Continued from Part 1
Synopsis: While Part 1 of this study showed a modest negative correlation between Mercury retrograde and the stock market, Part 2 tests the alternative hypothesis of trend reversals. That is, changes in Mercury's direction from forward to backwards and vice-versa may be correlated with changes in prevailing price trends. Based on the sample size of the 79 most recent Mercury retrograde cycles, this study found very limited evidence to support the trend reversal hypothesis. Of the four possible reversal scenarios, the bearish reversal following the retrograde station had the largest effect.
Part 2: Disaggregation by price trend
Using the same logic as my Venus retrograde study, I divided the cases into batches of rising and falling price trends. My thinking here is that the retrograde and direct stations of Mercury might coincide with reversals in the prevailing price trends, up or down. The retrograde and direct stations occur about 24 days apart and offer a plausible way to measure changes in trend. We should note, however, that this is a different hypothesis from that explored in Part 1 which focused solely on the potentially negative influence of the Mercury retrograde cycle.
Whatever the intrinsic influence of retrograde Mercury, it is also possible that the retrograde cycle could operate as a neutral timing pivot. Recent experience would seem to support this notion as the interim low on August 5, 2024 came on the same day that Mercury stationed retrograde. Stocks then rose sharply during its subsequent retrograde cycle. In that instance, Mercury retrograde was actually very bullish as the three-week pullback from mid-July to early August occurred during the preceding 'shadow' period while Mercury was still moving forward in the sky.
Of course, this is only one case. We need to find out how reliable this timing pivot idea might be by testing it across many cases. In order to maximize the size of any potential reversal, I limited my case selection to larger preceding moves of greater than 1%, either higher or lower. I assumed that if the station coincided with a price pivot, it would be more likely if price was clearly trending higher or lower beforehand.
Scenario 1: Bearish Reversal after retrograde station (RxS)
The first batch of cases tested the most plausible scenario: if stocks were in an up trend before the Mercury retrograde station (RxS), then we could measure the extent to which stocks declined afterwards. The criteria for case selection was choosing cases that featured a larger than 1% rise in the 6 days ("-6dr") before the retrograde station. My assumption here was that the shorter intervals of 2 days before or 4 days before were too brief for a trend to be established. This was especially true since I am using calendar days and not trading days. If the time interval or retrograde station fell on a weekend or holiday, then these shorter time intervals may not sufficiently long enough to reflect the preceding price trend.
The other criteria for selection for positive price trends for most of the other time intervals leading up to the retrograde station. This excluded the selection of cases where a positive price change appeared in only one or two intervals, including the 6-day interval. If price fluctuated too much before the retrograde station and even went negative, then that suggested a weaker trend that might be less likely to reverse after the Rx station.
The first thing we should notice is that these criteria only yielded 14 cases out of a total of 79. Clearly, a strong upward price trend across 6 or 8 days before a Rx station is not that common an occurrence. We should keep this in mind when trying to extrapolate the results given this limited applicability.
But we can see that in the days following the Rx station, stocks did on average move lower. The most concentrated down move occurred with the 6 day time interval after the station ("RxS 6dr") which saw an average decline of -0.65%. A slightly larger decline of -0.78% occurred across a much longer time interval of 17 days. This interval is labeled "RxS 17dr". Using 20-day interval after the retrograde station ("RxS 20dr"), the average decline fell to -0.61%. It fell further to just -0.28% when considering the entire retrograde cycle of these cases ("RxS DS")
Shorter time intervals after the Rx station were less negative although the 4-day interval did produce 10 negative outcomes out of a possible 14. This was the largest number of negative results of any interval. But with the other intervals only registering 8 negative outcomes out of 14, it is clear that this bearish reversal effect is not strong. If the strategy of 'going short' at the retrograde station finds only limited support, there is a better argument for at least reducing long exposure at the time of the retrograde station given the increased downside risk.
Verdict: a weak to moderate effect
Scenario 2: Bullish reversal after the retrograde station (RxS)
Now what about the opposite scenario when a negative price trend precedes a Mercury retrograde station? This is what happened in the first week of August 2024 as stocks had been falling for several weeks ahead of Mercury turning retrograde. Stocks bottomed the same day as Mercury reversed its direction and then began a strong rally. In order to test the viability of this bullish reversal thesis, we needed to isolate all the cases in which stocks were up strongly (>1%) over the preceding 6 days before the station. In all, there were 14 cases which met this criteria with an average decline of 3.22% during the 6 calendar days before the Rx station.
The average effect was weak, however. Only the 2-day interval after the Rx station exhibited a positive reversal, and even then only 0.11%. The 6-day interval also showed a marginally positive result (0.08%) although this is also negligible. The other time intervals were actually negative, suggesting that down trends before the Mercury retrograde station may tend to continue afterwards. However, we should note that a majority of cases were positive after the RxS, with most intervals having 8 or 9 positive outcomes out of 14 cases. It's clear that the negative outcomes tend to be large enough in size that they nullify the effect of the positive reversals.
Even if we acknowledge these majority of positive outcomes, the overall effect seems weak on average. This is not to say that the August 2024 bullish reversal was a one-off coincidence. But it suggests that other planetary factors need to be considered for this scenario. But there is reason to think the strength of the Mercury retrograde station alone is not strong enough to warrant its use as a trading strategy. Perhaps this is an argument for a more conservative strategy which reduces short exposure going into a Mercury Rx station during a market down trend. Certainly, the abrupt shift higher in the number of positive results (in 8 or 9 cases) suggests that the down trend is more likely to end following a Mercury retrograde station, even if the market is not strongly bullish thereafter.
Verdict: weak effect
Scenario 3: Bearish reversal after the direct station (DS)
Using the Mercury direct station (DS) as a potential pivot, we considered the possibility that stocks could reverse lower after an up trend. Remember that the direct station occurs about 22-24 days after the retrograde cycle as Mercury resumes its normal forward motion. Most financial astrologers would likely see the period after the direct station as having a bullish bias but let's see to what extend it might act as a neutral pivot point.
Using the same 1% criteria as a strong up trend 6 days before the station, we created a batch of 14 cases. In one case (the last one from Oct 2001), I bent the rules a bit and accepted a >1% change for the 4-day and 2-day intervals as qualifying for this scenario. The average gain 6 days before the DS was 2.39% and 1.82% for 4 days before the DS. While we observed a rise in the number of negative cases (from 0-1 to 5-7), we did not see an average decline after the direct station. Instead, the average change was merely somewhat less than before the DS, the 2-day interval ("DS 2dd") equal to 0.40% and the 4-day interval ("DS 4dd") change of 0.46%.
Clearly, the effect of the direct station was weaker than the retrograde station both in terms of the raw number of reversals and in terms of the average change. Even the lesser strategy of reducing long exposure in anticipation of lower returns finds only limited support in this test.
Verdict: little to no effect
Scenario 4: Bullish reversal after direct station (DS)
Finally, let's see if there is any evidence to support the commonly-held view that the Mercury direct station can be bullish after a down trend. Cases were chosen where a clear down trend was in effect before the direct station. I included instances of >1% decline in at least two out of three time intervals, 6 days, 4 days, and 2 days before the station. This batch included 17 cases with an average decline of -2.16% over 6 days before the DS and -2.55% 4 days before the DS.
This bullish reversal effect was stronger than the above bearish reversal scenario with more than half (11 out of 17) of the case having a positive outcome 2 days after the direct station. The upside was still quite minimal, however, with the average gain just 0.34% for the 2-day interval. All other intervals showed smaller effects and some were negative, such as the 6-day and 15-day intervals.
This effect must be considered weak at best. The more conservative strategy of reducing short exposure at the direct station finds some support here since a majority of those cases ended up positive. Thus, the bulk of the downside was concentrated in the time before the direct station implying that traders who were short may consider closing out their positions given the greater risk of a bullish reversal.
Verdict: weak effect
Conclusion
This study suggests the Mercury retrograde may only have a limited influence on stock market returns, even using the disaggregation approach based on prevailing price trends. While there is some evidence that price trends are more likely to reverse at the time of the retrograde and direct stations of Mercury retrograde cycle, the immediate effect seems fairly small. Investors should therefore be cautious about estimating the size of any potential reversal. This is worth remembering as we approach the next Mercury retrograde station on Saturday, March 15, 2025. A trend reversal after that date may more likely than chance would predict -- up or down -- but its probability is not convincingly high.
Of the four scenarios considered here, the strongest effect was the bearish reversal after the Mercury retrograde station. This was perhaps an unsurprising confirmation of the widely-held view about Mercury's apparent backwards motion. But while there was evidence for it, the effect was quite modest (less than 1%) and there was a fairly high failure rate (29-43%, depending on the time interval).
Although the size of the reversal was quite small, there is nonetheless evidence to support the conservative strategy of reducing positions (long or short) at the time of the retrograde and direct stations. If stocks have been trending higher before the station, then they are less likely to continue trending with the same momentum. Moreover, they are increasingly vulnerable to reversal. The same is more or less true for down trending markets as they are somewhat less likely to continue to decline after a Mercury station and are more likely to reverse higher, albeit only modestly.
N.B. For a discussion of the current decline in the stock market, please see my March 10 post on the NYSE horoscope at my Modern Vedic Astrology website.
