Are gaps (aka gasps) important?

An ID 200 tics/bar chart is showing a refresh phenomenon indicating too few tics/bar for reliable data.
More...

I have not seen this phenomenon (a chart changing after a refresh) since moving to DTNIQFeed/QT months ago BUT it is clearly evident in the pair of charts below [PM II follows PM].

Until the exact reason for this is determined, one should use ID (intraday) gap data for YM with appropriate caution. The most obvious reason is, as I said, a refresh artifact but if so, it would be peculiar to YM. Highly liquid instruments like SPY don't show this which makes me doubt the refresh explanation (this arises from data lag). What one is left with then, is late trades, over which the trader has no control. There may be other explanations but I am not aware of them.

On the other hand the interday data is as reproducible as one can get, IMO, and thus is very believable (aside from whether or not one perceives gaps to be important). So we are left today with a close below the lower gap margin [actually an outside down day on increased volume], having failed rather miserably on the upside. Note that the gap construction makes NO PREDICTION about which way the market is going but rather points out certain obstacles which could be in the way of it going where it wants to go.

lj
 

Attachments

Quote from ljyoung:

I don't understand what you are saying here (?do you mean "can't move"?). If that is what you meant to say, then are you saying further that your definition of a gap is something (a price void perhaps) which occurs when there is a market closure (what is a market closure? How often does that happen?) or market suspension (when does that happen?) or that volume is too light (too light compared to what?).

TIA

lj
More...

sorry, can't move. The rest remains
 
The rest remains
More...

So if what you saying is that a gap is a failure of price to move because of a market closure, market suspension or "too light volume", then I would again ask what do you mean by "a market closure, a market suspension and too light volume"? Are these the only circumstances under which you have observed a gap? Are these the only circumstances under which you would expect a gap to be present?

It seems important for the purposes of discussion to establish what each of us is talking about and I still don't know what you are talking about. Let me give you my definition of a gap:

A gap is a price void.

lj
 
Price tumbles in the after RTH market. The broker-dealers are eating their own. Where will price stop? At the new gap? Well, actually it's not new, just previously unnoticed. It was so small, it hardly deserved mention. It actually occurred before the "big" one on 5/1-2. Let's see what happens when the Brits get moving in a couple of hours.

lj

PS: F/U on the ID YM wipe. I checked GOOG (which is what I trade) for the dreaded refresh artifact. It is pristine. No wipe. I will recheck SPY and DIA tomorrow. BTW, DIA shows a strong sell signal (volume indicator) with a [double-inside + an outside] to give a classic "pop-gun" configuration which is indicative (but not predictive of) a large move in price.
 

Attachments

The pesky thingy, an overnight gap was filled rather briskly this AM and so the question naturally arises, "So what?" There are more to gaps than just the absence of price. There is the relation of a gap to previous price activity, previous gaps, the timing of the gap, the manner in which it is filled, the response of price to the filling of the gap and on and on. But one thing is for certain - the market maker or specialist did not simply forget that there were 2 pips or 500 pips between two succesive tics. There is a reason for "hopping" over a pack o' prices and most times one doesn't know what it is. So what.

My futures real time feed expired yesterday and so I'll be switching over to DIA. It is actually more interesting , IMO, because there are soooo many gaps. The premarket is a time where this gapophilicity is especially notable when compared to the other two index ETF's of interest to me: QQQQ and SPY. I wonder why that is?

lj
 
The premarket is for price discovery (PD). It serves other functions as well, but PD is a big one and DIA is usually all over the place. Confusion? I don't think so.

Rule: Three closes is safest. Three popped up for Mandelbrot too. And 1-2-3 too.

lj

That 132.98 spike has gotta be a "bad tick" right?
08/02/07 11:00:57 132.98 133.57 133.58 1300
Maybe not.
 
DIA has been stuck in a nasty little inside since the day o' the dump and needs to do something about it. There is still, IMO, a very strong volume-related sell signal but I do not know when it will be acted upon. BTW that 135 resistance is perhaps not unrelated to the bottom of the lower "gap post" to the left in the daily chart.

lj
 
IMO, gaps are important. IMO, gaps cannot be viewed in isolation. IMO, Friday's selloff stinks to high heaven. What's the BS cheese really up to and how can I make money on what I think? A touch and go on Monday would be telling as would a decisive close below the points on high volume. George Douglas Taylor said, "I make 40% of my money when the market doesn't do what I thought it was going to do". Smart man. Good advice.

lj

BB can't cut but maybe the statement will be mushy.
 
You must log in or register to reply here.
Back
Top