If I did not use a mechanical method for ranking replacement candidate stocks against my portfolio, I would not let so many stocks go onto my potential replacement list. Today I updated my industry model, and here it is:

Industry Groups January 2008

(If you have any difficulty downloading that, let me know. I’ve been having trouble with that.)

From that, I ran a bunch of screens, adding in some technology industries that have been hit of late. Here are the additional tickers that will be added to my candidates list: AMIE ASYT BBBY BC BELM BGFV BIG BNHN BRLC BWS CAB CBR CHRS CHUX CMRG CRH CTR CWTR DBRN DECK DFS DSPG DSW ESEA EXM EXP FHN FINL FRPT FSS GASS HGG HLYS HTCH HZO IDTI IKN IM IMOS JAS JNS KSWS KWD LF LIZ LNY LSI MIPS MRT NSIT NSTC NTY ODP OPMR OVTI OXM PERY PLAB POOL RCRC RENT ROCK RSC RT RUTH SAIA SHOO SIG SMRT SNA SNX SONC SSI TJX TOPS TUES VLTR VOXX ZQK

Now, the mechanical ranking system is supposed to be a simple way of prioritizing value stocks, and typically it does pretty well in directing my attention to the stocks that I should analyze, not necessarily the ones I should buy. That’s true of any screening method, no matter how simple or complex. You always find some companies that look really good initially, but got there because of data errors, accounting mis-characterizations, or a business situation that was vastly different when the accounting snapshot was taken.

Now, after all of this work, I’m only trading 3-4 stocks into and out of my portfolio of roughly 35 stocks. But the idea is to end up with a portfolio with better offensive and defensive characteristics, such that the relative performance will be good, and should the market turn, I will be in the industries and companies with a lot of potential to outperform.