Overnight follow through

Something that has been studied (or commented on) by Dr. Brett numerous times, how does the opening price in relation to the prior days closing price predict the direction of the days trade.  Confirming Dr. Brett’s findings, the answer is not much.  On a percentage basis looking at the difference between the prior day’s close and today’s open, the likelihood of it being an up day (closing higher than the open) is about the same as a down day. Separating the overnight percentage change into deciles,the median change between close and open (3rd column) is nearly 0%, with a large up move at open the only case that there might be a follow though (bottom row of the table).  The fourth column is median high – open percentage change.  The fifth column is the percentage change from open price to the low, six column is the total range, and the final column is the percentage that the move down is greater than move up (comparing open to high and open to low).

open – prev close close – open high open open – low range down > up
1 -1.16% 0.04% 1.31% -0.92% 2.50% 42.44%
2 -0.56% -0.05% 0.74% -0.85% 1.70% 50.00%
3 -0.27% 0.05% 0.78% -0.65% 1.56% 45.38%
4 -0.12% -0.01% 0.59% -0.73% 1.49% 52.52%
5 0.00% 0.17% 0.67% -0.71% 1.48% 49.16%
6 0.11% -0.07% 0.56% -0.78% 1.52% 57.98%
7 0.23% -0.01% 0.52% -0.78% 1.52% 57.14%
8 0.40% 0.03% 0.54% -0.77% 1.52% 57.14%
9 0.65% 0.00% 0.65% -0.90% 1.74% 57.56%
10 1.18% 0.31% 0.85% -1.10% 2.42% 51.05%
median 0.05% 0.02% 0.66% -0.78% 1.54%
std dev 0.61% 0.11% 0.22% 0.12% 0.37%

Some more observations – the intra day range is usually larger than the prior close/open change. It is fairly common to close any gaps (median high vs open is greater than open vs prior close when open vs prior close is less than zero, and median low vs open is greater than open vs prior close when open vs prior close is greater than zero).   I would interpret the last row (10th decile) to indicate a less than 50% chance the gap will be filled  (when the change between open and prior close is greater than about .8% – I think is closer to about 40% that the gap will be filled).   There have been much more exhaustive studies of the likelihood of filling the gap.

The last column would indicate a slight better odds of the size of the intraday down move (low vs open) is going to be larger for days that open higher.

Biggest take away – generally speaking, overnight changes don’t predict the direction of the day’s action from open to close.  Trading in the opposite direction of the market open will often result in larger intraday moves than the overnight move, excepting the most extreme of moves (up moves of > 1% still have a high chance of large intraday down moves (that are larger than the intraday up move), but it not enough to fill the gap most of the time).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: