I saw @ukarlewitz on twitter a few days ago discussing how down Fri, down Mon eventually leads substantial market decline. The last trigger of this signal was Monday 8/12/2013. Here is his post:
I decided to look at the short term impact of these down Fri, down Mon instances. It is very surprising to find that in any given year, having a substantial number of “Down Fri, down Mon” seems to have a bullish implication. I am talking about 90% winning if bought on Monday on “Down Fri, down Mon” day (selling using Oops in FAQ). If we then add a filter for 30 yr bond yield to be higher on Monday compared to Friday and 30 yr bond yield’s close is higher than open on Monday, we get 100% win rate on 49 trades (the longest trade is 19 days).