Before running your first scan, consider how the size of your trading account will affect your watch list. In most cases, available capital will rule out trades over or under certain price levels. For example, many retail traders avoid illiquid stocks over $50 per share, while heavy hitters may find their best opportunities in the highest-priced equities.

Full-time traders should prepare a range of scans that target particular plays or patterns. For example, having an updated momentum list ready to go when speculation hits the markets can mean the difference between profit and loss. Alternatively, stocks that wobble back and forth between well-defined price levels make great trading vehicles in quiet times.

The Nasdaq's popularity for short-term execution makes it the preferred choice for daytraders, but swing traders may find that the NYSE offers better profit opportunities with less risk. In any case, it's best to stick with liquid stocks at all times, regardless of which exchange you're trading.

Your scan output provides the raw material for your premarket examination. The size of your watch lists should match the time you've set aside for trading and technical analysis. If you can't manage a lot of information, prepare a small list of 20 to 50 stocks, and follow these exclusively. Make small adjustments to the list each evening, but keep a core group you can track for months at a time.

Your scanning routine needs to identify setups, measure reward/risk ratios and find the most advantageous execution prices. Build an automatic filtering process that evaluates each stock objectively for these three characteristics. Remember that a single flaw that doesn't fit your trading style may negate an otherwise perfect opportunity.

For example, a triangle pattern that shows multiyear resistance just overhead should be removed without hesitation. Likewise, the best positions will come when the scan output fires on all cylinders. In other words, price, volume and technical indicators all line up and tell you it's time to act.

Understand the real purpose of market scanning. Traders spend long hours trying to build scans that spit out perfect gems and no losers. This strange obsession actually undermines their considerable efforts. Scans are wake-up calls for your watch list -- nothing more and nothing less.

The best scans uncover patterns that need to mature before it's time to enter the market. In fact, a good scanning routine usually increases the trader's workload, rather than decreasing it. The real effort comes when stalking the watch list on a daily basis, waiting for price movement to signal the perfect time to act.

Let's examine one of my favorite scans and offer step-by-step instructions so you can use them to find your own trading gems.

Some traders think scanning is a one-step process that should highlight a few stocks they can play immediately. Unfortunately, this reflects a lazy attitude that expects the scanner to do the analytical legwork for them. In reality, the best scans will increase the trader's workload, not decrease it.

Avoid the mistake of adding too many criteria into the database search. Your job isn't to find a perfect needle in the haystack, because the barn floor is already littered with sewing kits. Just bend over, pick one up and get to work.

There are three distinct elements to effective market scanning. Each step builds on the prior one as it filters stocks that shouldn't make the final cut. It's important not to skip steps because of the logical way the workflow builds layers of restrictive filters. Let's look at each step in detail.

1. What to Look For

Boolean logic, or equivalent math, comprises the first step by telling the scan what types of things to look for. This is where you choose a subset of common functions such as price, moving averages or indicator values. Once you've chosen these functions, combine them with math operators (+, -, *, /) and tie them together with logical operators (> , < , =).

Your formula needn't be complicated to be effective. In fact, the most potent scan statements are also the simplest ones to write. The previous illustration shows the language for my daily range scan. "H - L" measures the daily range by subtracting the high from the low of that session. The scan takes this number and filters out all instances where it's less than or equal to 10% of the closing price.

A stock needs a volatile day to meet the requirements of this scan. For example, a $20 stock must move through a daily range in excess of $2 to yield a scan hit. One obvious problem in constructing this scan is that low-priced stocks move through wider ranges than high-priced stocks. Fortunately, the next building block addresses this anomaly.

2. Filtering

The second step takes the initial output and filters it through chosen exchanges and liquidity levels. This is also the time when fundamental and technical analysis can be added into the mix. For example, Worden's TC2000 enables traders to filter-scan output through market capitalization, price-to-earnings ratio or earnings per share.

Notice how the initial logic is now tied to a database subset called "All Stocks." This grouping removes mutual funds, derivatives, industry averages and indices from the search. The volume notation further limits the study to the top 40% in market liquidity as determined by average volume. Finally, this step removes all issues trading below $5 per share.

3. Sorting

The third and final step seems to give traders the most trouble. By this time, the scanning process has already produced a list of stocks that meet the narrow criteria of the first two steps. The final phase takes this culled list and sorts it in a way that lets the best trading opportunities "float" to the top.

This is very important, because some scans will yield hundreds of potential candidates. Realize that this overabundance doesn't indicate the need to apply more restrictive criteria. In fact, traders should keep their scans loose and inclusive at all times. The final step addresses this surplus through common-sense sorting techniques.

My favorite method is to sort the candidates by relative strength attributes. TC2000 calculates this in several ways, including RSI and comparative performance to the SP 500.

After years of study, I've found that the best results come from sorting the price vs. 200-day moving average column. This represents percentage distance above or below this long-term average.

Note how the list can be resorted quickly by market cap, price percentage change, or even three-month trading range. This lets traders view data output in a variety of ways. Even so, there's one more thing they need to do: examine the price chart. As simple as it sounds, market scans do absolutely no good unless there's a well-organized pattern to trade.