Last week I wrote about the edge that develops when the market makes consecutively lower closes. Excessive bearishness was the soup du jour, and yet the study showed a bullish edge.
Today (Wednesday) marks three consecutive higher closes, and the soup du jour is clover (bulls love clover), so I want to take a look for any edge that might exist.
The Method
This study will short the [[SPY]] at the close if the following two conditions are met:
1. The ETF will make three consecutive higher closes, AND
2. The three day rate-of-change will be greater than 5%.
The exit is a simple time-based exit.
100K per trade was used, with no compounding of gains. No commissions or slippage was added.
There were 22 trades generated from all of the available SPY data.
The Results:
The numbers running across the bottom of the graphs are the day on which the trade was exited.
Net Profit ramps up for the first three days and then begins to fade away.
The win/loss ratio is very volatile. This may be a consequence of the small data set.
Summary:
While the results do suggest that there is an edge, there were too few trades to convince me that it is robust.
I want to highlight the spikes in the graphs. Ideally these graphs would show a smooth fall-off starting from day three on. The volatility in the graphs point to the possibility that the best day to exit may change quickly and often. Again, if there were more samples, these graphs may be smoother, and I’d be more likely to trust the results.
Something that is odd is that this study also shows a spike around the 12 to 13th day, similarly to the 3 consecutive lower closes study. Maybe it is just a coincidence.
Disclosure: Short the SPY as of Wednesday’s open.
Danny also mentions the 3 consecutive up closes. His custom buy/sell strength indicators are one of the few indicators that I check nightly.
good stuff, wood! i love reading these test results.
I agree with Aris.. Thanks again Woodshedder! These are Awesome
link to my shit you bastahd.
Wood – I really like seeing the results — looks like next week just might suck… Thanks!
Thanks Aris and Mitch. I like the graphs as opposed to spreadsheets.
Danny, nice avatar. Hey, email me the link you want me to use. I looked at your site but couldn’t find the posts very quickly.
where the fuck did that avatar come from? not me ! not me! … oh. everyone has a little bug. i like it….
I would feel more comfortable shorting the close. Interesting none the less.
What about back testing a rising VIX on a massive up day? ( + another variable to make the test more relevant)
Might be useful?
Ketchup, this particular study did test shorting the close, but you would have had to shorted today’s close to get the fattest part of the edge.
Your VIX study was already completed:
July 15 (Bloomberg) — The VIX rose with the Standard & Poor’s 500 Index, a sign from the options market that the steepest three-day rally for stocks since June is poised to end.
The Chicago Board Options Exchange Volatility Index, as the VIX is known, added 1.7 percent to 25.44 at 2 p.m. in New York. The S&P 500 gained 2.4 percent. They have moved in the same direction 6 percent of the time since January 2003, according to data compiled by Charles Schwab Corp. The S&P 500 reversed course the next day 66 percent of the time, including seven of the past nine instances.
“That is remarkable,” Randy Frederick, head of trading and derivatives at Charles Schwab in Austin, Texas, said of the tandem move by the S&P 500 and VIX today. “The VIX is expecting something here, either a pull back this afternoon or tomorrow.”
Both indexes rose on July 6. The next day, the S&P 500 retreated 2 percent. The volatility benchmark, known as Wall Street’s “fear gauge” because it almost always increases as stocks fall, reflects expectations for price swings for the next 30 days and is calculated from S&P 500 options that are one or two months from expiration. Higher levels signal more risk in equities.
The government’s weekly report on jobless claims tomorrow may be spurring concern among investors, Frederick said. The U.S. unemployment rate has increased to a 26-year high of 9.5 percent.
“We’ve had some unexpected numbers recently,” he said. “There is probably an anticipation that could happen again.” Consumer confidence unexpectedly declined this month, according to a Reuters/University of Michigan index on July 10.
I mentioned it tonight.
http://ibankcoin.com/dannyblog/?p=3586
______
the articles are like brothers from different mothas
thanks Shed, appreciate it!
Well, when I go into battle, I want “Three White Soldiers” marching with me. No, that is not some racist comment from some nuckle-dragging neanderthal from Sacramento. Rather, this is a candlestick formation which tells of BULLISH conditions ahead – not bearish. The first of the three candles (in today’s world of color relationships – up is now green rather than white – the “white” came from Japanese candlesticks) is a reversal off a downtrend…..hey, didn’t we just have a 2week downtrend bottom before the first of the “Three White Soldiers”? To be valid, all three need to close near their highs, which was the case on the $SPX and $INDU. It is a sign of strength ahead. The pull back from this formation probably has more to do with consolidation/retracement after the moves than it does with it being a bearish indicator. Need to look longer out after the formation to see how the next week or two evolved. Nonetheless, your backtesting results and analysis are interesting.
Let’s see what kind of monster I get.
And, that didn’t work.
Wood,
I love your analysis, but get confused on the charts. In the Net profit chart, are you saying that if you short after the conditions are met and get out of the short at the end of the first day, second, third ,etc., the point on the chart shows the net profit or loss over x number of times.
So that getting out of that position after the third day according to your chart would be the best.????
H F-O, exactly.
Alvari, the edge is gone after 5 days.