Finding the Magic in MFI

As my readers know, the Magic Formula Investing approach has had pretty mixed results for many investors since it's debut in January 2006.  In addition, my tracking portfolios have also had their fair share of struggles.

That being said, my empirical evidence does suggest it is working better for

• larger cap stocks (say over $700m) and
• for stocks that pay at least a 2.6% dividend (which honestly is largely a subset of 1st bullet),
• for stocks that have just two or three letters in their ticker

At first I guffawed a  bit at that last point.  But I think it is related to the first two points, in that 2 and 3 letter stocks are often more stable and "old economy" than 4 or 5 letter tickers. So I think stable 4 letter stocks like INTC, MSFT and CSCO are fine... but perhaps I am rationalizing.

So in my first new tranche, which I started in mid-August is off to a terrific start.  Recall, I used these points in my selections:  MSFT and NSU paid the dividends. DLB, KFY and UIS are all three letter tickers with market caps 700m+.  NSU has been the big winner so far... it was so cheap.

August 15 Tranche
Stock	Start	Current	Dividend	Pct Gain	R3K Gain
DLB	$34.43	$33.51	$0.00	-2.7%	0.4%
KFY	$13.98	$14.91	$0.00	6.7%	0.4%
MSFT	$30.19	$30.33	$0.00	0.5%	0.4%
NSU	$3.62	$4.28	$0.00	18.4%	0.4%
UIS	$19.96	$21.50	$0.00	7.7%	0.4%
Totals				6.1%	0.4%

So I am already starting to think about my mid November tranche.  Stocks on my short list include (note there are a few "riskier names there and I would only use one of them):

SAI
GTAT
CSCO
HFC
TW
CA
CHE
AAPL
BKE
STX
GA
CYOU
KLIC
KLAC
SOHU
LPS
JOY