This is now my third time writing about the tightening in the daily range of the S&P. The market has been stuck in tight range for a couple weeks now. Just the other day I had written a blog stating that it had been the 8th consecutive trading day of tightening in the daily range. I define tightening as days when the range ends up being less than the 20 day average daily range (ADR). Yesterday the S&P futures traded in a 10.75 handle range, another narrow range day, marking now the 9th session in a row of daily range tightening.
I re-ran my model, again using S&P futures data from January 2000, to look for all instances when we have seen nine consecutive sessions of narrow range trading (i.e., the actual range is less than the 20d ADR). I want to present a side by side comparison of the model results (what happens on the subsequent 5, 10, and 20 days) when I run it for eight consecutive sessions vs. nine consecutive sessions:
The main item here that I want to point out is that the probability of the market moving higher for all time periods drops when the analysis is run for nine consecutive days (as opposed to eight). Particularly, we see that following eight consecutive days of tightening the probability of the 10d change being positive was 43.1% and for the 20d change was 41.7%. Compare these figures to the 10d and 20d change following nine consecutive days of tightening: 38.5% and 36.5%, respectively. The decline in the odds of the market going higher seems to be part of a larger trend- the longer we stay stuck in a tight range, the higher the probability of a pullback. The first time I ran this study was when the S&P had been trading in a narrow range for just five days in a row. Take a look at what the odds results were previously:
The probability trading higher in the subsequent 10d and 20d was 57.4% and 51.7%, respectively. Again, note that the odds of the market being up go lower the longer we remain (over consecutive days) in a tight range. The average (and median) changes help confirm this trend as those numbers point lower as well. In particular, the 20d average change and median both lie around -3.3%, a noticeable drop. Initially, the 50/50 odds generated after five consecutive sessions gave me no conviction to either side. But now after nine days of tightening in the daily range, with the odds favoring the market being down in the next 10 and 20 days, I will say I believe this slow down in the market will result in a pullback and not a resumption of the uptrend.
*If today ends up being yet another day of daily range tightening, and it appears that it will as we have been trading in a 11.75 handle range (the 20d ADR will be approximately 16 handles and change), I will re-run the model to accommodate for 10 consecutive days.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.