Is Price Behavior Different in Bull Market's vs Bear Markets? The answer is Yes. And No.
While there are clear differences in price behavior in Bull vs Bear Markets, most notably the longer term trends of up vs. down in Bull Markets vs Bear Markets. Shorter term price movements among historically correlated price conditions tend to follow the same behavior patterns.
We performed back tests on ten diverse Price Conditions in the S&P 500 using Quantitative Trading Strategies. We conducted this on a Think or Swim trading platform from Jan 3 – May 13 in OnDemand mode. This subject test period was relatively volatile and bearish as compared to the last four years. The S&P 500 was down about 17% during this period and provided a strong test against the historical data.
The largest variations in correlation of a condition occurring were due to the number of events and not the time or the volatility. Price Conditions which had more than 50 events over the past four-year period had very little variation to the test sample by year or volatility. Following are key observations from that study:
Despite the higher-than-normal volatility, win/loss percentages were consistent with correlation levels 80% of the time
All tests and trades performed were used in the analysis. Trade 4 results showed the poorest relationship, but only had 7 trades that were applicable during this period and as such may be discounted
Some variation between the correlation levels and win rates is due to factors in the trading methodology. For example, debit spreads may have a higher win rate than naked calls/puts although the ROI is not as good. Similarly, credit spreads may have a higher win rate versus debit spreads on the same price condition but a lower overall ROI
A range of different Price conditions were used to provide a broad sampling of conditions and trade types. These ranged from one day, multi-day, & multi week, to price reversals, credit spreads and naked calls & puts
All ten price conditions showed very strong positive ROI despite the negative market conditions and relatively high volatility
Trade size was kept constant at $2,500 although some opportunities warranted smaller or larger amounts to aid in managing trade risk
All trades followed strict rules for opening & closing – no subjectivity was allowed
As discussed above the two test deviations on correlation occurred where the number of events had the highest divergence. Given the test sample during the Jan - May period was relatively small as compared to the historical data there were relatively few events in tests 4 & 9 and results are not conclusive. However in the remaining 8 tests the correlations were very similar despite the difference in market conditions.