With over 10,000 stock charts to look at every day, professional analysts use sophisticated technical scanning programs to quickly find stocks with the best potential for profitable moves. Now StockCharts.com makes many of those same scans (sometimes called "screens") available to you on a daily basis. In this article, we'll discuss how scans work, how they should (and should not) be used, and the specifics of each scan that we publish.

Let's start with a definition:

By technical formula, we mean a calculation that uses a stock's price data (open, high, low, close) and/or volume data over a given period of time to compute some value. By large population we mean an easily defined collection of stocks that is too large to screen by hand. By filter criteria, we mean the range of valid values for the results of the calculation.

Now let's look at the description of a very simple stock scan:

• All NYSE stocks whose current 50-day moving average is anywhere above their current closing price.

In this example, the technical formula is the 50-day moving average formula, the large population is the set of all stocks traded on the New York Stock Exchange (NYSE), and the filter criteria is any value above the given stock's current closing price.

To "run" this scan, a computer must have at least 50 days worth of price and volume data for every stock listed on the NYSE. For every stock listed on the NYSE, it must add up the last 50 closing prices, divide the result by 50, and then compare that result to the current closing price. If the results is above that closing price, the particular stock is added to the results of the scan. If not, the stock is ignored. The computer then repeats the process for the next NYSE stock.

So how do you know if a stock scan is useful or not? There are several factors to look at:

• Simplicity
• Selectivity
• Technical Accuracy
• Philosophic Accuracy

Scans with complex technical formulas, complex filter criteria, and very large source populations can take a very long time to run. Faster computers have made this less of a problem than in the past, however – all other things being equal – simpler is better.

Selectivity refers to the number of stocks that a scan returns. Highly selective scans will often return no results – on the day the scan was run, no stocks met the scan's filter criteria. On the other hand, loosely selective scans often return more than 10 percent of the original population. That's more than 1000 stocks if the original population is "all listed US stocks". Useful scans tend to be fairly selective, but will rarely return an empty result list.

Scans are designed to find stocks that are in a specific technical situation. If a scan is supposed to return stocks that have recently made strong upward moves, but most of the stocks it returns have not done so, then the scan should probably be refined until it is more accurate.

Scans are designed with a particular investing philosophy in mind. Value investors use scans that try to find weak stocks that are showing signs of a turnaround. Growth investors use scans that try to find stocks that have recently moved higher on strong volume. Short-term traders use quick, volatile indicators in their scan's technical formulas while longer-term investors may use slower, less jittery indicators. And so on.

Philosophic Accuracy refers to a scan's ability to find stocks that meet the designer's investing philosophy. This is probably the most important factor for evaluating a scan but unfortunately, it is also the hardest to measure objectively. Each investor has to look at the results from a scan over a long period of time and determine for themselves if the majority of the stocks it returns are "useful" to them. A scan that one investor finds valuable may seem like a waste of time to another.

Before we get into the specifics of the scans that we provide, let's discuss the right way (and the wrong way) to use technical scan results.

Probably the most important thing to keep in mind is that scans find stocks that are in potentially promising positions. As with most forms of financial analysis, nothing is guaranteed. While a well-designed scan can filter out lots of stocks in unpromising positions, that doesn't mean that all of the stocks it does find will be useful to you.

Here are the steps we recommend you follow in order to get the most out of our scan results:

1. Determine your trading philosophy - Are you looking for short-term or long-term opportunities? Are you looking for strong stocks that may get stronger or weaker stocks that may be turning around? Do you care about the price of the stock? What about average daily volume? And so on…
2. Understand the scan's indicators - Are the strengths of the indicator(s) in the scan compatible with your trading philosophy? Are the time periods used to calculate the results similar to your time horizons?
3. Review charts of stocks selected by the scan - What percentage of the stocks selected by the scan have charts that look promising to you? How many false results were generated?
4. Track the results for several trading periods - How many of the "promising" stocks actually moved in profitable ways? Would any of the unprofitable stocks have caused you to lose more than you are willing to risk in the market? Did the stocks move primarily due to technical or fundamental factors?

After a while, you will begin to see patterns in the behavior of the stocks that a particular scan selects. You will develop a sense for the stocks that may repeat those patterns and that can help you trade or invest with more confidence. While risk and uncertainty can never be eliminated, taking the time to become familiar with the results of a particular scan will usually reduce both.