|
PureBytes Links
Trading Reference Links
|
For Tom Nagel et al
This is a quickie response to an excellent post from Tom. I say quickie,
because shortly my machine with the modem is having (what the Royal Navy
calls a "dockyard job", in that it requires more memory. (I am not to sure
that the owner couldn't do with a bit more too, but that is another
matter...!)
First of all, let me say that I didn't really start this little strand
'tongue in cheek' but I do admit that I knew the title would set a few guys
going in the defence of the varied mass of indicators that exist within SC
(as far as I'm concerned) and TS for all those who can back-test and do all
the other wonderous things I can't.
By the way, when I say indicators, I mean these sort of things:
stochastics, RSI, MAs, MACD and all the other wonderful and wiggly lines
designed to help you know where the market is going - only, as I say, as far
as I am concerned they all lag behind the times (rather like me when I'm
shooting clays - when I miss it is usually because I am behind, very rarely
ever in front!)
None of the RT replies or the mass in private mail that to me came up with a
single 'non-lagging' indicator. I simply don't believe its feasible. And
I don't count things like a retracement tool which shows a .382 or .618
retracement (or extension). I don't mean volume (which on the Bonds, I
don't get, but I do track the tick volume to get a feel for the noise).
Volume is a tool within the market and very useful if you have the real
thing, real time.
Now, please don't get me wrong about the value of indicators. As I have
said, in the time frames I trade, I feel that they are uselss. I am sure
they help others to make decisions they otherwise might not make without
them - let's put it that way!
Me, I don't need stochastics to tell me the market is overbought or
oversold - it is staring me in the face! - what I need (if the wretched
thing exists) is a wonderous tool to tell me when it is going to stop being
overbought or oversold and, well, get on with it... For me, the only
thing to tell me is the price action. And, despite far flung searches for
something else, that's it. When it throws up a particular pattern at a
particular point in time (sup/res), that is the nearest I am going to get to
being told that the market is heading off in direction in which it is
pointing... and, with much wriggling around, it will reach another sup/res
point and try and tell me all over again.
Let me give the 'today' example - for those that can put up a USM8 five
minute chart>
After the movement following the GDP Report, the market headed south through
R1and then did a reversal at the Pivot. The retracement tool, measured from
the High to the base of the reversal bar at 7.55 CST was immediately put up.
The market cut the .382 but couldn't go further to show that it was very
bearish. The third bar started the continuation of the move south (not
before 4 ticks could be extracted). It then put in a huge down-thrust at
8.30, closing on the Pivot. Out comes the retracement tool again. The
market went to .382 again and held its postion for 20 minutes, which was a
pretty long time to make you mind up that it was a southern type of day!!
The market makers then took their one tick for stops and headed south -
racing through the Pivot and going with indecent speed through S1 and S2
before drawing breath just one tick from Yesterday's Low. It then did a
"lets go sideways, while the senior traders take break" and made a pretence
of a double bottom, which would have fool no one. The third attempt and
through price action, was on the border of being a false move - but the big
boys returned at midday to make the move count for something.
While I finish this, the market isdrifting sideways and down - after having
made a really decent range day. But even from the last retracement at the
bottom, there were another 10 ticks to pick up. In the next half hour the
odds are in favour of a small retracement while the daytraders take their
profit and give back a "thank-you" to the pit for so doing.
Now what indicator would have given you the info that you needed to trade
that? For the hell of it, I have just put up the RSI and this is what it
did: it was mid channel for both of the early retracements and it cut the
30 line half way through the fast move down and stayed below it until the
retracement at the bottom. It then went on wiggling around betwee 50 and
30, which is what it is still doing now (13.20CST). Incidently the tick
volume was of particular help during the second of the early retracements,
in that the relative volume dropped completely. Now that did help to
confirm that traders were not interested in the upside - but wouldn't real
volume have been so much better...
Anyway, to wind this up, as far as I'm concerned. I flew a kite in hope
that someone might come up with a marvellous indicator, none of us knew
about. Alas. But it was worth getting some good brains coming up with
what they did. But it would seem that my manual will not have many
indicators in it....
Happy trading
Bill
>Bill Eykyn sent thru a great post about his views on indicators and their
>perceived value compared to the actual price.
>I feel that this post may have been written slightly "tongue-in-cheek"
after
>seeing quite a few postings from RT's who believe wholeheartedly in using
>indicators as part of their mechanical trading systems...I am one of
>those.....
>
>Bill wanted to provoke thought so these are mine...
>IMHO he is completely correct when he says "Except for price action, don't
>all indicators lag ?" but just because indicators lag does not mean that
>they are not valuable tools in the technical arsenal. I would suggest that
>many profitable moves can be "predicted" by synergising certain lagging
>indicators with price-pattern analysis and that by using price action
>exclusively these moves will not be caught (with the same degree of
>reliability).
>In my experience these lagging indicators (when used with price-pattern
>analysis) will give highly reliable signals over all time-frames...and I
>think that one-minute data can be used just as you can monthly data...at
>least in the FX markets.
>Maybe I am wrong in this observation but I feel that every large &
>significant move is accompanied by an upward shift in momentum (relative
to
>the zero-line) so I am wondering how does price activity alone gives
>information on the momentum in the market.
>I personally use an empirical sort of momentum measure by counting time
>versus price in the last up/downmove and then comparing to the the current
>down/upmove of price v. time (but this is really a simplified momentum
>INDICATOR which is easily derived from observation of price).
>
>In Bill's approach, it is price analysis at the beginning (the set-up) and
>price analysis at the end (the entry or exit) but at least for me I prefer
>lagging indicators as the set-up and price-pattern analysis at the end (the
>entry or exit) thereby protecting one from foolish forays into the market
>based on the lack of fine-tuning inherent with lagging indicators....maybe
>I've shot myself in the foot with that last comment !
>
>In these cash FX markets which I watch there is no real-time Open Interest
>or Volume to use as confirmation, but you can't beat price pattern
>recognition for stop-loss placement in particularl......maybe that brings
up
>another eternal question...the question being should we use the same
>criteria for entry as for exit or is it preferable that they be different.
> I think I'll leave that as more food for thought.
>
>Conceptually, it's all interesting stuff... I hope you agree.
>Regards,
>Tom Nagle
>Tullett & Tokyo Forex International,
>Cable House,
>54-62 New Broad Street,
>London EC2M 1JJ.
>Tel: +44 171 827 3409
>
>www.tullett.com
>
|