Poke holes in this investment strategy.

[theGreek]

Hey everyone, I'd like to share with you on some thoughts about a crazy idea I have.  I want as many opinions as possible so I don't waste any more time thinking about it :/.

Situation (highly theoretical)
-----------------------------

Prediction of highly liquid, fairly high price, very high volume NYSE stocks.  Example set: IBM, Costco, Exxon Mobile.  These stocks follow the main market very closely.  All these stocks are also 30% marginable.
These stocks average maybe 0.5%-1% a day in either direction.  Nothing huge.

Magically we have a box that can predict the next day's High of Day AND Low of Day to within about half a percent.  The magical box only works with the above type of stocks.

****Intra-day trading ONLY.****  NEVER holding anything over night.

Question: how best to make money?

Thoughts:
1) If the LOD occurs first, buy at LOD, sell at HOD.  If it doesn't hit HOD, sell at close.
2) If you buy at LOD, set some stop limit say 2% below LOD (for the magic box is not always very magical).
3) If the HOD occurs first, short the stock at HOD, and buy back at LOD.  If you don't hit LOD, buy back at close.
4) If you buy at HOD, set a buy back on stop limit at 2% above HOD (magic box not always correct).

5) Use margin all the time to maximise profits.  Stock movement of 0.3%, gain 1%.  Same for the inverse.


This is all highly dependent on how good the magical box is obviously.  Let's just assume its correct for now - I am more interested to hear your opinion on how this will spectacularly fail or spectacularly be amazing.  Please poke as many holes in this strategy as possible *assuming* the magical box is fine within 0.5%.

Ask/fire away!  Thanks.

[thrilldealer]

I don't get what you're trying to get out of this 

Anyways, that just looks like a normal common sense day trading strategy to me, if you did have such a box i'd say go for it, but of course there's no such thing.