i Zacks Investment Research, Inc. Research Wizard www.zacks.

com/RW
Table of Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
The Importance of Screening and Backtesting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Trading the Strategies and Calculating Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Chapter 1:
Upgrades and Revisions 2 - A Winning Strategy for Beating the Market . . . . . . . . . . . . . . . . . . . 7
Chapter 2:
Magic Numbers? and Relative Valuations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Chapter 3:
Filtering the Zacks Rank: Tips on Trading the Zacks Rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Chapter 4:
Return on Equity (ROE) - Part of a Winning Screening Strategy . . . . . . . . . . . . . . . . . . . . . . . . . 19
Chapter 5:
Earnings Acceleration: The Difference Between Good Stocks and Great Stocks . . . . . . . . . . . 24
Chapter 6:
Big Money . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Chapter 7:
Increasing P/Es for Stocks on the Move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Chapter 8:
Cheap Stocks and Big Returns (a.k.a. Breakouts) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Chapter 9:
Something for Everyone: Growth and Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Chapter 10:
Creating a Custom Consensus of Your Winningest Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Chapter 11:
More Winning Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Chapter 12:
Additional Comments on Backtesting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
You Can Do It! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
ii Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Introduction

The Screens that I go over in this book are just some of the screens that
weve published in our Screen of the Week articles.

Some of these screens are proven profitable trading strategies that I
created with and backtested in the Research Wizard.
And others are simply unique ways to screen for winning stocks. (In
fact, some of these screens shatter common myths on evaluating stocks,
complete with eye-opening statistics.)
There are also other screens that didnt make it into this book, for no
other reason other than we wanted to keep this book short and to the
point. (Its not meant to be an encyclopedia of screens, our editor kept
reminding me.) Plus, we didnt want to give all of our secrets away all at
once.
But the screens presented here (and the ones we left out) are ALL
available in the Research Wizard program.
So please read on and learn how you can pick better stocks that make
more money, now!
Kevin Matras
Zacks Investment Research, Inc.
iii Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
The Importance of Screening
and Backtesting
Why Should I Use a Stock Screener?
The short answer is:
Because theres over 10,000 stocks out there and you need a way to find the good ones.
The longer answer is:
Other than buying the stocks that are talked about on TV or written about in the paper (not to mention tips
from a friend), how else are you going to find stocks that meet certain fundamental characteristics.
Even if you dont use a screener now, most people still
do their own screening one way or another. They may
hear that a stock has a certain Growth Rate, or a certain
P/E Ratio or Sales Surprise, or whatever. They then find
themselves listening for or reading about stocks that
meet this criteria.
Well if you want to find stocks that meet certain criteria,
you can find them quickly and easily with a stock
screener.
But, just because you narrow down 10,000 stocks to
only a handful, doesnt necessarily mean that youve
picked the best stocks on the planet.
You might have picked the worst ones.
But how will you know?
Backtesting!
Once youve created a screen, you can then backtest it to see how good (or bad)
your screening strategy has performed.
In other words, does your screen generally find stocks that go up once theyve been identified, or does your
screen generally find stocks that get buried once theyve been identified?
This is good stuff to know.
With backtesting, you can see how successful your stock picking strategy has performed in the past, so youll
have a better idea as to what your probability of success will be now and in the future.
just because you narrow down
10,000 stocks to only a handful,
doesnt necessarily mean that
youve picked the best stocks
on the planet.
You might have picked
the worst ones.
iv Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
The Importance of Screening and Backtesting continued...
Of course, past performance is no guarantee of future results, but what else do you have to go by?
Think about it; if you saw that a stock picking strategy did nothing but lose money, year after year,
period after period, stock after stock, over and over again (you get the point), theres NO WAY youd
want to trade that strategy or use that screen to pick stocks with.
Why?
Because its proven that it picks bad stocks.
Sure, it may turn around and start picking winners, but it may also continue to pick losing stocks the
way it always has.
One the other hand;
... what if you saw a strategy that did great year after year, period after period (you know where this
is headed), youd of course would want to trade that strategy.
Why?
Because its proven to be a profitable trading strategy.
And while it may start picking losers all of a sudden (now that youre using it ), it may also continue
to pick winning stocks, just like it had been doing over and over before.
Keep in mind, a screening and backtesting program isnt a box of magic.
But its a great way to see what works and what doesnt BEFORE you put
your money at risk!
Ill end this with a recollection of a conversation I had with someone a while back who was stuck in a losing
stock.
I asked him why he was still in it if it kept on losing money.
He said that he didnt think it would go much lower from here.
I asked him if he thought it would go this low when he bought it.
(He of course said no.)
I then asked him if he thought itd go up from here.
His answer was probably not right away and then he added that it could possibly still fall a
bit more from here.
I told him there are plenty of stocks going straight up; Why dont you get out of that
one thats losing you money and get into a better one.
His answer was; he didnt know of any better stocks to get into.
v Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
The Importance of Screening and Backtesting continued...
I then asked him; what if you did know of a better stock to get into, would you do
it?
His answer of course was: YEAH! But he quickly added that he didnt know how to find
better stocks.
That last comment said it all.
He was in losing stocks because he didnt know how to pick better ones.
But if he had a proven, profitable, stock picking strategy, he could.
Dont get me wrong, just because you have a great
strategy for picking winning stocks, it isnt going to
preclude you from ever having another loser. On the
contrary, even some of the best strategies only have
win ratios* of 70% or 80%. (NOT 100%.)
But if your strategy picks winners far more often than
losers, once you find yourself in a losing trade, you can
quickly cut your losses and feel confident that your next
pick will have a high probability of succeeding.
And thats why someone should use a Screener and a Backtester.
No Hype
This book of screens will not be hype for the Research Wizard: Stock Picking and Backtesting Software.
But I should add that most of these screens would be impossible to do on any other program other than the
Research Wizard.
With over 8,600 stocks
650 different fundamental data items
The ability to create you own custom calculations and comparisons
Access to historical data
Not to mention backtesting
... this is one of the most capable screeners available.
Please read on for some great stock picking strategies and ideas.
* Win Ratio: see Trading the Strategies and Calculating Performance

if your strategy picks winners

far more often than losers,
you can feel confident that your
next pick will have
a high probability of success.
vi Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Trading the Strategies
and Calculating Performance
Trading the Strategies
All stocks are purchased with an equal dollar amount. At the end of the holding/rebalancing period**, the
screen is run again, keeping the stocks that remain qualified, selling the stocks that no longer qualify and
buying the new stocks that newly qualify.
** Holding/Rebalancing Period: the amount of time a stock will be held once it qualifies the screen. In
most cases, the holding/rebalancing period is four weeks (unless otherwise indicated).
* Win Ratio: the number of winning (profitable) holding periods out of the total number of available
holding periods within the backtested time span. For example; if there were 39 winning holding
periods out of a total of 52 available holding periods, the win ratio would be 75%.
Calculating Performance
At the beginning of each holding period, a list of stocks
(portfolio) is generated. The periods returns are calculated
using the % change in price from the beginning of the
holding period to the end of the holding period, plus any
applicable dividends. The returns for the portfolio is the
arithmetic mean of the returns for the individual companies
in the portfolio.
Compounded performances (when stated), were calculated
by taking a hypothetical starting equity amount and
calculating the total return for the period. Each subsequent
period then used the resulting equity balance as its start to
calculate that periods total return.
No consideration has been given to commission costs,
slippage or any other real-world constraints, in any of the
performance calculations.
Disclaimer: Stock trading/investing involves risk and you
can lose some or all of your investment. Hypothetical results
may not always be duplicated in the real world. Backtesting can also at times produce an unintended look-
forward bias. In addition, hypothetical trading does not involve financial risk, and no hypothetical trading record
can completely account for the impact of financial risk in actual trading, not the least of which is the ability
to withstand losses or to adhere to a particular trading strategy in spite of trading losses. These are material
points which can also adversely affect actual trading results.
1
Chapter
7 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Upgrades and Revisions 2
- A Winning Strategy for Beating the Market
Lets get started with one of my favorite screening strategies - Upgrades and Revisions 2.
It focuses primarily on stocks with upward Earnings Estimate Revisions and Rating Upgrades.
It generates on average of approximately 8-10 stocks a month, has an excellent win ratio (74%) and has shown
consistently impressive returns, year after year after year.
The Parameters to this Strategy are:
Zacks Rank = 1
(The screen starts out by selecting only those stocks with a Zacks Rank of 1 -- Strong Buy. With the
Zacks Rank proving itself to be on of the best (if not the best) rating system out there (avg. annual return
of 34.24% since 1988), this is a great way to start things off.)
% Change Q1 Estimates over the last 4 weeks >= 0
(Only earnings with fresh upward revisions for the current quarter, or at the very least no downward
revisions, qualify. This is one of the most important filters and one of the reasons why this screen is called
Upgrades and Revisions.)
% Rating Change over 4 weeks >= 0
(Since analysts ratings have such a huge bias to the upside, Im excluding anything thats been even
slightly downgraded. Plus, since stocks receiving upward estimate revisions will generally see the brokers
that cover them upgrade their ratings, this makes sense. This too is another big item and where the
Upgrades name came from.)
P/E using 12 mo. Forward EPS Estimates <= 65
(Its been shown that stocks with forward P/E ratios over 65 tend to underperform the market. So were
keeping the lid on at 65 and under.)
5 Year Historical EPS Growth >= 17
(Yes, 17. When this screen was created, the average stock in the Zacks universe had a 5 year historical
growth rate of 11%. Basically, this strategy is looking for companies that are doing better than the
average historical growth rate, and at present, its seeking companies that are more than 50% above this
threshold.)
% Change Actual EPS (F0)/(F-1) >= 0
(In addition to the 5 yr. historical growth rate, were also looking for positive EPS growth over this last year
as well. (F0) being the last completed year and (F-1) being the year before the last completed year.)
% Change Actual EPS (Q0)/(Q-1) >= 0
(Were also looking for positive EPS growth last quarter over the one before that - (Q0) being the last
completed quarter and (Q-1) being the quarter before the last completed quarter.)

8 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW

Upgrades and Revisions 2 continued...
And
% Change Actual EPS (Q-1)/(Q-2) >= 0
(Positive EPS growth for the quarter over quarter period before the last completed quarter. (Q-1) over
(Q-2).
In short, the above two expressions demand two quarters of positive EPS growth.
And as a whole, the 5 yr., 1 yr. and last 2 quarters of EPS growth paint a solid earnings picture over the
longer term historical and most recent historical past.
Last EPS Surprise >= 0
(The more likely a company has surprised in the
past, the more likely they are to surprise in the
future. This is true for both positive and negative
surprises. So no negative surprises allowed.)
Price/Sales ratio <= 4
(I had great success with the Price/Sales ratio
being between 2, 3 and 4. In general, a lower Price/
Sales ratio is better. But I went with 4 because it
didnt narrow down the stock selection so much.)
Price >= 3
(And finally, all the stocks have to be trading at a
minimum of $3 or higher. Although, with all of the items above, the typical pick has an average stock price
of approx. $30.)
The Results
I ran a series of tests over each of the last 5 years (2001 thru 2005). I rebalanced the portfolio every four
weeks and started each run on different start dates so each test would be rebalanced over a different set of
four-week periods. This exercise was done to eliminate coincidence and verify robustness.
Over the last five years, this strategy has shown an average annualized gross return of 54.4% a year, with an
average win ratio (winning periods divided by the total number of periods) of 74%. And again, it produces on
average of 8-10 stocks for your portfolio each month.

To break it down further; in 2001, the average annualized
gross returns were 55.8%, with an average win ratio of
71%.
In 2002, the average annualized gross returns were
42.1%, with an average win ratio of 68%.
In 2003, the average annualized gross returns came
in at a whopping 108.5%, with an average win ratio at
92%.

This screen, by far, has the most

hoops to jump through.
But the cumulative effect
it has on finding the right stocks
makes it one of the most
rock-solidest strategies!
This screening strategy
has shown consistently
impressive returns, year after
year after year.
9 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Upgrades and Revisions 2 continued...
In 2004, the average annualized gross returns came in at 49.3%, with a win ratio of 72%.
And in 2005, the average annualized gross returns were up 16.5, with a win ratio of 65%.
(The below chart shows the average annualized returns.)
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
For further illustration, I ran a continuous backtest starting on 1/5/2001 and going thru 12/30/2005,
rebalancing the portfolio every four weeks. I also compounded the periods returns to greater simulate how one
would invest. (For example; if the starting equity was $10,000, and the first periods gain was 10% ($1,000),
the next period would invest with $11,000. If that next period is off -10% (-$1,100), the period after that would
invest $9,900.) Also, in this illustration (and all of the other performance illustrations), no consideration has
been given to commission costs, slippage or any other real-world constraints.
The total return from its start date on 1/5/2001 thru 12/30/2005, is 560.2%, with a win ratio of 74%
(48 winning periods out of 65 total periods). A hypothetical $10,000 invested (sans commissions and etc.),
increased to $66,022.13. By comparison, a hypothetical $10,000 invested in the S&P during that time would
have increased to only $10,281.83 (or 2.8%). (See the next page for performance statistics.)
10 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Upgrades and Revisions 2 continued...
(The below chart and table shows the compounded returns from 1/5/2001 thru 12/30/2005.)
(1/5/2001 thru 12/30/2005) Upgrades and Revisions 2 S&P 500
Compounded Total Return 560.2% 2.8%
$10,000 Starting Equity $66,022.13 $10,281.83
Win Ratio 74% (48/65)
Avg. # of Stocks Held 8-10
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
Trading the Strategy
All stocks are purchased with an equal dollar amount. At the end of the holding/rebalancing period (4 weeks),
the screen is run again, keeping the stocks that remain qualified, selling the stocks that no longer qualify and
buying the new stocks that newly qualify.
Get It
This screen is available in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder (SoW -- short for Screen of the Week)
Selecting the file: bt_sow upgrades and revisions2
11 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Upgrades and Revisions 2 continued...
More Upgrades and Revisions
There are two other versions of the Upgrades and
Revisions strategy available in the Research Wizard as
well.
One of the versions generates on average of only 4-5
stocks per period as opposed to 8-10.
All of the parameters are the same except that the
Zacks Rank of 1, cant be more than four weeks old (i.e.,
the stocks had a Zacks Rank of 2 or higher four weeks
ago and have since been changed (upgraded) to a 1
within the last four weeks.
(This change will increase your turnover rate (and
commission costs, etc.), since every four weeks, a
completely new list of stocks is generated.)
This was actually the very the first version of the
Upgrades and Revisions strategy. I later revised it by
taking out the no more than four weeks old item and named it Upgrades and Revisions 2.
(Aside from the number of stocks and the turnover rate, the performance results are very similar for both.)
This screen can also be found by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: bt_sow upgrades and revisions
Even More Upgrades and Revisions
And lastly, theres another version that can be found in our Profit Tracks folder. This strategy and the other
Profit Track strategies in this folder, typically produce on average of 25-30 stocks per period.
More changes were made to this version so as to increase the number of qualified stocks coming thru. But the
main principle of Upgrades and Revisions still drives this screen.
The biggest change however was including Zacks 2s, in addition to the 1s (both Buys and Strong Buys).
This produced a bigger stock list for larger investors, but still limited it to the best of the best. (See the next
page for its performance statistics.)
All stocks are purchased
with an equal dollar amount.
At the end of the
holding/rebalancing period,
the screen is run again,
keeping the stocks that
remain qualified, selling the
stocks that no longer qualify
and buying the new stocks
that newly qualify.
12 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Upgrades and Revisions 2 continued...
(The below charts and table depict the compounded returns.)
(1/5/2001 thru 12/30/2005)
Upgrades and Revisions
(Profit Tracks)
S&P 500
Compounded Total Return 449.6% 2.8%
$10,000 Starting Equity $54,960.50 $10,281.83
Win Ratio 72% (47/65)
Avg. # of Stocks Held 25-30
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
This screen can be found in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the ProfitTracks folder
Selecting the file: upgrades and revisions
No matter which Upgrades and Revisions strategies you choose, they all have excellent performance results.
2
Chapter
13 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Magic Numbers? and Relative Valuations
So many people I speak to, seem to believe there are some magic numbers out there that equate to stock
picking success.
Two things in particular that I hear over and over again when Im helping people with their screening strategies,
relates to P/E Ratios and Price/Book Values.
For some reason, many people believe that P/E Ratios of 20 or less and Price/Book Values of 1 or less are
these so-called magic numbers.
Unfortunately, statistics prove otherwise.
I first wrote about this in June of 2003.
In that article, I presented some very eye opening
statistics.
-- from 6/30/03;
Looking at the best performing stocks
over the last six months (as qualified by
stocks that were trading at $10 or higher
at the beginning of the period and that have
increased by 50% or greater), the average
P/E Ratio over that time (using 12 mo. EPS
Actuals), was more than 50.
Thats more than 2 1/2 times the magic P/E numbers.
True, there were some in there with P/Es under 20, but you wouldve missed a lot of
fantastic winners if you excluded those over 20.
As for the average Price/Book Value, it was more than 4. (Not 1 or even 2, but over 4!) In
fact, not one of the stocks on that list [6/30/03] has a Price/Book Value of 1 or less.
So if you insist on looking for stocks with low valuations (P/E, P/B), try comparing them
to their relevant Industries.
Why? Because while a small number of stocks traded at P/Es of less than 20 with no
stocks having P/Bs of 1 or less, ... 70% of those stocks on that list of winners had P/Es
under the average for their Industry and 48% had P/Bs under the average for their
Industry.
Very eye opening!
I republished this article again in October of 2003 and August 2004 with updated stats.
many people believe
that P/E Ratios of 20 or less
and Price/Book Values of 1
or less are magic numbers
(to stock picking success).
Unfortunately, statistics
prove otherwise.
14 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Magic Numbers? and Relative Valuations continued...
-- from October 2003;
Once again, looking at the best performing stocks over the last six months [from 10/2003]
(as qualified by stocks that were trading at $10 or higher at the beginning of the period
and that have increased by 50% or greater), the average P/E Ratio was over 30 at the
beginning of the study and over 45 at the end. (Thats respectively 1 1/2 times and more
than 2 times the magic P/E number of 20 or less.)
The Price/Book Value started at over 3 and wound up at over 4. Again, way over the
magic number of 1 or less. (That time, only 3% of the stocks on that list (thats only 9
stocks out of 288), had a Price/Book Value of 1 or less.)
When compared to their relevant Industries
however, close to 70% of the stocks on that
list had P/Es under the average for their
Industry and over 50% had P/Bs under the
average for their Industry.
In August of 2004, I re-ran this article, but was now
using the choppy to lower market of 2004 [lower at
that time] as opposed to the bull market of 2003. The
statistics however, remained virtually unchanged.
-- from August 2004;
This time, since the flat/lower market [lower then] made it harder to find a large sample of
stocks that increased by 50% or greater, I expanded the 6 month performance period to
a YTD (Year-to-Date) period (8 months since it was August), and I reduced the minimum
price down to $5 instead of $10. The volume qualifier stayed the same though: average
daily volume >= 50,000 shares.
The study showed that the beginning P/E Ratios averaged out to be over 33 and increased
to over 41. Again, 1 1/2 times and more that 2 times the magic number of 20. And this
time, the P/B value started at over 4 and rose to more than 5, with only 1 stock on that
list of 67 stocks having a Price/Book Value of 1 or less.
But when compared to their relevant
Industries, 70% had P/E ratios under the
average for their Industry and close to
60% had P/Bs under the average for their
Industry.
So instead of thinking about low valuations as an
absolute number, try thinking about them as a relative
measure.
Ive found that companies that are outperforming their
Industries on earnings but are undervalued to their group in terms of valuations are great candidates.
Looking at the best
performing stocks ... the
average P/E Ratio was
more than 2 times the
magic number of 20 ...
True, there were some in there
with P/Es under 20, but
you wouldve missed a lot of
fantastic winners if you
excluded those over 20.
15 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Magic Numbers? and Relative Valuations continued...
Parameters
Ive created such a screen and have included the following parameters;
P/E Using 12 month EPS < XIndMean
(Stocks with P/E ratios that are lower than the average P/Es for their relevant Expanded (X) Industries.)
Price/Book < XIndMean
(Price to Book values that are lower than the average P/B values for their relevant Expanded Industries.)
Current Price/52 week High >= .90
(Companies with upward price momentum and that are trading within 10% of their 52 week high.)
% Change Actual EPS (Q0)/(Q-1) > XIndMean
(Were looking for EPS growth over the last quarter over quarter period to be greater than the average EPS
growth for its relevant Expanded Industry.)
And
% Change Actual EPS (Q-1)/(Q-2) > XIndMean
(EPS growth over the previous quarter over quarter period to be greater than the average EPS growth for
its relevant Expanded Industry.)
In short, its looking for stocks that have shown greater EPS growth over the last two quarter over quarter
periods than their Industrys average.
Price >= $5
Average 20 Day Share Volume >= 50,000
(Average daily trading volume (over the last 20 days) is a minimum of 50,000 shares or greater.)
Get It
This strategy can be found in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: sow_relative valuations
Try incorporating some of these ideas into your own stock picking strategies.
And while relative valuations arent available in most screeners, they are available in the Research Wizard. (Not
to mention over 650 other fundamental database items too).

3
Chapter
16 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Year
Zacks Rank #1
Stocks Return
S&P 500
Return
1988 37.46% 16.20%
1989 36.09% 31.70%
1990 -2.97% -3.10%
1991 79.79% 30.40%
1992 40.65% 7.51%
1993 44.41% 10.07%
1994 14.34% 0.59%
1995 54.99% 36.31%
1996 40.93% 22.36%
1997 43.91% 33.25%
1998 19.52% 28.57%
1999 45.92% 21.03%
2000 14.31% -9.10%
2001 24.27% -11.88%
2002 1.22% -22.10%
2003 74.74% 28.69%
2004 28.79% 10.87%
2005 17.97% 12.01%
Annualized 34.24% 13.13%
Disclosure: Zacks Rank performance is the total return (price changes + dividends) of equal weighted portfolios, consisting of those stocks with the indicated Zacks Rank, assuming
zero transaction costs. Results reflect the reinvestment of dividends and other earnings. These returns are not the result of a backtest. Simulated results do not represent actual trading
and may not reflect the impact that economic and market factors might have had on decision-making if an adviser were actually managing a clients money. The stocks in the Zacks
Rank portfolios were available to Zacks clients before the beginning of each month (monthly rebalancing).
Filtering the Zacks Rank
- Tips on Trading the Zacks Rank
Im sure most everyone reading this knows that the Zacks Rank is probably the most effective rating system
out there. Good markets or bad, stocks with a Zacks Rank of #1, continue to outperform.
In fact, since 1988, the average annualized return of Zacks #1 Ranked stocks is up over 34.2% a year.
Wow!
But since there are typically over 200+ stocks Ranked a #1 at any time,
its important to know what other filters to apply to the Zacks Rank to
generate a smaller (more tradable) watchlist.
Two filters in particular, when added to the Zacks Rank of #1, not only
narrows down the number of qualified stocks to a practical portfolio
size (approx. 10-12 stocks), it often times increases its performance
as well.
Parameters
The two filters Im talking about are;
% Change (Q1) Est. over 4 Weeks > 0
(Positive current quarter estimate revisions over the last four
weeks.)
% Broker Rating Change over 1 Week > 0
(Positive avg. broker rating changes over the last week.)
And of course
Zacks Rank = 1
These two items added to the Zacks #1 Rank, produce powerful results!
Results
I ran a series of separate tests on the Filtered Zacks Rank strategy, over each of the last four years (2002,
thru 2005). I rebalanced the portfolio every four weeks and started each run on different start dates so each
test would be rebalanced over a different set of four-week periods. This is done to eliminate coincidence and
verify robustness.

17 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW

Filtering the Zacks Rank continued...
In 2002, the Zacks #1 Rated stocks returned just over 1%, with an average portfolio size of approx. 200
stocks. An impressive return when compared to the S&P 500s 22%.
But holding onto 200 or so stocks isnt doable for most investors. But when adding the two aforementioned
filters, the portfolio size shrinks to a tradable 10 stocks (on average), and a phenomenal 18.1% return.
In 2003, the Zacks #1 list (approx. 200 stocks) did nearly 75% in comparison to the S&P 500s almost 29%.
But the filtered Zacks Rank narrowed that list down to
only 10 stocks (on average) with a return of over 66%.
(And while its true the filtered Zacks Rank produced
a smaller return than the full Zacks Rank (66% vs.
75%), rebalancing only 10 stocks a month is far more
manageable than 200.)
In 2004, the annualized returns for the Zacks #1 Ranked stocks was up 28.8% with an average portfolio size
approx. 200 stocks. (The S&P was up only 10.9%.)
Yet the filtered Zacks #1s annualized returns were up 30.3%, with again, only 10 stocks to hold on average.
And it was right on track again in 2005. The annualized returns for the complete list of the #1 Ranked stocks
(again, approx. 200 stocks) was up nearly 18% in comparison to the S&Ps 12%! But the filtered Zacks
Ranked #1s were up almost 43%. (Thats more than twice the return on a fraction of the stocks.)
(The chart below shows these annual returns, compounded over the last four years -- 2002 thru 2005).
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
... the filtered Zacks Rank.
Trade fewer stocks
for bigger returns.
18 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Filtering the Zacks Rank continued...
If youre looking for a reliable way to trade some of the best of the Zacks #1s, try the filtered Zacks Rank;
trade fewer stocks for bigger returns.
Trading the Strategy
All stocks are purchased with an equal dollar amount. At the end of the holding/rebalancing period (4 weeks),
the screen is run again, keeping the stocks that remain qualified, selling the stocks that no longer qualify and
buying the new stocks that newly qualify.
Get It
This screen can be found in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: bt_sow_filtered zacks rank
4
Chapter
19 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Return on Equity (ROE)
- Part of a Winning Screening Strategy
This screen uses the Return on Equity (ROE) measure as one of the main components in this strategy.
This too is one of my favorite strategies and one of the most popular.
ROE is one of the quickest ways to see whether a company is creating assets or gobbling up investors cash.
ROE = income / common equity
For instance; if the ROE is 10%, then ten cents of
assets are created for each shareholder dollar that was
originally invested.
Knowing the company is generating assets on invested
capital rather than burning thru it is a great starting
point.
Parameters
ROE >= 10
(The median ROE value for all of the stocks in the Zacks Universe is under 10. So any company with
shareholder equity less than this benchmark is disqualified.)
Zacks Rank = 1
(The Zacks Rank (which is considered by many to be the best rating system out there), looks at upward
earnings estimates revisions (amongst other things), and will get us into companies whose forecasted
earnings are getting stronger.)
% (Broker) Rating Strong Buy = 100(%)
(Since broker ratings are typically skewed wildly to buy and strong buy, I decided to cancel out any
company where the brokers arent fully on board.)
Price/Sales <= 1
(A low price to sales ratio (1 and below for example), is usually thought to be of better value, since the
investor is paying less for each unit of sales.)
Price >= 5
(And for good measure, all of the stocks have to be trading at a minimum of $5 or higher. Most money
managers wont touch anything under $5.)

ROE (Return on Equity)

is one of the quickest ways
to gauge whether a company is
creating assets or gobbling up
investors cash.
20 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Return on Equity (ROE) continued...
The Results
I ran a series of tests over each of the last 5 years (2001 thru 2005). I rebalanced the portfolio every four
weeks and started each run on different start dates so each test would be rebalanced over a different set of
four-week periods. This exercise was done to eliminate coincidence and verify robustness.
Over the last five years, this strategy has shown an average annualized gross return of 81.1% a year, with an
average win ratio of 75%. And it produces on average of 4-5 stocks for your portfolio each month.
In 2001, the average annualized gross return was 59.7%, with an average win ratio of 63%.
In 2002, the average annualized gross return was 89.8%, with an average win ratio of 79%.
In 2003, the average annualized gross return was 125.7%, with an 88% win ratio.
In 2004, the average annualized gross return was 44.3% with a win ratio of 73%.
And in 2005, the average annualized gross returns were 86.2% with a win ratio of 73%.
(The below chart shows the average annualized returns.)
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
For further illustration, I ran a continuous backtest starting on 1/5/01 and going thru 12/30/05, rebalancing
the portfolio every four weeks. I also compounded the periods returns to greater simulate how one would
invest. (For example; if the starting equity was $10,000 and the first periods gain was 10% ($1,000), the next
21 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Return on Equity (ROE) continued...
period would invest with $11,000. If that next period is off -10% (-$1,100), the period after that would invest
$9,900.) Also, in this illustration (and all of the other performance illustrations), no consideration has been
given to commission costs, slippage or any other real-world constraints.
The total return from its start date on 1/5/2001 thru 12/30/2005, is 1,698.7%, with a win ratio of 80%
(52 winning periods out of 65 total periods). A hypothetical $10,000 invested (sans commissions and etc.),
increased to $179,870.45. By comparison, a hypothetical $10,000 invested in the S&P during that time would
have only increased to $10,281.83.
(The below chart shows the compounded returns from 1/5/2001 thru 12/30/2005.)
(1/5/2001 thru 12/30/2005) ROE Strategy S&P 500
Compounded Total Return 1,698.7% 2.8%
$10,000 Starting Equity $179,870.45 $10,281.83
Win Ratio 80% (52/65)
Avg. # of Stocks Held 4-5
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
Trading the Strategy
All stocks are purchased with an equal dollar amount. At the end of the holding/rebalancing period (4 weeks),
the screen is run again, keeping the stocks that remain qualified, selling the stocks that no longer qualify and
buying the new stocks that newly qualify.
22 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Return on Equity (ROE) continued...
Get It
This screen is available in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: bt_sow_roe
More ROE
Theres also another ROE strategy available in the Research Wizard and this version generates on average of
25-30 stocks per period as opposed to the one aboves 4-5.
There are a few different parameters as well, but the focus still remains on the ROE. And it too has very
impressive results and an excellent win ratio.
(The below charts and table depict the compounded returns.)
(1/5/2001 thru 12/30/2005) ROE (Profit Tracks) S&P 500
Compounded Total Return 365.6% 2.8%
$10,000 Starting Equity $46,559.88 $10,281.83
Win Ratio 72% (47/65)
Avg. # of Stocks Held 30
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
23 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Return on Equity (ROE) continued...
This screen can also be found in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the ProfitTracks folder
Selecting the file: roe
5
Chapter
24 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Earnings Acceleration:
- The Difference Between Good Stocks and Great Stocks
In good markets or bad, strong earnings are one of the most important things that influence stock prices.
But instead of just looking at the most recent quarters earnings, try looking for earnings acceleration too.
Studies have shown that almost all of the most successful stocks in the past had displayed accelerated
earnings BEFORE their most impressive price moves.
Sideways percentage earnings growth (even if theyre
good) or decelerating percentage earnings growth
(strong or not) can potentially signal a period of
consolidation (or slowdown) which in turn can flatten
out prices or send them lower.
But increasing percentage earnings growth
(consistently improving from the companys prior
percentage of earnings growth) can often be the
difference between good stocks and great stocks.
The Parameters
In this screen, Im focusing on increasing percentage earnings growth and projected percentage earnings
growth.
I want the last two Quarter over Quarter % EPS Growth periods to be greater than the previous periods and the
next projected Quarter over Quarter % EPS Growth period to be greater than the previous period as well.
(For quick reference, Q(0) = the completed quarter. Q(-1) = the quarter prior to the completed
quarter. Q(-2) = 2 quarters prior to the completed quarter, and etc. Moving in the other direction,
Q(1) = the current quarter. Q(2) = the next quarter, and etc.)
So the parameters would look like this;
% Change Actual EPS (Q0)/(Q-1) > % Change Actual EPS (Q-1)/(Q-2)
% Change Actual EPS (Q-1)/(Q-2) > % Change Actual EPS (Q-2)/(Q-3)
Estimated EPS Growth (Q1)/(Q0) > % Change Actual EPS (Q0)/(Q-1)
(In the Research Wizard, you can compare an item to another item. This is done in the Calculation Expression
feature, and how the above customized items were created.)
In addition to that, Im only including stocks greater than or equal to $5 with an average daily share volume of
50,000 or more;
Price >= $5
Average 20 Day Share Volume >= 50,000

Studies have shown that

almost all of the most
successful stocks in the past
had displayed accelerated
earnings BEFORE their most
impressive price moves.
25 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Earnings Acceleration continued...
Get It
This screen is available in the Research Wizard by going to:
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: sow_earnings acceleration
So when a company reports earnings, take a look at their numbers, and then see how their current % EPS
Growth stacks up to their previous periods % EPS Growth (and the period prior to that). Moreover, take a look
at how their Earnings Growth is forecasted in their next reporting period as well.
Again, with statistics suggesting that accelerated earnings typically appear BEFORE the most impressive price
moves of a stock, this is definitely a screen worth watching, for finding winning stocks on the move!!!
More on Earnings (Increasing Earnings)
Another screen I like running in regard to earnings, deals with Increasing Earnings (if not necessarily the
percentage of EPS Growth).
Parameters
In this one Im looking for each of last 8 quarters of Earnings (thats right -- 8 quarters) to be greater than the
previous quarters Earnings.
Quarterly EPS Q(0) > Quarterly EPS Q(-1)
Quarterly EPS Q(-1) > Quarterly EPS Q(-2)
Quarterly EPS Q(-2) > Quarterly EPS Q(-3)
Quarterly EPS Q(-3) > Quarterly EPS Q(-4)
Quarterly EPS Q(-4) > Quarterly EPS Q(-5)
Quarterly EPS Q(-5) > Quarterly EPS Q(-6)
Quarterly EPS Q(-6) > Quarterly EPS Q(-7)
Quarterly EPS Q(-7) > Quarterly EPS Q(-8)
(These were also created with the Research Wizards Calculation Expression feature.)
This one isnt so much concerned with the increasing percentage of earnings growth, but rather, the simple
increase of earnings from one period to another.
The thing that makes this screen so tough (i.e., a great stock picker), is that it demands a consistency of
excellent performance from a company over an extended period of time.
I also want this quarters Estimated Earnings to be greater than last quarters Actual Earnings and the next
quarters Estimated Earnings to be greater than this quarters Estimated Earnings.

26 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW

Earnings Acceleration continued...
Q(1) Consensus Estimate > Q(0) Actual Quarterly EPS
Q(2) Consensus Estimate > Q(1) Consensus Estimate
(Calculated Expressions.)
And price and volume are;
Price >= $5
Average 20 Day Share Volume >= 50,000
The report thats generated will have the last 9 quarters
lined up in the row, from left to right. It also continues
out with the current quarters estimated EPS (Q1) and
the next quarters estimated EPS (Q2). (You can also
arrange the columns in any order you wish, so you can see the data you want in any way you want it.)
(With this display, you can instantly see the EPS trend.)
And if you have access to the Integrated Company Analysis (ICA) program, you can right click on the stock and
pull up a price chart with an EPS overlay. (See below.)

The thing that makes this

screen so tough (i.e., a great
stock picker), is that it
demands a consistency of
excellent performance
27 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Earnings Acceleration continued...
The solid red line plots the actual EPS and the dotted red line plots the Estimated Earnings Consensus. I also
included the stocks EPS Surprise history in this screen shot too. The green and red arrows show positive or
negative surprises.
This strategy will typically find stocks with fantastic earnings consistency and a history of beating earnings
expectations.
Get It
This screen is also available in the Research Wizard. Go to;
Screen on your Menu Bar
Open Screen Definition
Double-click the SoW folder
Select the file: sow_increasing earnings

6
Chapter
28 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Big Money
Ive published this screen numerous times in our Screen of the Week articles and this is one of my favorite
strategies. In fact, if youre looking for a strategy that holds only a few stocks in your portfolio at any time (3),
this IS my favorite strategy.
Its a price momentum screen that finds stocks on the move. (In fact, it looks for the top % price gainers over
the last 1 month, 3 month and 6 month periods.) But with a Price to Sales ratio added to it, these movers are
still considered bargains.

Parameters
Price/Sales ratio < .5
(A low price to sales ratio is typically considered a good bargain, since the investor is paying less for each
unit of sales. Another way of looking at this is that a Price/Sales Ratio of 1 means the investor is paying
$1 for each $1 in sales. A P/S ratio of .5 means youre paying 50 cents for each $1 in sales.)
Average Broker Rating < 2
(Strong Buys and varying degrees of average Strong Buys. Since brokers ratings are typically skewed to
the bullish side, I want to make sure my picks have strong outlooks from the analysts covering them.)
Average 20 Day Share Volume >= 50,000
(Average daily trading volume (over the last 20 days) is a minimum of 50,000 shares or greater. It has to
be tradable.)
% Change in Price over 24 weeks -- Top# 20
(Looking for the top 20 price performers out of the list of stocks that qualify the above parameters.)
% Change in Price over 12 weeks Top# 10
(Looking for the top 10 price performers out of the list of the 20 best from above.)
% Change in Price over 4 weeks Top# 3
(Looking for the 3 best price gainers out of that top 10 list from above.)
Results
(using a 4 week rebalancing period)
I ran a series of tests over each of the last 5 years (2001 thru 2005). I rebalanced the portfolio every four
weeks and started each run on different start dates so each test would be rebalanced over a different set of
four-week periods. This exercise was done to eliminate coincidence and verify robustness.
Over the last five years, this strategy has shown an average annualized gross return of 138.4% a year, with
an average win ratio of 69%. And as the last parameter of the screen dictates, it always only picks 3 stocks
each period.

29 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW

Big Money continued...
In 2001, the average annualized gross return was 128.9%, with an average win ratio of 64%.
In 2002, the average annualized gross return was
262.1%, with an average win ratio of 70%.
In 2003, the average annualized gross return was
205.5%, with an average win ratio of 79%.
In 2004, the average annualized gross return was
56.3%, with an average win ratio of 65%.
And in 2005, the average annualized gross return was
39%, with an average win ratio of 67%.
(The below chart shows the average annualized returns.)
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
Results
(using a 1 week rebalancing period)
This screen also tests well with one-week holding periods too.
Over the last five years
(2001 thru 2005),
this strategy has shown
an average annualized gross
return of 138.4%* a year
* (4 week rebalancing period)
30 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Big Money continued...
In fact, since its essentially a price momentum screen, running it on a weekly basis lets you keep closer tabs
on the price changes of the leaders.
Over the last five years (2001 thru 2005), this strategy has shown an annualized gross return of 145.2% a
year, with an average win ratio of 61%. And again, whether its rebalanced every 4 weeks or 1 week, it always
only picks 3 stocks per period.
In 2001, the annualized gross return was 76.1% with a win ratio of 61%.
In 2002, it was up 263.7%, with a win ratio of 58%.
In 2003, it was up 154.2%, also with a win ratio of and
62%.
In 2004, it was up 127.5%, also with a win ratio of and
63%.
And in 2005, it was up 37.4%, with a 60% win ratio.
(The below chart shows the annualized returns.)
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
Using a 1 week rebalancing
period over the last five years
(2001 thru 2005), this strategy
has shown an annualized gross
return of 145.2% a year
31 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Big Money continued...
An interesting observation to note, is that the 4-week outperformed the 1-week in 2001, and the 1-week beat
the 4-week in 2004. But both approaches did equally as spectacular in 2002. And they both performed pretty
spectaculary in 2003 as well.
But no matter how it was traded or when, all of the returns were quite impressive to stay the least.
Alternative Trading Methods
Even though I originally designed this screen to be rebalanced every 4-weeks, it also tests well on a 1-week
basis too (as you can see).
I have also found that using a combination of both
(4-week and 1-week holding periods) can also be very
effective as well.
This is done by running the screen on a weekly basis,
but holding the stocks for a 4-week period. In other
words, youd run the screen at the beginning of the
week, buy all 3 stocks, and hang on to them for the next
4 weeks. Then, next week, youd run the screen again,
picking up any new stocks that made it thru, and being
sure to hang on to those for the next 4 weeks. And you
would do this week after week.
Eventually, after each 4-week period is up, youd run your screen again and see if those original stocks still
qualify -- keeping the ones that did, selling the ones that didnt, and buying the new ones that now do.
So even though a 4-week holding period is being used, the screen is being run weekly and only a portion of
the portfolio is being rebalanced each week.
Get It
The Big Money screen can be found in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: bt_sow big money
using a combination of both
(4-week and 1-week
holding periods), can also
be very effective too. Running
the screen every week,
but holding onto the stocks
for a 4-week period.
7
Chapter
32 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Increasing P/Es for Stocks on the Move
Studies have shown that many of the best stocks, year after year, over the past decade, saw their P/E ratios
increase by more than 100% from their breakout point. (This is also evident (increasing P/Es) in the statistics
presented in Chapter 2s Magic Numbers? and Relative Valuations.)
A good screening tool can help you find these stocks early in their breakout cycle before they really take off.
Heres an example to explain this scenario. Say a stock
price is at $20 and its earnings over the last four
quarters = $1 per share. Then its P/E ratio will be 20.
($20 divided by $1 = 20)
If the earnings rise (to $1.25 for instance) and the
stock doesnt, the P/E ratio will fall. ($20 divided by
$1.25 = 16)
If Earnings go up but Prices dont, the P/E ratio will decrease. (But typically, as earnings increase,
so should prices.)
If the stock rises and its earnings stay the same, the P/E ratio will increase. So if the stock is now at $30 and
its earnings remain the same at $1, ... the P/E ratio will have increased to 30 as well. ($30 divided by $1 =
30)
If Prices go up but Earnings dont, the P/E ratio will increase. (But this scenario is probably
short-lived because the demand for a stock (prices) only goes up when earnings are going up, or at
least expected to.)
But now lets say next quarter earnings come out and its four-quarter combined numbers show the EPS at
$1.25 with the stock increasing to $30. The P/E ratio will now be 24. ($30 divided by $1.25 = 24 -- a 20%
increase in its P/E ratio from the first example.)
If Earnings go up and Prices go up too, the P/E ratio will also increase. (The interesting dynamic
is that as earnings increase, so should prices. And as forecasts for continued earnings arrive, the
demand for the stock should continue to send prices even higher. This type of scenario (higher
earnings and higher prices) has longevity and is common in most trends.)
This increase in price and earnings is an ideal way to
spot stocks in favor and that are anticipated to continue
to trend higher. And instead of looking for nominal P/E
changes, screen for P/E increases in excess of 20%
which should provide the greatest upside potential.
Just like a 20% increase in the price of a stock can
alert you to a new potential uptrend, you can also use a
20% increase in the P/E ratio to alert you to potentially
significant price and earnings events too.
Studies have shown
many of the best stock saw their
P/E ratios increase by over
100% from their breakout point.
A meaningful increase in a
stocks P/E can potentially alert
you to a significant price
and earnings event.
33 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Increasing P/Es for Stocks on the Move continued...
Parameters
When I first wrote about this in Jan. of 2003 in our Screen of the Week series, the Screen I was using was:
P/E(Recent) >= 1.2* P/E(Recent-3M)
Which means current P/Es that are 20% or higher than their P/Es from 3 months ago.
P/E(Recent) < 2* P/E(Recent-3M)
I also included a limiting expression which made sure the current P/Es were not greater than 100%
higher.
(Its important to note that most screeners dont have historical P/E ratios (or other historical measures
for that matter), but the Research Wizard does. And with it, you can then compare an items value to its
value from a different time period. This is done in the Calculation Expression feature, and how the above
items were created.)
The stocks also had to be at least $10 or higher with an average daily trading volume of 100,000 shares or
more.
Price >= $10
Average 20 Day Share Volume >= 100,000
In future articles, I added some additional filters to the screen, ensuring that all of the stocks saw an increase
in their earnings. (If prices are rising without an increase in earnings too, theres no real reason for prices to
continue to rise.)
EPS Q(0) > EPS Q(-1)
I wanted to see an increase in the most recently completed Quarters Earnings over the previous.
EPS Q(-1) > EPS Q(-2)
And an increase in the Quarter-over-Quarters Earnings before that.
And for good measure, I wanted;
EPS F(0) > EPS F(-1)
The most recently completed Annual Earnings to be greater than the Previous Years Earnings.
EPS F(1) > EPS F(0)
Along with projections for This Years Earnings to be greater than Last Years.
I also dropped the minimum price down to $5 or greater, but kept the average 20 day share volume at
100,000.
Price >= $5
Average 20 Day Share Volume >= 100,000
Use this screening strategy alone or with other criteria to help spot winning stocks
BEFORE they become big winners!

34 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW

Increasing P/Es for Stocks on the Move continued...
Get It
This screen is available in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder
Selecting the file: sow_increasing pe
8
Chapter
35 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Cheap Stocks and Big Returns (a.k.a. Breakouts)
This screen is also one of my favorites and one that I wrote back in 2001.
The premise behind this screen is to try and find cheap stocks (stocks at or less than $15) that are trading at
(preferably, consolidating at) just under their 52 week high, in an effort to get on board before they break-out
to new highs.
In other words, I want the stocks to be near their highs,
but most of all, Im looking for stocks that have been
turned back from their recent highs, ... have consolidated
their advances on their trek back to those highs, ... and
are just now starting to make a run at their 52 week
high again.
Im a big fan of getting into basing patterns (relatively
narrow trading ranges) after an uptrend has been
established. Especially near recent price highs, since stocks making new highs tend to make even higher
highs.
And while its true, some of the stocks that make it thru this screen have already broken-out, I have found this
screen to be an ideal strategy for finding low priced stocks with a high probability of success.
Parameters
Price <= $15
(Cheap stocks.)
Zacks Rank = 1
(This helps solidify the current earnings picture.)
Current Price/52 Week High >= .90
(They have to be trading within 10% of the 52 Week high.)
% Change in Price over 4 Weeks >= 10% but not more than 20%
(Im 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 peoples attention while follow-thru at 20%, typically signals the
beginning of a trend or a breakout, I wanted to be alerted BEFORE a breakout was seen.)
Beta <= 2
(Active stocks are good, but wildly volatile ones are not.)

I have found this screen to be

an ideal strategy for finding
low priced stocks with a high
probability of success.
36 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Cheap Stocks and Big Returns continued...
The Results
I ran a series of tests over each of the last 5 years (2001 thru 2005). I rebalanced the portfolio every four
weeks and started each run on different start dates so each test would be rebalanced over a different set of
four-week periods. This exercise was done to eliminate coincidence and verify robustness.
Over the last five years, this strategy has shown an average annualized gross return of 66.2% with an average
win ratio 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 55.7%, with an average win ratio of 70%.
In 2002, the average annualized gross return was 24.4%, with an average win ratio of 66%.
In 2003, the average annualized gross return was 139.2%, with an average win ratio of 88%!
In 2004, the average annualized gross return was 75.4%, with an average win ratio of 75%.
And in 2005, the average annualized gross return was 36.3% with a 69% win ratio.
(The below chart shows the average annualized returns.)
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
37 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Cheap Stocks and Big Returns continued...
This strategy also tested well using a one-week rebalancing period. In fact it does fantastic.
Get It
This screen is available in the Research Wizard by going to;
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder (SoW -- short for Screen of the Week)
Selecting the file: bt_sow breakouts
Additional Comments
Note: Even though this screen will generally produce on average of 4-5 stocks per period, there will be times
where literally no stocks will qualify due to the narrowness of the parameters.
Tip #1: But since this screen has such a great track record and such a high success rate, if nothing comes
through on my first pass, Ill run it day after day, until the screen spots something.
(This is one of the reasons why I run so many backtests using different start dates in my analysis. I want to
make sure that the strategy has a history of picking good stocks at any time as opposed to just the 3rd
Monday of every other month, etc.)
Tip #2: Even though I designed this screen to be a trading strategy, Ill also use this screen as a high probability
stock picker for when I have excess cash on the side and Im looking for cheap stocks to add to my portfolio.
9
Chapter
38 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Something for Everyone: Growth AND Value
This screen has something for both Growth Investors and Value Investors alike.
Growth Investors focus on companies with great earnings growth, but that alone isnt good enough for
many stock pickers any more. They want good growth at reasonable prices (low P/Es).
And while Value Investors tend to focus on low P/E stocks, too many are low because they lack earnings
power.
So instead, try combining the best of both worlds and
focus on the companies with the highest growth rates
with the lowest P/E ratios. (If youre looking for stocks
with low P/Es, this is one of the best ways to find
them.)
And this screen does just that.
Parameters
Companies with 5 Yr. Historical Growth Rates to be in the top 20 percentile of all companies.
(Using a Uniform Rank of 1-99 (99 being the best growth rates), I screened for stocks ranked 80 or better,
meaning they had growth rates better than 80% of all of the other companies.)
Companies that also happened to have the lowest P/Es too lower than 80% of all other companies.
(Using a Uniform Rank of 1-99 again (this time 99 having the lowest P/Es), I screened for stocks ranked
80 or better, meaning, they also had lower P/Es than 80% of all of the other companies too.)
Price >= $5
(I then required those qualified stocks to be trading at or above $5.)
Average 20 Day Share Volume >= 100,000
(With average daily trading volumes of 100,000 shares or more.)
Zacks Rank <= 2
(And a Zacks Rank of 2 or less. Only buys and strong buys allowed).
Results
I ran a series of tests over each of the last 5 years (2001 thru 2005). I rebalanced the portfolio every four
weeks and started each run on different start dates so each test would be rebalanced over a different set of
four-week periods. This exercise was done to eliminate coincidence and verify robustness.
Over the last five years, this strategy has shown an average annualized gross return of 41.4% with an average
win ratio of 65%. This strategy holds on average of 15-16 stocks in its portfolio each month.

the best of both worlds

focus on companies with the
highest growth rates with
the lowest P/E ratios.
39 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Something for Everyone: Growth AND Value continued...
In 2001, it showed an average annualized gross return of 42.6% with an average win ratio of 60%.
In 2002, the average annualized gross return was 19.1% with a 69% win ratio.

In 2003, it produced an average annualized gross return of 89.9% with a 71% win ratio.
In 2004, it produced an average annualized gross return of 43.8% with a 71% win ratio.
And in 2005, it showed an average annualized gross return of 11.7%, and a 53% win ratio.
(The below chart shows the average annualized returns.)

Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
For further illustration, I ran a continuous backtest starting on 1/5/2001 and going thru 12/30/2005,
rebalancing the portfolio every four weeks. I also compounded the periods returns to greater simulate how one
would invest. (For example; if the starting equity was $10,000 and the first periods gain was -10% ($1,000),
the next period would invest with $11,000. If that next period is off 10% (-$1,100), the period after that would
invest $9,900.) Also, in this illustration (and all of the other performance illustrations), no consideration has
been given to commission costs, slippage or any other real-world constraints.
The total return from its start date on 1/5/01 thru 12/30/05, is 326.1%, with a win ratio of 65% (42 winning
periods out of 65 total periods). A hypothetical $10,000 invested (sans commissions and etc.), increased to
$42,611.65. By comparison, a hypothetical $10,000 invested in the S&P during that time would have increased
to only $10,281.83.
40 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Something for Everyone: Growth AND Value continued...
(The below chart shows the compounded returns from 1/5/2001 thru 12/30/2005.)

(1/5/2001 thru 12/30/2005) Growth and Value Strategy S&P 500
Compounded Total Return 326.1% 2.8%
$10,000 Starting Equity $42,611.65 $10,281.83
Win Ratio 65% (42/65)
Avg. # of Stocks Held 15-16
Hypothetical returns may not always be duplicated in the real world. There is risk of loss in trading stocks.
This is an excellent Growth and Value screening strategy to find new stocks with a high probability of success.
Check it out for yourself.
Get It
This screen can be found in the Research Wizard by going to:
Screen on your Menu Bar
Open Screen Definition
Double-clicking the SoW folder (SoW -- short for Screen of the Week)
Selecting the file: sow_growth and value
41 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Something for Everyone: Growth AND Value continued...
More on Ranking
The above screen uses just one of the more sophisticated Ranking methods found in the Research Wizard.
(Uniform, along with Histogram and Ordinal -- the definitions of which and how they can be used are found in
the User Manual.)
But theres also a quick and easy visual Ranking method too that can be instantly applied to any screen, report
or portfolio with a click of the mouse. And its called Hot Maps.
Check and see where your stocks rank out of all of the other stocks out there, as well as in your own portfolio
and screens.
See which stocks are the best and which stocks are the worst as the Hot Maps graph, assigns varying shades
of green for most attractive and varying shades of red for least attractive. (See below.)
(Screen shot example of a Hot Maps graph.)

The program also lets you chart your own regression analysis studies too with its Scatter Plot feature. (See
below.)
42 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Something for Everyone: Growth AND Value continued...
(Screen shot example of a Scatter Plot chart.)

Very cool.
10
Chapter
43 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Creating a Custom Consensus of Your
Winningest Screens (a Screen of Screens)
Every week in our Screen of the Week articles, I either go over a unique way to screen for stocks or Ill publish
a proven, profitable screening/trading strategy.
For many customers interested in these trading
strategies, the only real decision is in choosing which
one(s) to use.
Well instead of choosing just one, why not look at them
all and create a Custom Consensus of some of our (or
your) winningest strategies.
This too is a great strategy for picking winning stocks
from many diversified approaches. Because aside from using the Zacks Rank for many of the screens, there
are many other filters layered on top in each screen to find the best stocks from different styles.
Most of the screens Im currently using in my Custom Consensus strategy are outlined in this book. But they
all come loaded with the Research Wizard program. (And many more.)
The Consensus Strategy Screens are:

Big Money (Chapter 6)

Breakouts (Chapter 8)

EPS Growth / Past and Present (Chapter 11)

Filtered Zacks Rank (Chapter 3)

Increasing Cash Flows (Chapter 11)

PEG (SoW version not in book, but in program)

ROE (Chapter 4)

Sales and Margins (Chapter 11)

Upgrades and Revisions2 (Chapter 1)

Instead of choosing just one
strategy ... create a
Custom Consensus of our
(or your) winningest strategies.
44 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Creating a Custom Consensus of Your Winningest Screens continued...
The concept of the Consensus strategy is pretty easy but very powerful!
Simply run each screen (generating a list of qualified tickers for each screen), and then count how many times
a stock appears in all of those screens.
If it appears two or more times (i.e., shows up on two or
more screens), its now a part of the Consensus strategy/
portfolio.
Ill typically create this list old school style, by simply
writing down the stocks that qualify each screen on a piece
of paper and then circling the duplicates. Then Ill simply list
those dupes and voila. (See the image to the right.)
(And while its not necessary to construct the portfolio, Ill
usually tally up the total number of times they came thru. But
no one stock should get a greater weighting than another
just because it came thru more times. They should all be
invested in equally.)
Now even though two passes is enough for me, I have
some customers who run additional screens and require a
stock to qualify three or four times before its considered a
Consensus pick. (But again, two passes is fine for me.)
By the way, you dont have to write your lists down on a
piece of paper if you dont want to. You can also export your
stock lists to Excel (or Notepad or Wordpad, etc.), and tally
it all up in there.
Moreover, with the Research Wizards Scripting feature,
you can run as many screens as you want and have each list
automatically exported to an Excel file or even printed, all
with one simple click of a button.
Additional Comments
As you can see, its quite easy to create a Custom Consensus or a Screen of Screens strategy. You just have
to make sure youve got good strategies to draw from.
And while thats no guarantee of success; strategies that have proven to pick good stocks in both up markets
and down markets in the past, should continue to do so in the present and in the future.
11
Chapter
45 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
More Winning Strategies
As I mentioned in the Introduction, we purposefully narrowed this books focus down to only ten of our best
screening strategies. But since there are many other strategies that have done quite well and that are popular
with our users, I decided to list some of them along with their historical performance.
PEG
(earnings growth at reasonable prices)
Get It
file: peg
46 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
More Winning Strategies continued...
Increasing Cash Flows
(cash is king)

Get It
file: bt_sow_increasing cash flows
47 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
More Winning Strategies continued...
EPS Growth, Past and Present
(strong earnings growth, then and now)

Get It
file: bt_sow_eps growth past and present
48 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
More Winning Strategies continued...
Winning Ways
(winning ideas from some of our winnigest strategies)

Get It
file: bt_sow_winning ways
49 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
More Winning Strategies continued...
Sales and Margins
(money made and money kept)

Get It
file: bt_sow_sales and margins
12
Chapter
50 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
Additional Comments on Backtesting
A question that I get asked a lot in regard to the backtester is;
Will it also show the stocks that came thru in each historical period?
The answer is;
Yes!
Itll give you a summary of your strategys performance, and also break it down period by period. (See below.)
(Screen shot of Backtest report.)


And you can also see what stocks came thru in those periods too within the Backtest Details window. You
can take a look at one period at a time or all of the periods at once. It even shows you how many times a stock
qualified your screen and how many consecutive periods it qualified too so you can determine its turnover.
(See the next page for a screenshot of the Backtest Details window.)
51 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
(Screen shot of the Backtest Details window.)

And all of the Backtest reports (along with any other data), can be easily exported to Excel with a click of a
button.
And of course, once in Excel, you can then do even more analysis including creating charts and graphs like
youve seen throughout this book.
52 Zacks Investment Research, Inc. Research Wizard www.zacks.com/RW
You Can Do It!
Try out some of our strategies for yourself or create your own.
And then backtest them to see how good they are.
Find the right stocks with right tools and START MAKING
BETTER DECISIONS TODAY!
You can do it!