Monday, October 22, 2007

Big Gap Down After a Big Down Day: What Comes Next?

My historical studies typically take the most distinctive aspects of the present market and then ask, "When this has occurred in the recent past, what has typically followed in the markets?" As it looks right now, the S&P 500 Index (SPY) looks set to gap open much lower following a very weak day on Friday. What has typically occurred when a large gap to the downside has followed a big down day?

I went back to 1996 (N = 2934 trading days) and found 26 occasions in which SPY was down by more than 2% the prior day and then gapped open to the downside by -.5% or more. From the downside gap open to the close of that same day, the market was up 15 times, down 11, for an average gain of .86%. The following day (from the close of the gap down day to the close of the next day), the market was up 18 times, down 8, for an average gain of 1.18%. All in all, from the downside gap open to the close of the following day, the market was up 20 times, down 6.

In short, there has been no bearish edge to selling into a sizable gap down when the prior day has been down sharply. Indeed, on average, the trader would have made money by buying the open and holding through the following day.

While that's not a pattern I'll be trading mechanically, it is one that primes me to look for buying setups during the day--especially price lows that are accompanied by fewer stocks making new lows and less extreme negative TICK readings. Should we not get those setups, that too would be an important indication, as it would suggest that the market is so weak that it cannot live up to historical precedent.


Short-Term Waterfall Declines in the Market


Roger Enright said...

What software or program do you use to do these studies?

Brett Steenbarger, Ph.D. said...

Hi Roger,

All my studies posted to TraderFeed are conducted in Excel using historical data from my real time feeds.


Brandon Wilhite said...

Dr. Brett,

Currencies gapped down last night sharply against the USD, but since then the USD has firmed considerably (contrary to what was implied on the radio report this morning). If USD had continued to sell-off after the gap, I would think that we might be in for another period of panic and volatility (esp. with the apparent sell-off in other markets last night), like I had mentioned in a previous comment. As it is though, it seems we're in a more orderly market right now. It will be interesting to see how the US Equities do today.


p.s. I'd love to see your Excel techniques. Whenever I do the kinds of studies you do, it seems just a tad clunky to me, so I'm thinking maybe you've got a better way.

Jeff said...

Thanks for the insight, Brett. I've been away from the markets for most of the year as other priorities needed to come first, but looks like I picked an exciting time to return. I'm looking forward to the Excel class.

Jeff Pietsch CFA, Esq said...

Also see MarketSci's (recently quoted in this blog) post where they find that exceptionally big down days together with exceptionally high volume lead to short term gains.

Brett Steenbarger, Ph.D. said...

Hi Brandon,

Yes, absolutely, my Excel stuff is clunky. I'm sure there are far more elegant ways to look at these things. I am so accustomed to cutting, pasting, sorting, and analyzing in Excel that it takes me little time. But clunky is a great word for it!


Brett Steenbarger, Ph.D. said...

Thanks, Jeff. If I can get out from under the workload with traders, I'd love to get the Excel class going!


Brett Steenbarger, Ph.D. said...

Hi Jeff,

Yes, that's a great piece of research. Those high volume declines often lead to reversal moves, though not always immediately and sometimes with a need to absorb some drawdown along the way--