the stock screens of Kevin Matras

Every week, Zacks presents a stock screen by Kevin Matras. It always touts a very back-tested high return. (I suppose if it didn't output a high return, they wouldn't present it.) As far as I remember, the screen always has the Zacks ranks as one of the criteria (makes sense) and usually an indicator that the stock is moving up (momentum), so as such they probably should all be viewed as short term (three to six months or so) screens. This week's screen looks interesting.

"Cheap stocks and Big Returns"

This screen is one of Kevin Matras' favorites for when he's
looking for low priced stocks. The premise behind this screen
was to try and find cheap stocks (stocks at or less than $15)
that were trading at (or consolidating) just under their 52
week high, in an effort to `get on board' before they broke-out
to new highs.

In other words, Kevin wanted the stocks to be near their highs,
but most of all, he was looking for stocks that had been turned
back from their recent highs, had consolidated their advances
on their trek higher to those highs and were just now starting
to make a run at the 52 week high again.

Kevin is a big fan of getting into basing patterns (relatively
narrow trading bands) after an uptrend has been established.
Especially near recent price highs, since stocks making new
highs tend to make even higher highs.

He has found this screen to be an ideal strategy for finding
low priced stocks with a high probability of success.

Parameters:

- It searches for stocks trading at or below $15.

- They also have to have a Zacks Rank of 1.

- They have to be trading within 10% of the 52 Week high.
(Expressed as Current Price / 52 Week High greater than or
equal to .90) (Simply put, Kevin is looking for stocks high up
in their uptrend.)

- The % Price Change over the last 4 weeks has to be greater
than or equal to 10% but not more than 20%. (Kevin is looking
for stocks on the move, but not ones that have moved so much,
so quickly, a correction could be in store. Since 10% seems to
get people's attention while follow-thru at 20% typically
signals the beginning of a `trend' or a `breakout', Kevin
wanted to be alerted BEFORE a breakout was seen.

- And lastly, for good measure, he wants the Beta to be less
than or equal to 2. (Active stocks are good, but wildly
volatile ones are not.)

Results:

Kevin ran a series of tests over the last 4 year time span
(2001, 2002, 2003, and YTD 2004 -- thru 11/26/04) as well as a
series of tests for each of the last 4 years individually. He
rebalanced the portfolio every four weeks and started each run
on a different start date so each test would be rebalanced over
a different set of four-week periods. (This was done to
eliminate coincidence and verify robustness.)

Over the last 4 years, this strategy has shown an average
annualized gross return of 66.9% with an average win ratio
(winning periods divided by the total number of periods) of
74%. And it holds on average of only 3-5 stocks in your
portfolio each month.

In 2001, the average annualized gross return was 53.5%, with an
average win ratio of 71%. (This year's avg. # of stocks held
per period was 8.)

In 2002, the average annualized gross return was 36.2%, with an
average win ratio of 70%. (4 stocks / avg. per period.)

In 2003, the average annualized gross return was 167.2%, with
an average win ratio of 87%! (3-5 stocks on average.)

And so far, 2004's YTD (thru 11/26/2004) average annualized
gross return is up 45.5%, with an average of 3 stocks per run.
(The S&P is up 7.9% comparatively.)

As of Mon., 12/13/04, 5 stocks had qualified. They are

AKS AK Steel Holding Corp.
BABY Natus Medical, Inc.
CIB Bancolombia.
TZIX TriZetto Group, Inc.
UHCO Universal American Financial Corp.

Note: Even though this screen will generally produce on average
of 3-5 stocks per period, there will be times where literally
no stocks will qualify due to the narrowness of the parameters.

Tip: Since this screen has such a great track record and such a
high success rate, if nothing comes through on Kevin's first
pass, he'll run it day after day, until the screen spots
something.

(This is one of the reasons why Kevin runs so many backtests
using different start dates in my analysis. He wants to make
sure that the strategy has a history of picking good stocks
`at any time'.)

Since I don't do short-term trading, I have never tried any of these screens. But they do look like they have potential if you stick to the system.

[6/15/05] How have the picks done?

AKS has gone from about 14 to about 7

BABY went from about 8 to 10.5

CIB went from about 12 to 16

TZIX went from about 9 to about 13.5

UHC went about 15 to near 24 today

All in all, it looks like the strategy worked out despite one pick losing 50% of its value.