Trading with Market Statistics VI. Scaling In and Risk Tolerance

[Dogpile 10]

Walter, the chart you just posted before was from July 25th. Look at the similarity to todays chart:

 

Bands narrow, chops around a bit, breaks trendline lower, forms bear-flag technical pattern, then pushes down a few more times... deja vu.

 

I labeled a few other entries I took today. Today I was watching S&P futures for that initial break but had my finger trigger on Dow futures as it came down through that trend-line. I then added on the S&P bear-flag, covered, and re-shorted the test back up to VWAP. Just an incredible day in an incredible market. I am not sure how long we can continue to have these types of days but I was trading in 1997-98 and that stayed volatile for quite a while.

[walterw 10]

NIce Charts Dog ¡¡ certainly this combination of price pattern plus statistical analisis is quiet a good way to trade, are you making some correlation analisis between futures on the breaks of this patterns ? validating or not ? cheers Walter.

[Dogpile 10]

walter, no I wasn't commenting on correlations -- but I will

 

seems to me like S&P futures is the contract to watch most closely as it seems to really interact with VWAP nicely. I have begun trading S&Ps more -- I used to trade YM and NQ almost exclusively but find myself trading YM and ES more right now. I was watching NQ but I didn't actually trade it even once today. I was trying to short a re-test of the morning high but it kind of spiked down hard and never really got my short signal so went back to the Dow and S&P charts.

[BlowFish 10]

Dbntina,

 

Any place that had the sense to ban me would be the sort of community I'd like to join! Actually no such bad boy behaviour. I used to use TS2000i (TS8 has only recently became available in the UK due to brokerage issues). I switched to Multicharts a while back but am not completely happy with that. It does compile most easy language natively though. I think you need a current TS8 account now to access the forums.

 

Would you mind sending me the PvP code? I would happily make suggestions and am a reasonably proficient code monkey. (albeit with little time at the moment).

 

You also might like to look at the ghkramer code for ideas http://www.enthios.com/FrontPage.htm There are a couple of tricks he uses that are quite neat but he re-builds the arrays on every bar! Not needed for our purposes. I have a fair understanding of how it works but there is too much going on to write you notes. If you do take a look and want to talk about anything PM me and we can talk on Skype or something.

 

Cheers.

[Blu-Ray 10]

No problem Blu-Ray...will send it tomorrow morning...would like to get some people to test it out...and I can make changes if we need to and get it documented and once we get the bugs out I will post for everyone here...next version I will get the histogram put on and then we are in business...then the following version will have an option to combine multiple days of data as an option...then I am through with this....

 

dbntina

 

Thanks dbntina, I'm looking forward to it.

 

 

Blu-Ray

[BlowFish 10]

This chart beautifully illustrates the behaviour that I was talking about (and trying to ensure I understood) a few posts back. This mornings DAX again.

 

From the session open it just dropped with no base to build any sort of volume for the PvP. Volume increased on the fall and so PvP led the Vwap the whole way down. Now according to Newbies (ok my) understanding of things so far, the skew is up technically so Newbie and even Trader would stand aside. But, having watched this behaviour for a couple of days in a strong down push they are wondering if Advanced Trader might be taking some other action?

 

At least they remember 'when in doubt stay out' and don't trade against the Skew.There patience is rewarded with a text book trade (shown on the chart). However they resolve to find out a little more about this phenomenon in a strongly trending market.

 

Cheers!!:beer:

[Blu-Ray 10]

Nice analysis there Blowfish, it looks like you've got that PVP pretty much nailed.

 

Blu-Ray

[Dogpile 10]

Blowfish, I don't know the mechanics of the european markets because they are strongly affected by the US market but here in US, a large opening gap changes the dynamics of the morning session.

 

Gaps here in the S&P futures tend to fill at least partially. Given this tendency, the initial play is generally to fade the gap for a trade and then see how market responds and go back with a trade in the direction of the gap once the market has pushed into the gap. You can see this dynamic play out on the DAX at the open. Think about it, the market wants to screw those who panic with 'bad marks' to open it too low allowing nimble players to buy into the panic and take a ride back up before the institutional players enter the market (institutions generally trade in afternoon -- which often creates 'afternoon trendiness' versus morning '2-way action').

 

General rule is that the morning session will often see a flush to force people who haven't yet panicked to panic. This will often set a high or low for the day in first few hours of trading- thus the PVP/VWAP will not be so relevant for the opening 30 mins+... then gets more relevant as the distribution fills out.

 

my point is just that the morning and afternoon sessions tend to be fundamentally different.

[BlowFish 10]

Hi Dog,

 

There's the thing - there was no gap to speak of yesterday! (chart a page or so back). Good observation for today however. The DAX is open now for the full European and US (main) sessions. Personally I think gaps occur in index futures for a couple of reasons. For example if there is key global economic news and if that happens when a market is shut (or in its overnight electronic session) when it opens it is gonna open more in line with its overseas counterparts. There is key news before the US opens (even US news) and that moves markets (or gives an excuse for them to be moved around!). And as you say it is the best time to jerk around the un informed investor especially if they have market on open orders to be exploited!

 

To be honest I am more interested in the characteristics of the PvP in certain situations (like the one I mentioned) rather than the fundamental 'macro' stuff that caused them (those conditions). For example if you get a strong directional day (whatever the cause) you are not going to have much basing action to build a solid PvP. The basing could have occurred one or more days before. Incidentally as the PvP starts anew each day it is gap neutral.

 

Cheers.

[jperl 11]

From the session open it just dropped with no base to build any sort of volume for the PvP. Volume increased on the fall and so PvP led the Vwap the whole way down. Now according to Newbies (ok my) understanding of things so far, the skew is up technically so Newbie and even Trader would stand aside. But, having watched this behaviour for a couple of days in a strong down push they are wondering if Advanced Trader might be taking some other action?

 

Ok, now I understand what you are referring too. When the market is near the PVP, a trade against the skew is possible by an advanced trader. It's called a break-out trade which we discuss in the next thread coming up.

[jperl 11]

Yep. That's what I was looking for - thxs Jerry.

 

Note the pullback at that point.

 

Here is another one that touched the 3rd SD Cooter. NQ touched, but continued on down.

[Guest cooter]

Did it not revert back to the mean?

[O66 10]

wasnt easy today

in the afternoon we had vwap below pvp but price action above. Shorts didnt work.

[jperl 11]

Did it not revert back to the mean?

 

Eventually it did, but would you hang on that long.

[jperl 11]

wasnt easy today

in the afternoon we had vwap below pvp but price action above. Shorts didnt work.

 

Yes, today was a day for breakout trades against the skew. We haven't discussed that yet. Coming up next.

[dbntina 10]

Jerry,

 

Looking forward to that as I think that will be very enlightening.

 

Today I saw the first short with the skew on the NQ would have worked.

 

As it came back up a short at the 1SD and then the VWAP would have been stopped.

 

Then the skew still existed and if you took 2 new shorts...probably would not have worked out for you...then the rest of the day PVP and VWAP pretty close.

 

Very interested in knowing how to tell when to abandon trades with the skew and favor trades against skew (breakout)...chart included.

 

Looking forward to the next thread...Jerry,

 

Great information,

 

dbntina

[Dogpile 10]

let me add something here dbntina. Linda Rashke rule is that 'Retracement trades' (entering on a pullback -- such as shorting at the -1SD level) after the market makes a very large opening gap is very dangerous. Believe me, I learned this the hard way a while back.

 

when you have a really large gap (like todays) -- this implies a large increase in volatility -- as investors will be very emotional. The dynamics of the morning session are different when you have a big gap and I would always be very careful after a big gap.

 

my point is that gaps call for a different mindset. I trade nasdaq a lot and can tell you that trading the morning session on NQ can be very tricky as it trades quite differently than ES.. add a monster gap in, and you should expect some serious spikes. trading a retracement with a wide stop is just not a good strategy for that type of environment. The market was trading FAR below recent value and was set-up to make a low and trade up after such a large move. The market was trading at a huge discount to the previous days price already.

[jperl 11]

 

Today I saw the first short with the skew on the NQ would have worked.

 

As it came back up a short at the 1SD and then the VWAP would have been stopped.

 

Then the skew still existed and if you took 2 new shorts...probably would not have worked out for you...then the rest of the day PVP and VWAP pretty close.

 

Very interested in knowing how to tell when to abandon trades with the skew and favor trades against skew (breakout)...chart included.

 

dbntina

 

I was going to wait until a later thread to discuss this, but since you brought it up here, dbntina, I will talk about it now. The question is, when price action retraces to the 1st SD or the VWAP which is a signal to pull the trigger, how would you know not to do it?

I use a simple technique due to Larry Pesavento which he calls the "Shapiro Effect". The Shapiro Effect states "If a trade is good now, it will be good in 5 minutes".

Don't take this statement literally, but use it interpretaively. I use it in the following way. Let say you are looking for a short entry on a retrace to the 1st SD or VWAP. The first UP BAR that touches the SD is your signal. Don't pull the trigger yet. Apply the Shapiro Effect. Wait for a down bar that confirms the signal. A confirmed signal would be one that drops below the low of the signal bar. Then pull the trigger.

Look at the attachment. This is the same 2 minute chart of NQ for Aug 10 that you posted dbntina.

At point "a" we get a signal to go short at the 1st SD. Apply the Shapiro

Effect. There is no confirmation of the low on the next bar. Don't pull the trigger. It took 7 bars later to confirm that low, which was below the 2nd SD. I usually don't trade there.

At point "b" (the VWAP) we get a signal to go short. Apply the Shapiro Effect. 2 bars later there is a confirmation of the low of the "b" bar. Pull the trigger short.

At point "c", we get a short signal at the 1st SD. Apply the Shapiro Effect. There is no confirmation of the low. Don't pull the trigger.

At point "d" we get a signal to go short. But there is no confirmation of the low. Don't pull the trigger.

 

You can see how this works. Applying the Shapiro Effect will keep you out of most bad trades. The downside of course is you might miss some good trades.

[dbntina 10]

Thanks JPERL makes alot of sense...Shapiro effect combined with the knowledge of when to look for breakouts and avoid what happened today should be very helpful...thanks for the reply!

 

Dogpile good info as well...you make a good point gaps fill more than they don't and large gaps and large volatility is another cause for caution...good points.

 

These are wonderful threads Jerry and I am learning alot of good guidelines...thanks for sharing your market wisdom and everyone's comments.

 

dbntina

[walterw 10]

No problem Blu-Ray...will send it tomorrow morning...would like to get some people to test it out...and I can make changes if we need to and get it documented and once we get the bugs out I will post for everyone here...next version I will get the histogram put on and then we are in business...then the following version will have an option to combine multiple days of data as an option...then I am through with this....

 

dbntina

 

Hi dbntina, thanks for great contributions here, as you may notice TL is a very diferent forum for the level of collaboration and quality of content... I think we are all priviledged to have Jerry with us... so far all this threads had been evolving wonderfully..

 

My question regarding to your indicator now including dynamic PVP, when would this be available, would love to try it too ? thanks in advance, Walter.

[dbntina 10]

Walter,

 

No problem...the code plotting the actual tick by tick pvp is finished and plots on a minute chart with the VWAP and SD bands like the chart I posted earlier for example.

 

Blowfish and a few others are testing it out. Blowfish even made some changes to the code that will allow it to run even faster. He made it so it only checks the current tick for a change rather than running through all the arrays every tick...good thinking on his part. Should be very efficient now.

 

I will forward to you Monday morning unless I head into work this weekend. The code is for TS however, I think Blowfish altered the code to work on Multicharts also.

 

Cheers,

 

dbntina

[Guest cooter]

Blowfish and a few others are testing it out. Blowfish even made some changes to the code that will allow it to run even faster. He made it so it only checks the current tick for a change rather than running through all the arrays every tick...good thinking on his part. Should be very efficient now.

 

Efficient, yes. But truly accurate?

 

Since every tick should be indicative of a change in aggregate volume, wouldn't the result differ accordingly?

 

Or put another way, are you going to have a "zero volume" tick?

[dbntina 10]

Cooter I am not following what you are saying...

 

Here is what the indicator does. It takes the volume at every price and keeps a cumulative running total of volume. It plots the price that has the most volume, the pvp.

 

As far as being accurate...I don't think it can get any more accurate then tick by tick...unless the tick volume is not correct. It will be as accurate as the accuracy of the tick volume I guess.

 

I don't understand what you mean by zero volume tick.

 

dbntina

[Guest cooter]

Cooter I am not following what you are saying...

 

Here is what the indicator does. It takes the volume at every price and keeps a cumulative running total of volume. It plots the price that has the most volume, the pvp.

 

As far as being accurate...I don't think it can get any more accurate then tick by tick...unless the tick volume is not correct. It will be as accurate as the accuracy of the tick volume I guess.

 

I don't understand what you mean by zero volume tick.

 

dbntina

 

If Blowfish modified the code to only check the current tick for a change, isn't something missing if it's not running thru the array on every tick?

 

What change is the new code referencing? Price, volume or both?

[nickm001 10]

I think cooter is correct... if there is no change in price ( but trade still happened) is that volume included in the calculation or not? It should be.

 

Some posts in TWS forums ( think from solidus) showed that 60% of the volume happens with no tick( price) change. This is where the CPU load saving comes from.