Why don't HFTs just start ramping AGNC into earnings now?
Yeah, ill second that, this threading system is a mess.
Thanks once again for your input, I was hoping maybe some other people would chime in to liven up the debate but i guess not. By the way just fyi I never once junked you.
Also i wasnt trying to psychoanalyze you in any way, its just quaint to see there are people out there who still think fundamentals (especially on the macro scale) apply in the new normal.
I dont really have any real idea what your understanding of algos is, but even someone with basic understanding of how algos work should understand that the 3 flash crashes in the stock arent really robotic abberations, so much as algos competing with one another as they do everyday. The flash crashes are only small glimmers of the markets plumbing.
Lastly as I said, following when correlations change has been very useful to me in foreseeing algo movements (which are often based on rudimentary TA) but thats voodoo to you.
Imho one should follow the market with one eye on the fact that most of it is rigges by algos and one eye on the fundamentals of great companies (those first principles you mention) which is why we are on the AGNC board to begin with.
Anyway, there doesnt seems to be a real dearth of discussion so I doubt ill be back to yahoo, therefore farewell and the best of luck to you
I wasn't referring only to the flash crashes, is my point. There were a couple of days of dead-money plateaus that could only have been maintained by the bots.
Anything else the HFTs do that doesn't produce a stable geometric pattern (the day they made three Dow components cycle sinusoidally in sync, for instance) is indistinguishable from market activity. And everything they do is likely to cost them a metric ton of money before they harpoon a whale and make it all back. Their unrealistic contribution to valuation is distortive noise, and the value wins out eventually. The only people they really screw are those unlucky enough to time their transactions out of sync with the bot.
And there's so very little of TA that has any merit (hint: moving averages mean jack squat because the market doesn't do coordinate arbitrage, people do).
So you're left with what is the company worth and what will it be worth in the future and what is your personal time horizon for the trade and what is everyone else thinking about those things at any moment in time. Trying to second-guess a random collection of randomly-programmed computers that can't get out of their own way? Pointless. Unless they give you a blatant signal like a flash-crash to buy into.
PS- By the way, to be totally fair to the fundamental point we should note that its interesting to look at the mREIT universe as a whole. They all sort of crashed, but some deserved it much more than others for fundamental reasons. Annaly did a very poor job going into QE3 and had to cut the divy. while IVR's crash was just a short spike and went on to all-time highs as it was the first to recover (presumably the 'story' was that having the lowest CPR, it would be the best able to withstand competition from the Fed). Given AGNC's low CPR, its chart should look much more similar to IVR than NLY, but ended up falling somewhere in between. Also I dont really think ANYONE was expecting an SPO, that would have been suicidal. If anything, Kain should have realized some of that comprehensive income and bought back AGNC shares at 10 or even 20% of the float. Can you imagine how fast the shares would have rebounded had they done that? More's the pity because I doubt management will ever get another opportunity to sell assets at north of 100 cents on the dollar and buy its own shares for 80% of BV...