The day after Thanksgiving, affectionately known as Black Friday, is an odd day of trading. For whatever reason, the exchanges open for business for part of the day even as most people still have vacation and/or shopping on the brain. Because of this partial trading and the accompanying light trading volume, the performance of stocks on Black Friday tends to get discounted as an anomaly. It turns out the data tell a much more nuanced story; traders should NOT discount Black Friday trading.

First, some basic numbers. The S&P 500 (NYSEARCA:SPY) closed with a healthy 1.3% gain on Black Friday 2012. Since 1950, the maximum gain on Black Friday is a 1.8% gain in 1971. The second highest is a 1.7% gain in 2007. The Black Friday performance for 2012 ranks as the sixth best since 1950. The chart below shows a distribution of Black Friday performances (click to enlarge images):

*S&P 500 Price Change on Black Friday (1950 - 2011)*

**Source: Stock prices from Y!Finance, Black Friday calculated as the day after Thanksgiving which is assumed to be the fourth Thursday of every November**

{Note that (x to y%] represents a range where x% is the lower bound but not included in the range, and y% is the upper bound and IS included in the range}

Notice the slight bias to large one-day gains over large one-day losses. The average performance is 0.35% and the median is 0.32%. With only 16 down days compared to 46 up days, betting on a Black Friday rally is a pretty good bet. HOWEVER, 3 of the 16 down days occurred from 2009 to 2011. This year, broke the longest Black Friday losing streak since at least 1950.

This tendency towards upside should be reason enough to pay attention to Black Friday. Yet, there are two more relationships to look forward to in the following week. Of the 62 Black Fridays since 1950, 36 of them (58%) are followed by Mondays where the S&P 500 closes lower than it closed on Black Friday. However, IF the S&P 500 manages to close higher on the following Monday, then odds are very high that the index will not look back for weeks and maybe months. The chart below shows the wide chasm in post-Black Friday performance.

*Number of Calendar Days It Takes for the S&P 500 to Close Below Black Friday Close (1950 to 2011)*

In the majority of years, the S&P 500 manages to close below its Black Friday close by the following week. If the index stays aloft, then the odds appear very good that it will take weeks for the S&P 500 to finally return. Special recognition goes to 1951, 1988, 1991, and 2011 (yes, LAST YEAR). The S&P 500 never looked back during these years, never closing below the Black Friday close. The stock market crash in 2008 finally took the S&P 500 below its close on Black Friday 2003, 1774 calendar days spent pushing upward and onward. This stretch is the longest for this data set; no other period during the 1982-2000 bull market ever came close (the longest was 47 days from 1995 to 1996).

The stats above were calculated without regard to whether Black Friday delivered a one-day gain or loss. Once I narrow the focus to Black Fridays with gains, the S&P 500â€˛s subsequent performance becomes even more interesting. This time, I count the number of calendar days it takes to completely erase Black Friday's gains. Of the 46 years with gains, 28 of them (61%) feature years where the gains are erased by the following week. Make it past this first week and the S&P 500 motors on weeks, sometimes months, without erasing those gains. Five years delivered Black Friday gains which have survived the test of time: 1953, 1958, 1962, 1982, and 1995.

*Number of Calendar Days It Takes for the S&P 500 to Erase Black Friday Gains (1950 to 2011)*

You likely missed betting on an up day for Black Friday 2012 given the (anomalous) streak of three poor year prior. The good news is that odds favor getting another chance to buy at pre-Thanksgiving prices. However, if this coming week does not deliver the lower prices, then prepare for an extended rally (like the legendary Santa Claus rally). This valuable lesson would have paid off very well last year.

**For those interested in some more formal statisticsâ€¦.**

Black Friday's one-day performance is statistically distinct from the performance of all other Friday's since 1950. I used a t-test with a two-tailed distribution and two-sample unequal variance (heteroscedastic), available in Microsoft Excel. For all other Fridays, the average one-day performance is .06%, the median is .09%.

*S&P 500 Price Change on All Fridays Excluding Black Friday (1950 - 2011)*

*To reduce chart clutter, I did not spell out the ranges of percentages. Instead, the numbers on the x-axis represent the upper bound of a 5-percentage point range. The lower bound is exclusive of the range, and the upper bound is inclusive of the range.

Be careful out there!

**Disclosure: **I am long SSO. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.