Does the 1% Loss on the Final Day of the Year Give Any Clues About January?
by Jerome "Mel" Hickerson
Looking at historical data, does the 1% loss on the final day of the year give us any clues about January?
Thursday's late day plunge certainly changed my plans on what I was thinking to write for this weekend update. I not only totally missed seeing that downward move coming, I knew it was statistically a rare event. So I shelved the plans for this article and switched gears because of the end of year action.
I went to the historical data with several things in mind:
1) How often do we see a 1% drop on the final day of the year?
2) How often do we see a 1% drop during the "Santa Rally" timeframe?
3) Are there historical patterns within the data that might lead us to draw any worthwhile conclusions from the action?
The first thing necessary is to define what time period we are discussing within the term "Santa Rally" timeframe. There are many definitions around; the Stock Traders Almanac defines the time period as the last five sessions of the year and the first two sessions of the new year. I settled on a definition very close to that one; my definition used here is the session before Christmas through the first two sessions of the new year. This will almost always be the same as the Trader's Almanac definition but a few years it will include an extra session.
All data used is from 1950 forward, using the SPX. So including this year, we have 60 years to examine.
Raw data:
• There were 436 trading sessions during the Santa Periods.
• 24 sessions closed 1% or more negative (5.5%).
• 53 sessions closed 1% or more positive (12.2%).
• 35% of the 436 ended negative; 65% ended positive.
• The Santa period was negative 14 times, positive 45 times.
• The Santa period was 1% or more negative 9 times, 1% or more positive 35 times.
• The most negative Santa period was -3% (01/1991); the largest positive period was +7.38% (01/2009).
• Contrary to everything I've read, a negative Santa Rally period led to a negative January only 50% of the time.
• A positive Santa Rally led to a positive January 65% of the time.
• 20 of the 60 years ended on a negative session; 25% led to a negative January.
• 40 of the 60 years ended on a positive session; 45% led to a negative January.
• 4 times in 60 years the year ended on a 1% or more negative session (1996, 2000, 2001, and 2009). The following January ended +6.13% (1997), +3.86% (2001) and -1.56% (2002).
• 3 times in 60 years the last day of the year ended on a 1% or more positive session (1957, 1974, 2008). The following January ended +4.28% (1958), +12.28% (1975) and -8.57% (2009)
• 23 of the 60 Januarys ended negative (38%) and 37 ended positive (62%).
• Santa periods with both 1% gains and losses in the same year led to a negative January 4 times of 9 occurrences.
• Santa periods with at least one 1% negative session but no 1% positive sessions led to a negative January 5 times of 7 occurrences.
• Santa periods with at least one 1% positive session but no 1% negative sessions led to a negative January only 5 times in 20 occurrences.
Now, lets see if we can draw any corollaries from the available data.
• There is a 2-1 bias for positive closes during this time frame. This is statistically significantly greater than the 52.7% of all closes since 1950 which have closed positive.
• Closes during the Santa period with a 1% loss are uncommon. The 5.5% occurrence is statistically significantly different than the 10.49% of all sessions since 1950 which have closed down 1% or more. The 12.2% of positive closes during the Santa period does not statistically significantly differ from the 9.98% of all closes since 1950 with a 1% or greater gain.
• The entire Santa Period gains three times as often as it loses (76%). This is a statically significant difference from a random seven day period which closes positive only slightly more than half (53%) of the time.
• Individual sessions are more likely to close positive (65% vrs 52.7%) during than this time frame than random.
• January has a positive bias (62% positive). But this falls to 50% if the Santa period is negative.
• When the final day of the year closes negative, January closes negative only 25% of the time. When the final day of the year closes positive, January closes negative 45% of the time. This is mildly significant statically but due to the small sample size does not allow us to draw any relevant conclusions.
• There seems to be some statistical relevance to the 1% losses and gains when there are no offsetting 1% gains and losses during the same period.
Conclusions:
• History suggests a positive bias during this time frame; this includes the next two trading sessions.
• The year ending session Thursday, while certainly a little eye-popping, does not lead us to draw any negative conclusions from the historical data. If anything, the data weakly suggests a positive January based on the year-ending close.
• The axiom "If Santa should fail to call, bears come to Broad and Wall" seems to have only a weak basis in the data. The failure of a Santa Rally leads to a negative January 50% of the time versus a 62% positive bias for all Januarys since 1950.
• The Santa period this year has seen a 1% losing session without (yet) a 1% gaining session. Our 59 year sample of data shows that 71% of the time this leads to a negative January versus the 62% positive bias for all Januarys. We are dealing with small samples of data but this seems to bear watching.
A lot of numbers but few worthwhile conclusions. Mostly, this shows that the losing session on Thursday may have little or no significance going forward. The last bullet under Conclusions may be the one thing worth keeping an eye on.
Have a safe, happy, and prosperous New Year everyone!
-----------
"Mel"
ETF’s we trade:
Ultra S&P500 ProShares (NYSE: SSO)
Ultra Dow30 ProShares (NYSE: DDM)
Ultra QQQ ProShares (NYSE: QLD)
PS UTLRSHRT QQQ (NYSE: QID)
UltraShort S&P500 ProShares (NYSE: SDS)
UltraShort Dow30 ProShares (NYSE: DXD)
PowerShares QQQ Trust (NASDAQ: QQQQ)
Direxion Daily Small Cp Bear 3X (NYSE:TZA)
Direxion Daily Small Cp Bull 3X (NYSE:TNA)
