Is Abnormally Profitable Recent Performance On "Mutual Fund Mondays" Indicative Of Market Manipulation Or Just Herding?
Courtesy of Tyler Durden
With Erik Hane
Zero Hedge has previously discussed the bifurcation in market performance when comparing regular hour trading with that of the afterhours session, noting that in the September through December 2009 period, the market would have been flat if one were to strip away the benefit of gains after the market closed. Today, we take a look at a different set of data, namely observing a very peculiar market phenomenon associated with the term coined as Mutual Fund Mondays, especially over the past 6 months. Whereas on a long-enough timeline, the market performance on any given Monday (or, more specifically on any given first day of the business week), has averaged to about a 50% win/loss ratio, this is certainly not the case in the September 2009 – March 2010 period. In fact, during this time period, when the broader market has risen by only 10%, the positive contribution from Mondays has been 20%, implying that on all other days of the week the market has, on average, lost 10% in the past 6 months. Furthermore, the win/loss ratio for the first trading day in the last 26 weeks has been 85%: a nearly 3 standard deviations from the norm outlier. Let’s dig in.
To analyze this particular data set, we have split up the SPY performance from the past 52 weeks, and have primarily focused on the market performance (as captured by the broad SPDR ETD) on first day of any particular week. We notice two distinct regimes: one from March 1 2009 through the end of August, and a much more aggressive one beginning in September 2009 and continuing through today.
First, below we present is a price/volume chart of the SPY performance in the two periods under consideration.
While market volume is not a topic in this particular discussion, the secular decline as the market has kept going higher has only been interrupted by the brief correction in early February when the market was forced to undergo a breif 10% correction, which forced volume to spike up substantially. Now volume is back to normal.
Yet what we would like to bring your attention to are the following to charts, which scatterplot the performance of any given first week day mapped over time.
First – the period from March through September 2009:
And next, the period beginning September 2009 and continuing through today:
Here are the notable observations:
Between March and September 2009 the broader market increased by just under 50%, between September and March 2010 the market returned just one-fifth, or about 10%. This is well-known to everyone. Yet what may not be well known is that while historically, the win/loss ratio of the first day of any given week averages to 50%, and this is precisely the win/loss ratio for the period March-September 2009 (13 out of 26 weeks positive), the win loss ratio in the September-March interval is 85%, or higher on 22 of 26 Mondays (in some cases Tuesdays when the Monday is a holiday). In fact, the average daily return for "Mondays" in the first interval is a mere 0.09%, while in the second interval it is a stunning 0.75%! In fact, since early December, or 12/7/2009 to be specific, there has been just one "Monday" in which the SPY has closed down. In addition, the bullish performance on 1/19/2010, 1/25/2010 and 2/1/2010 are in stark opposition with the broader bearish downdrift that was taking place in the market during the period from January 15 and February 5. These three days account for 3 of the 5 green days in the this broader trading period.
Looking at this data in aggregate, in the second interval period, the cumulative return of just Monday has been 20%, or double the broader market return of 10%.
This data indicates that traders should buy the market (SPY) just before the open on Monday in the afterhours session (or just before the Friday close for those who have no access to AH trading) and to sell just before market close on Monday. If the trend discussed above persists, two months of this kind of repetitive activity will generate a 20% return all else equal.
And while this observation indicates that due to the high statistically significant nature of this outlier phenomenon the market is possible being nudged between the weekly close and the close of the first day of the following week from September 2009 through today, as there is nobody in regulatory capacity who either cares about this event, or would be willing to take any action against it, the best possible outcome would be for everyone to jump on board with the algo or outright mutual fund buying that is taking place at this time, and simply ride its coattails to power and riches. Then again, as with all technically trending patterns, that have no validation in reality or fundamentals, the persistence of this pattern will only continue until enough market participants spot it, and there is no greater fool to take advantage of on the other side of the trade (except for gullible taxpayers of course, represented in this, and every case, by the Federal Reserve).
