Good Black Swans

T

tomorton

Most of us will be aware of the danger of Black Swans - in trading, an unexpected, sudden and extraordinary price move. But if they can't stopped and they can't be predicted, can we do anything to be on the right side of a market Black Swan? So when price crashes, we're already short. Or when price rockets, can we be already long?

I'm going to say most Black Swans in trading could have been profitably anticipated (I'm not saying predicted) by being long when price and the 20EMA were above the 50EMA, or by being short if price and the 20EMA were below the 50EMA.

Some classic bear examples -
EUR/CHF, 20/01/2015 - closed -17.2% on previous day. But the 20EMA had crossed below the 50 in June 2014.

USD/JPY, 28/04/2016 - closed -3.08% on previous day. The 20EMA had dropped below the 50 in December 2015.

GBP/USD, 27/06/2016 - closed -3.03% on previous day. The 20EMA had dropped below the 50 the previous day - this was the Brexit result day and although the rule didn't anticipate the result (who did?!?), it would have stopped further losses.

Dow Jones, 29/09/2008 - closed -6.98% on previous day. The 20EMA had dropped below the 50 in June.

Gold, 15/04/2013 - closed -8.73% on previous day, The 20EMA had dropped below the 50 in February.

Maybe this will convince a few people to go over to trend-following, but it might at least convince a few to get out of the water before a Black Swan.
 
Jack1960 said:
You will lose your shirt by following this logic.
More...


This is certainly not a strategy. But its a tactic, call it a safety valve if you like. The point is that the majority of what we might call Black Swans don't come out of the blue and destroy a bull market. More often they come in the midst of a clear and prolonged downtrend.

Same applies in reverse to extraordinary unexpected price rises - they're usually not counter-trend incidents.

I can't say anyone will be more profitable if they always apply this set-up, but by getting out of counter-trend positions they should avoid most nasty Black Swans.
 
Martinghoul said:
There's a fair bit of data-mining happening here, methinks...
More...


Absolutely, I want to demonstrate a point about Black Swans, not about data-mining.

Look at the top 20 daily falls recorded on the Dow. The majority occurred when price and the 20EMA were below the 50.
 
tomorton said:
Absolutely, I want to demonstrate a point about Black Swans, not about data-mining.

Look at the top 20 daily falls recorded on the Dow. The majority occurred when price and the 20EMA were below the 50.
More...
But lots of other times price and 20 were below 50, but they are not Black Swans?
 
tomorton said:
I'm going to say most Black Swans in trading could have been profitably anticipated (I'm not saying predicted) by being long when price and the 20EMA were above the 50EMA, or by being short if price and the 20EMA were below the 50EMA.
More...

Just a quick glance, from Feb 2016 to today you would've shorted the SPX 4 times and lost between 1% and 3% each time
 
There have to be other factors that would influence whether you go short after seeing a 20/50EMA dead cross. It cannot be an automatic reaction.

But the real point is, if you had taken that pattern as a signal to get out of a long position, you would have avoided most of the most notorious black swan price crashes.
 
Wikipedia carries a handy list of the worst stock market crashes and bear markets in Europe and United States (https://en.wikipedia.org/wiki/List_of_stock_market_crashes_and_bear_markets).

Looking at the Dow, would getting out of Dow longs have been indicated prior to these events by a 20/50EMA dead cross or a price close below the 50EMA?


· Start of 2015-16 stock market selloff, 18 August 2015 – Yes, 20/50 dead cross 25 June


· 2010 Flash Crash, 6 May 2010 – No


· Start of Global Financial Crisis of 2007-08, 16 September 2008 – Yes, dead cross 04 June


· Start of US bear market of 2007-09, 11 October 2007 – Partial success claimed. This date was actually the high so pre-dates the bear price action. The first serious price struck on 19 October, when the close printed below the 50EMA. Taking this as an “exit Dow longs” signal would have been valid, but anyway a 20/50 dead cross printed on 08 November, after which the index fell a further 50% by March 2009.


· Start of global stock market downturn of 2002, March 2002 – Yes. Price closed below the 50 on 03 April 2002 and a 20/50 dead cross followed on the 25th. The index fell a further 28% from the initial exit signal.


· Economic effects arising from the September 11 attacks, 11 September 2001 – Yes. Dead cross had already printed on 27 June


· Collapse of dot-com bubble, 10 March 2000 – Yes. The Dow had reached an all-time high on 14 January, and price closed below the 50 on the 21st. A dead cross followed on the 31st, after which the index fell a further 32% to October 2002.


So, claiming 5 Yes’s and a partial Yes from the most recent 7 events. Of course, not all these could be regarded as Black Swans. The burst of the dot-com bubble was hardly unforeseen at the time.


So, what about single-day price crashes on the Dow, say over the last 10 years? I’ll have a look at these later.
 
You must log in or register to reply here.
Back
Top