possibility of being a bullish week I would have laughed at you. Today we are up on the week and so bullish is definitely in the cards...

Well it was an interesting week with two big back-to-back down days on Monday and Tuesday and then a remarkable recovery mostly on Thursday. The range for the week in the S&P500 cash index was 43.29 points and the change on the week was a drop of just 0.73 points so not even 1 point.

As I mentioned in the last posting, after Tuesday evening I never thought that there was a chance that this triple witching week would end up with a chance of being an up week. Well it didn't end up being an up week but it was very close.

Triple expiration this week.