P & F & UK Shares

orchard

Active member
Messages
163
Likes
2
I have been following the thread on P & F on real time indices with interest. I am really interested in using P & F with UK shares using EOD data (I'm learning the lingo :) .
I have studied P & F to some degree and intend to paper trade using various combinations. At the moment these consist of 3% x 2 and 3 reversals and 5% x 2 & 3 reversals.
I have literally just started, but would any kind soul out there give me any pointers?. Otherwise it could take me months or years to get any meaningful data to work on.
I have information about simple and complex signals etc. which I will follow, but my main query is about the % value to use and the reversals.
I am able to put any combination on the comp (using sharescope) but would also be interested in if there is another indicator which works in conjunction with P & F to give a better overall picture.
Sorry about picking your brains again ;) but you shouldn't have been so helpful in the past :D
 
At the risk of plugging products here have you looked at Pfscan? I have no link other than being a satisfied user honest :)


It uses the SS datafeed and produces automatic P&F alerts for any group of shares (eg FTSE100/ 250) it also then allows you to use various filters to cut done those signals.


You can try it free, download from here:

www.pfscan.com


Cheers
 
Orchard - be careful if you are using ShareScopes P&F indicator, I have spoken to them in the past, I understand that they only use the open & close price of the day and this results in an inaccurate P&F indicator :(
I use Pfscan as Helen does...believe me its good and serves my purpose ;)

http://www.pfscan.com

ST
 
Thank you both. With 2 such recs how could I not buy??

Have done so, just awaiting confirmation. (Also learned about paypal!!)

Once again, many thanks.
 
Hi,
sorry for the delay in sending the code, I was in sunny London and incommunicado ... or is that the pear?... for 48 hrs.
As for buying on the strength of the recs - okay guys, take your picks from the 'Pfscan Loyalty Scheme Catalogue', is it to be the novelty coaster or the chiclets dispenser this time?
Dave :D
 
Hi all and especially Orchard who it seems is also weighing up the merits of P&F . . . . .

I have recently read Marc Rivalland's book Swing Trading and have to admit mixed opinions.

A review of his book is not the purpose of this email. What the book did leave me with was a desire to find out more about P&F. My concerns are, in order of magnitude:

i) Since there is no time axis on a P&F diagram the part of the chart that generates the signal could take 30 mins or a few weeks to print. Does this not mean that you miss signals when using EOD data?

ii) Stops. Where to put them and how to monitor them especially if a 3 box reversal could end up being a profitable reverse trade. Problem of stop placing is compounded by i) above.

iii) Which set-ups to trade and when to take profits?

Has anyone any experience of P&F on UK FTSE 100 shares rather than FTSE index which I would never have time to trade.

As an aside - I might be missing the point but isn't the decision whether to use SB firm or actual futures governed by your account and trading size. If you always pay 6p spread to the SB firm on a FTSE cash punt and your stake size is just £0.50 then this has to be worthwhile. If the commission charge from future broker was £18 per round trip then you would have to be regularly betting £3/point to make it worthwhile. Anything above £3/point with SB firm and you are better off paying futures broker's commission?

This ignores SB firm bias. Am I way off beam?

V grateful for any help on P&F charts.

BTW Orchard - what do you make of pfscan?
 
Hi Fastnet.

Still experimenting with P & F.

Re Pfscan - great - easy to load, easy to use, lots of information and fast.

I use sharescope so the connection between the two is simple. Also I only use FTSE 100, otherwise you get drowned in info and shares.

Once this crazyness re the war is settled down I might even unfreeze my finger and start actual trading :cheesy:

Why not try Pscan free for 30 days. DaveJB is more than happy to answer queries before you make the leap into buying. (sorry DJB :!: )

Best of luck
 
Thanks for the advice Orchard.

This might seem a bit selfish but I was wondering more about the actual reliability and success rate with P&F charts generally including the problem with lack of time axis.

Before I sign up to yet another subscription site (despite free 30 days) I just wondered if anyone was using P&F at the exclusion of all else to trade profitably overall.

P&F as a concept is very intuitively attractive since it removes distracting background noise and charts clear formations. However so many systems that have at first appeared very appealing (trend channels/ buying on retracements etc) are actually very difficult in practice especially on EOD date.


I have lost cash in the past and vowed that if I ever got involved in anything outside an ISA again I would be very disciplined with backtesting before I began.

So - two main parameters that need to be satisfied - i) Is P&F profitable overall (ignoring drawdowns) ii) how easy to use in practice?

Cheers
 
...sorry...I clicked the button too early and posted the incomplete message...

I wanted to add that there are many packages that are now avilable, some giving you facility to download from web and a use of 4 weeks for you to decide...

I would suggest that you look at all and then decide...to see which one would suit your trading style...

the other ones are
http://www.xlchartpro.iinet.net.au/index.html
http://www.ez-pnf.com/

regards
 
Orchard,
no bother at all - I prefer people to trial, if they pay first and dislike it I'm left with a tendency to refund... which suits 'pirates' who then have a valid access code for free. (Trial, then buy, then dislike it I consider as a shot at your own foot ;) )
I'd agree also with Zambuck, selecting a trading system is very much a personal thing - what suits one doesn't suit others (I like Metastock, for example, yet I never seem to meet anyone else who does!) I'd also say - as I usually do at this point - that trading is about entry signals, and exits, and managing the position - whilst Pfscan's backtesting is improving it is still more an 'entry generator'. I consider this a small part of a successful trading system, which is why I'm trying to round it off with the other parts of the whole shebang... you need to consider the whole system, and pick the parts to fit it, which trying a few programs out will help with.
Dave
 
Hi Orchard and DJB -

I wonder if you could offer a little advice on actually using P&F for FTSE 100 shares in practice?

I can only use EOD data. I find P&F charts very intuitively attractive but am concerned that the lack of a time scale means that, in practice, you might miss some of these great set ups.

For example when you see a triple top with ascending bottoms forming would I have time, in general, to act on this signal or would price normally break through resistance (the point where I would go in based on Marc Rivalland's book) and go straight through so that by the time I updated the charts in the evening the opp for profit has diminished?

I realise that this is a slightly naive question but please bear with me . . . . .

I shall be using SB and only trading one or two signals that I feel are most reliable. I shall also filter the signals to trade only with the trend etc etc.

Also, DJB - how would I go about trialling pfscan?

I have realised at least that updating P&F by hand is very time consuming and leads to many errors if you're not in the habit!

Cheers for any advice
 
Okay,
trial info (easier to refer back if it's at the start) www.pfscan.com/downloads - download it and it runs 30 days in trial mode. Email me [email protected] for more info etc. (Apologies for 'advertising' on here, I try to avoid it!)
Pfscan uses EOD actually, so that's covered - what you're really after is to take your P&F signals and confirm they are right often enough by a large enough % to allow you to produce a tactic for trading them... so you could take your triple tops and set rules that you think might improve the chance of a good run up - for example share is above 200MA (so less likely to be a short reaction to an overall drop), Bear resistance trendline overhead is at least 10 clear, and so on. Having filtered out TT's that have resistance close overhead and other likely candidates for a bull trap style setup you'd then have to have your stops and exit tactic ready to maximise the return.
I'm not saying this 'off the cuff tactic' is going to work, I'm knee deep in testing things like this right now and am nowhere near a conclusion, but Pfscan (advertising again ;-) ) is being moved in the 'let's check this out' direction... so I'm hoping to have some up to date info to base such ideas on soon. Rest assured that if I discover anything worthwhile I'll certainly... err, sell it for a lot of money ;-)
 
Thanks DJB -

You obviously have a lot of experience with P&F.

To further my understanding of these charts I thought it would be a useful exercise to sit down with the past 6 months daily OHLC data from GSK (picked as random example of FTSE 100) and construct the chart.

I am using the advised 1.5-2.0% box size of 20pts. 3 box reversal.

Now one problem that I quickly came across was that of a day where the daily range includes a downside move that exceeds 3 boxes but also an upside move that exceeds the day before. This clearly requires more than one additional column per day. However with just OHLC day it is impossible to tell in which order the up/down columns should be or whether there are just two or more than two columns required?? It quickly becomes quite confusing especially when plotting by hand.

Since P&F are independent of time my basic understanding is that extra columns should be added using high/low data whenever they occur. By just adding one column per day you could easily omit important resistance levels that the price might have touched and bounced off intra-day. This, in turn, could mean that a breach of a double-top would actually be a breach of a triple top etc . . . .

How do the PFSCAN charts get around this problem? - do they add one column per day or follow the exact price action?

Also - do you know where I might find free historic P&F diagrams that I could back test?

As always I am very grateful for any help/advise from you or anyone else that has already come across these problems for the first time.

Best regards
 
Hi fastnet,

PnF daily charts should look at the days action and see if more boxes should be added in the current direction. If they should they are added. A move in the other direction is not then considered.

However if there are no further boxes to be added in the current direction the opposite direction is then considered.

What you are suggesting is not daily PnF but intraday where clearly you need intraday data and not EOD.

JonnyT
 
Thanks for your reply JonnyT -

From your email I understand that if, for example, the price is currently describing an upward column but then, intraday drops once or twice and bounces off the same resistance that is more that 3 boxes away (assuming 3 box rev) BUT the price gave a high for the day slightly higher than the previous day then only the continued movement upwards would be added??

Would this chart not then fail to register the obvious strong resistance at the lower level?

I can see that this way of assembling the charts makes the process much easier but does it not interfere with the principle behind P&F which is to show areas of resistance and equilibrium on a chart independent of time. If a chart is completed with strictly one column per day it can not be independent of time.

As I say I am new to all this. Please understand that I am not being deliberately argumentative but instead just trying to reconcile in my own mind the effect this would have on the reliability of the signals? Have you ever compared the two methods?

I should expect that with large box sizes and relatively nonvolatile markets this problem would not often arise.

Also could you please confirm that, using your method described above, that you add a box in the current direction based on the HIGH point in that direction and not the closing price. Basically an extra box (or more) is added in the current direction even if the low is far enough away to warrant a reversal.

I would appreciate any comment you have on the apparent deviation form the underlying idea behind P&F and the extent of its affect on results/signal reliability.

Thanks again
 
If you are charting off end-of-day data, then you will only see support/resistance lines that are apparent with market close data.

If you want to see intraday features, you need to use intraday data.

And yes - one data item can only do one of:
a) Extend existing column
b) Start new column in opposing direction.

First see if it can do a. If not, see if it can do b. It cannot do both, nor can it start more than one column. Unless you are using some weird variant of PnF.

-svengali
 
Hi fastnet,

<i>From your email I understand that if, for example, the price is currently describing an upward column but then, intraday drops once or twice and bounces off the same resistance that is more that 3 boxes away (assuming 3 box rev) BUT the price gave a high for the day slightly higher than the previous day then only the continued movement upwards would be added??</i>

Correct, that is a rule of PnF charts.

<i>Would this chart not then fail to register the obvious strong resistance at the lower level?</i>

Correct again, the answer is to use intraday data and charts else change the granuality of the chart.

<i>I should expect that with large box sizes and relatively nonvolatile markets this problem would not often arise. </i>

Correct

<i>Also could you please confirm that, using your method described above, that you add a box in the current direction based on the HIGH point in that direction and not the closing price. Basically an extra box (or more) is added in the current direction even if the low is far enough away to warrant a reversal. </i>

Correct, this is a basic rule of PnF charting.

JonnyT
 
Thanks for your clear and concise replies.

I wonder if you could point me in the direction of historic P&F charts that I could use to back-test the reliability of the signals generated.

From what I have read it seems that the most reliable signals are failed double tops with at least 3/4 boxes in opp direction or failed triple tops.

Triple tops with ascending bottoms are also supposed to be reliable.

Any comments? - it's great to have found someone else that uses P&F - I'm not sure it's that popular

Thanks
 
Would anyone be willing to discuss success rates of P&F set-ups on FTSE100 shares using EOD data?

Main issues are the most reliable signals, stop points and filtering techniques - ie only going long on bullish signal when FTSE trend overall is up etc.

Unless someone can inform me otherwise it looks as though there isn't a free source of P&F charts for UK shares which is going to make backtesting signals pretty time consuming since I'll have to draw the charts as well!!!

Does anyone trade large cap shares usiing P&F? I can only see threads about there use with intraday FTSE/S&P trades.

As always I'd be grateful for any help -

Thanks
 
Top