A recent study of historic “market” patterns points to a retest of lows.
Bespoke looked at bear markets since 1928 (the past 92 years) to determine how many times a bear market made new lows after initially rallying out of bear market territory. They found that markets made new lows after rallying more often than not (table below).
Bespoke also found that more recent results were less ominous. However, after taking a look at the Bespoke results, we have a different conclusions.
Original Bespoke data and results
- Number of bear markets counted = 25
- Number of bear markets making new lows = 14
- Percent of bear markets making new lows = 56%
- Number of bear markets since 5/14/40 to make a new low = 6
- Percent of bear markets making new lows since 5/14/40 = 42%
But Bespoke counted as separate bear markets periods that probably should be interpreted as just part of the same, but longer bear market.
If we make sure that a new period must be at least 12 months past the previous period, then there are only 15 separate, not 25 separate, periods and 9 of 15 bear markets made new lows or 60%. (see second annotated table 2).
Using this new definition of the market periods means that in the past 20 years (since 2000), ALL of the bear markets have made new lows (see annotated table 2) TPA thinks this analysis makes a lot more sense, since don’t we include the period from 1/6/09 to 3/9/09 in the 2008-2009 bear market? It does not make much sense to talk about these last 2 months as a separate bear market. This perspective on the historic pattern points to a much more likely retest of the 3/23 lows.