Opening Momentum

Opening Momentum

Here's something that's intrigued me for some time:

At stock opens UP by, say 2%. (That's like a 200 point gain in the DOW.)
Do investors think: "WOW! I better sell to lock in my gains" (That'll drive the price down, eh?)
Or do they think: "This stock is on a roll -- I better buy some!" (That'll drive the price up, eh?)

>So, what happens after the Open?
That's the question I asked myself. I've asked it before, here.
I'm convinced that there's something, call it Opening Momentum, that'll drive the stock higher during the day.

In particular, I wondered about my coal stocks, like Grande Cache
What's intriguing is trying to (somehow) display this "momentum".
I started by plotting the Open-to-High gain versus the (positive) gains from Close to next day's Open.

The result is displayed in Figure 1. The scatter plot is all over the map. It tells me nothing.
>What about other stocks?
Yeah ... so I look at a few more stocks, from the DOW:

Figure 1

>Well, your coal stock does look more ... uh ... scattered, right?
Yes, it's different -- but it doesn't display what I expect: namely, that the High for the day is significantly higher than the Open.
>On those days when the Open is higher than yesterday's Close.
Yes, that's what I'm considering: only those days when the stock opens UP.
Then I decide to plot the Close to next-day-High vs the Close to next-day-Open.
>On those days when the stock opens UP?
No, I'll look at all Opening changes ... to see what happens to the High when the stock Opens UP or DOWN.
This is what I get, for those DOW stocks:

Opening Momentum

>What's the red line?
The "best straight line fit" to the points: the regression line.
The grey line is where High = Open. Clearly, no point will lie below that grey line.
>Huh?
The High for the day is certainly not less than the Opening price ... else the High ain't the High, eh?

>So does your coal stock looks any different?
Stare at Figure 2 and tell me if you see any difference.

>I'm looking ... seems the same to me.
Look at the regression line. It sits higher, above the grey line.
It can't be any lower -- but it could be significantly higher, eh?

>Only if the High is significanlty higher ... uh, is that what you're looking for?
Yes, I reckon that must be it: the Intercept of the Regression Line.