Elements of a Good Swing

In my previous post, Anonymous learner asked about my criteria for a good swing trade.

"You mentioned you couldn't find any decent swing setup. How do you usually look for such swing setup? And in your opinion, what kind of criteria fulfilled would you consider it to be decent? Thx Learner."

After studying some of the charts of the various trades that I have made recently, I can't say that I have a bulletproof, repeatable method for swing trading.
I started out "re-learning" how to swing trade by studying the examples of those traders who do it better than me. The Market Speculator, Dummy Spot, Dave Landry, AlphaTrends, RealTimeTrader, and TAZ Swing Trader, these were my "mentors" that I read (and still do read) regularly (links at the right sidebar). Please note that even within swing trading everyone has their own style. For instance, AlphaTrend's style is different from Dave Landry's style, and neither is better than the other. Some traders like to focus and follow one specific style, but as you might have noticed by now, I like an open, flexible, style. I would study the charts they provided, or the trades that they listed, and internalize what made their trades work out well. So now, when I look for a good swing setup, I kind of use the following routine:

1. Begin with the End in Mind.
This habit of highly effective people sounds a bit hokey for trading, but once you think about it, it really makes sense. Good trading begins with good habits. Have a pre-defined set of trading rules. (How successfully you abide by them is a totally separate issue). Make a log of all (or at least most) trades for later review. Plan and prepare ahead of time for each morning's trading session.

2. Look at multiple sources for the potential swing candidates.
Daytrading candidates also make for potentially good swing candidates, so I don't discriminate.
For me, I use the HCPG newsletter, Richard's new 3o-day high list, Alan Farley's picks, Brian Shannon's nightly videos, my previous trades (which I typically keep on my watchlist for a few days anyways), my own custom-cooked up new 52-wk highs scan from msn moneyCentral, and some microcap picks from stockChart.com's high volume gainer scan. If you surf through enough stock trading blogs, you will find plenty of ideas for swing trading candidates. More often than not, I am not able to go through all of my sources. I usually wind up with too many candidates and I have to start culling the list. I cull the list by looking at RSI, MACD, MA trendlines, chart patterns, and the "cleanliness" of the daily chart. Time permitting, I also look at the 10day hourly chart.

3. Evaluate the setup
About a half hour before market open (if I can wake up that early!) or the night before, I usually enter the stocks into my trading account's watchlist, and start pulling up charts for most of the stocks in my watchlist. I look at support and resistance levels. The videos from AlphaTrends do a pretty darned good job of educating guys like me on how/where to draw the support/resistance levels. After watching those videos a few times, I got the hang of it, and after that, I started coming up with the support/resistance levels without using the video. Once I have the resistance/support levels, I enter it in the form of an alert in my watchlist of stocks. Then I just sit back around and wait for the alert to go off (usually I pre-occupy myself with taking a nap, eating, or surfing, or once my wife made me do some house chores......). Often I will bring up onto my screen the real-time streaming charts of some of the stocks on my watchlist, esp. the ones where I know I should be stalking the stock.

4. The Key Decisions
If an alert goes off, I have to do the following quick:
a) pull up a chart of the stock in question (if it is not already up)
b) look at the behaviour prior to setting off the alert - was it clean or choppy?
c) look at the overall directional movement - price spike, breakout, fakeout, general trend up, or something else?
The key here is to understand what the chart is telling me that it is doing, instead of me trying to predict what the chart will do.
d) does it match one of the high probability patterns ?
e) is there volume?
f) where is the stop loss going to be?

I used to work out the stop loss levels on a spreadsheet prior to taking each trade, but I found that to be a bit cumbersome and slowing me down, so now I make a mental stop. This of course, has its drawbacks, so I will be going back to entering in a hard stop until I have the mental part of my trading back to a level that I expect of myself.


So, that covers most of the basics of how I find my swing setups.
Now, to the other part of your question - what makes a stock a good swing trading candidate? I'm probably far from being an expert in this area, but here's my thought process:

1. Market environment. Though there are exceptions, but in general, I look for long candidates in an uptrending market/sector, and short candidates in a downtrending market/sector. Now that we are in a trendless market, the charts better good, with a capital G. Clean, not choppy, no fake-outs, mini-spikes that reverse, etc.

2. Volume. Without volume, most of the high probability chart patterns become higher risk plays. The presence of volume more often than not leads to continuation moves.

3. Clear and definitive Trend.
You don't want to select a stock whose 10/20/50d MA's are crossing each other all over the place. Once a stock is clearly moving in a certain direction, it will tend to continue on in that direction. Trendlines help to identify where the buyers are hiding, and where the stock could be headed.

4. Past Experience.
Experience is the best teacher. Once I have made enough profitable swing trades, I learned to recognize what made it a success. On the flip side, once I have made enough unprofitable trades, I learned what made it a failure and what to avoid in the future.

I will create a new category, "swing", to identify my true swing trades (as opposed to hourly scalps that I have no intention of holding for longer than a day or two).