Bespoke had an interesting post this afternoon, Down 1.5% Two Days in a Row. They show what happened the next day. I wanted to see what happened looking farther out. See below for the answer.
This is a simple test. SPY closes down 2 days in a row, and both days lose 1.5% or more. What happens the next day?
I first tried to replicate Bespoke’s results. My results were very close. They had 31 trades (counting today) and my results had 30. This discrepancy may be due to the time frame they are using. They state, “Today marks the 31st time in the last ten years that the S&P 500 has been down 1.5% or more for two days in a row.” Therefore, I tested from 11.17.2001 to 11.17.2011. Bespoke may have used a slightly different period of time.
Being reasonably confident that we were both looking at the same data, I simply looked at what happened every day following the setup, for the next 20 days, and then averaged the results.
All buys and sells are assumed to have taken place at the close, with the buy made the same close as the day the setup is realized.
The Results:
The blue line is a close approximation of the test Bespoke published.
The red line uses the same time frame as the blue but requires SPY to have closed beneath the 200 day simple moving average. This requirement eliminated only 2 trades. Hence, this setup is characteristic of a bear market.
The green line is using all SPY data, which starts in 1993.
This is bullish over the short-term, but what happens if we use the same setup and look further out?
Looking out over 50 days, we see something that looks a lot like what we’ve been seeing over the past several months: a bear market trading environment.
Let’s see if we get the bounce tomorrow. Over 70% of trades closed higher the next day after the setup occurred.
Thanks! I really appreciate this work.
Wood, never. stop. posting.
Thanks.
Will do! Thanks Go2.
Awesome post and very informative, thanks!
Thanks! And coming from no less than a Prince!
Great stuff man. Always enjoy your posts for a good perspective on things.
Day 12 thru day 19 would seem to be the optimum swing buy time frame …
Do “days held” constitute market days only or calender days?
Nice work Wood, Thnx
Teahouse, those are market days. Thanks!
An average of 1% is meaningless if the difference between individual results suggest extreme risk. Two lines such as “Best 80% of trades above this profit” and “Worst 80% of trades below this profit” would be more revealing. It is the same as saying, “60% of trades were between these lines”.
Edg, pretty hard to mask extreme risk with a positive average and better than 70% winning trades, but I do understand your point.
Mr Woodshedder, I think you may have already mentioned this, I was wondering if you could tell me what platform/data feed do you use for testing?, thx
Interesting in light of the Mean Reversion posts recently. Good work!
Simple Conclusion
There are three reasons why stocks could rally:
1) Seasonality
2) Flash-in-the-pan type news from Europe
3) A bullish triangle (the market torched this scenario yesterday)
As mentioned earlier, there are plenty of reasons for stocks to drop, possibly even drop like a rock.
The trick to profiting from this scenario is proper risk management. The right time to sell long positions is either when the up side target is reached (that was at 1,275) or when support is broken.
If you sell and go short right after support is broken and place a protective stop-loss just a few points above support, you limit your risk to almost zero while prying open your profit potential.
Such key support mentioned over the past six months include S&P 1,325 (July 27 – S&P dropped 16% within the next ten trading days) and for those who missed 1,325, 1,298 provided a second chance (July 29 – S&P dropped 15% within the next eight trading days).
The S&P is now just above important support again; a drop below may set off another selling chain reaction.
Australian Dollar&, S&P are same index! almost tick for tick look at chart!
Rally Viewed as Corrective
.Companies:…AUD/USD . …RELATED QUOTES.
.Symbol Price Change
AUDUSD=X 1.0010 +0.0012
……
300 Minute Bars go to chart
Prepared by Jamie Saettele, CMT
The decline from 10750 is viewed as the first leg of either a 3 or 5 wave decline. In other words, lower prices are expected in the AUDUSD. This morning’s advance reversed sharply near the former 4th wave (10426), which is where one would expect a countertrend move to end. Coming under last night’s low of 10202 would shift focus to 10014 and 9880. Until then, 10500 cannot be completely ruled out.
I have to convey my respect for your kindness for all those that call for guidance on this 1 field. Your unique commitment to passing the answer up and down has been incredibly functional and has continually empowered a lot of people just like me to achieve their dreams.