My CAPS score has declined by 4000 points over the last 6 months. My real life short positions have also performed poorly, although long positions have more than made up for it.
I thought I should do a post mortem of some picks that have performed poorly, and figure out what mistakes I made. There are a lot of them, 64 picks on which I've lost over 100 points, including 1/4 of my open picks.
Facebook (FB), Baidu (BIDU) - I try not to red thumb "great" business at high prices, since they have an awful habit of growing into their valuation. I screwed up here. Especially Baidu.
Maybe I should close my red thumbs on LNKD, P, SSYS, DDD. Unlike most of my red thumbs, I can see they have the potential to be very good businesses.
Headwaters (HW) - Two mistakes here. Overemphasis on earnings versus cash flow, and shorting through a phase change from unprofitable to profitable.
Clearwire (CLWR) - They were acquired by Sprint after Softbank acquired control of Sprint. Don't overweight operating results versus asset value, both are important. When a corporate investor has a large stake, short positions should generally be avoided. They are likely to make irrational acquisition decisions because of empire building and sunk cost fallacy.
American Airlines (AAMRQ, now AAL) - I red thumbed them in bankruptcy and made several errors. Every bankruptcy is different, statistical approaches as opposed to individual approaches should be avoided. Didn't read any of the bankruptcy documents. Underemphasized cash flow.
Ariad Pharmaceuticals (ARIA) - They had a successful phase III trial and introduced a drug to market. Since then the results have disappointed. Phase III results are a dangerous time to short.
WMI Holdings (WMIH) - Didn't consider the value of NOLs. Also didn't consider the reflexive nature of the stock price. With their NOLs they are primarily valuable as a merger piece, therefore higher price = higher value.
Netsuite (N) - See Clearwire. It's super expensive, but Larry Ellison's involvement here is a red flag.
ParkerVision (PRKR) - Should have ended once they graduated from vaporware manufacturer to patent troll. Patents are unpredictable. My thesis changed, 90% of the vaporware manufacturers I've red thumbed have declined, but only 50% of patent trolls.