Once upon a
time, a very, very, very long time ago, a generation of traders (of which I was
one) was introduced to the concepts of crosses. This introduction was not
without some mishaps of course, but it was an invaluable outcome of our elders
and betters teaching us how to read markets, more specifically how to identify
drivers of moves in the main currency pairs. It is hard to escape the
conclusion that in the modern FX market, such skills are thin on the ground.

There have
been some pretty big moves in FX markets since the US election – not surprising
given the lack of preparation for a Trump victory – but it’s the drivers of
some of those moves that I find fascinating.

USD/JPY, which has had its own emotional rollercoaster over the past few days.
I looked at the pair gyrate as it did and then looked at the other currencies,
and saw AUD/JPY. This favourite of Japanese retail FX traders, fell about 6% in
the immediate aftermath of the result becoming clear (in about three hours in
fact), only to rebound 7% and more in the next 24 hours. It was clearly a big
driver of the headline pair.

I know Japanese
retail volumes can be large, but this flow, with all due disrespect to the
thousands of punters involved, is uninformed. They are reacting a crucial few
seconds after the professional institutional market, and their impact is more
often than not to extend a move, not create it.

This retail
army is, however, accurately reflecting a broader trend – the unwillingness to
pre-position due to uncertainty and the maintenance of too tight stop losses.
Moves such as described above only really happen because of a lack of
pre-positioning, if there were people in front of the trade they would take
profit at some stage and help slow the move.

The foreign
exchange market now exhibits all the characteristics of animals – it’s herd
mentality 101 – and while it is exhibited by Japanese retail clients in
particular, one can’t help escape the conclusion that the rest of the market is
now doing to same, which is inevitable I suppose when machines are making so
many decisions.

The machines
are data dependent and as such they also become reactive and extend moves. Of
course, much as is the case with technical analysis, if enough people follow
one path then the whole thing becomes a self-fulfilling prophecy.

It is
interesting to me that this same data dependency permeates other areas of the
world and was highlighted last week during the election results. The New York Times, which I have previously
stated to be the newspaper I consider to be the best in the world (and still
do), had a very sophisticated barometer operating as results came in. The
problem was this “sophisticated” barometer was also reactionary. It started off
with an 80% Clinton victory on exit polls (data), drifted around, then went to
60% and then very quickly as results (data) came in, it was at 20%.

At no time
did the “sophisticated” model actually make an accurate prediction until the
deed was done and the result known. What it told us was the same thing any
semi-sensible person (and I’ll refrain from an obvious joke here) could have
seen with their own eyes – the results were not coming in as expected and
Clinton was first, in trouble, and then second, losing.

data-dependency, especially in a world dominated by mean reversion logic, must
present great opportunities for human traders. I have to think it is only a
matter of time before an investment manager with serious AUM to play with,
establishes a mean reversion fund with human traders following FX markets
during events and stepping in with a series of bids or offers.

there is a real issue driving the market (which the humans will have spotted of
course), these new bids and offers, which won’t have to be that large, will
stop the market and trigger the mean reversion.

As noted
before, the machines don’t differentiate between retail and institutional
players, so stick a few bids or offers in on several public platforms and let
the machines do the work. There is a great opportunity to make serious money
because at the moment the market structure seems to be dominated by reactionary
machines and a bunch of sheep who will follow anything in the hope of making a
few pips.

It is often
said that the foreign exchange market, both in market behaviour terms and
market structure, is a pendulum. Well it seems to have swung too far toward the
machines and data dependency (which is also a result of the new compliance
regime – “the data made me do it!”). The opportunity is there for someone to
help redress the balance and probably make some serious money in the meantime.

And if the
balance isn’t redressed…AUDJPY, a 13% move over a 28 hour period – welcome to
your future.




Colin Lambert

Share This

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on reddit

Related Posts in