In a recent article, I answered the question that writers for this site are frequently asked: Which technical analysis tools do I prefer? Preferences differ among writers and subscribers, but one of my choices always seems to draw either extreme interest or extreme ire. That choice is the nested Keltner channels. These channels admittedly look as if I cooked up a batch of tri-colored pasta and flung it against the screen.

I understand the mixed responses, especially since so few traders use this tool. Other than the original article writers who introduced me to the topic, I know of only one other person who has used these Keltner channels nested in this way. However, I am forever grateful to the two writers who years ago introduced me to nested Keltner channels. I have two primary reasons for sticking with these technical analysis tools through the years. They allow me to make some initial judgments about whether support or resistance looks stronger, and they allow me to set potential price targets if that resistance or support is broken.

After I received a predictable and understandable "I don't like your charts" response from my last article about the technical analysis tools I use, I was reminded that it's perhaps been a while since I've discussed these powerful technical analysis tools. We've had a lot of new subscribers since that time. I thought some further explanation might be in order. Some of this will be a repeat to long-time subscribers. However, I'll review charts that were current at the time the article was written, so the updated potential price targets might prove interesting even if the discussion is old hat.

Keltner channels were introduced by Charles Keltner. Keltner used these channels, based on central moving averages, to identify breakout moves. The theory with Keltner channels is the same as with Bollinger bands and other channeling system. Most of the time, prices are likely to travel inside the channel. When they break out, they're often propelled by strong momentum. Breakout trades can be highly profitable if one can stomach the frequent draw downs that are also part of breakout trading systems. I can't. They're not for me. The occasional big winner in no way makes up for the zillions--it seems--of small losses, even if the system makes money in the long run.

However, just because I don't use these channels the way Keltner used them doesn't mean that they're not helpful. Take the propensity for prices to travel within a channel, for example.

Annotated 15-Minute Chart of the OEX:

Different underlyings or trading styles might require some fiddling with the interval length (10-minute versus 15-minute, for example) to find the appropriate channel that works best for the underlying you're watching. Sometimes, as support or resistance is approached, the various channel lines or central averages converge, showing that support or resistance might be massing. That allows a judgment about whether support or resistance might be stronger. Remember when viewing the daily chart that it was snapped a week before publication.

Annotated Daily Chart of the OEX:

When prices maintain closes outside the black-channel or any of the channels, it's then possible to set a potential price target at the outer boundary of the next wider channel.

Annotated 15-Minute Chart of the RUT:

As one can imagine, it can be of great help to see when resistance might be stronger than support or vice versa. Traders can also easily imagine how helpful it can be to set potential upside or downside targets. This is particularly true when prices are breaking to new record highs or lows or moving through price levels where no obvious support or resistance lies. Doubly helpful is the fact that the new potential target is also the place where new potential resistance or support might lie, with an emphasis on "might." A trader who has elected to enter a trade based on a breakout through one channel's boundaries might elect to take at least partial profits at the next channel's outer boundary and then tighten stops on the rest of the trade. I don't trade this way any longer because I don't trade directional trades and use these channels differently now than I did in the past.

However, as is true with all technical analysis tools, this one is far from foolproof. Although the RUT reached its upside target on the 15-minute chart and then some, that doesn't always happen, of course. As was true with the RUT, the potential resistance or support at the new target doesn't always hold up, either, especially when the channels have been evenly distributed, one within the other, all moving horizontally. A break out of one channel in those conditions is often a breakout on strong momentum. If the initial breakout is particularly strong or is sustained through several closes, the next channel's boundary can many times be exceeded, especially with the volatile RUT.

That's why I don't count on all massed resistance or support to hold or all subsequent upside or downside targets to be reached. It's the reason I don't count on all trendlines to hold or all moving averages to be retested in pullbacks. I think in terms of possibilities, vulnerabilities and probabilities, not something that can be counted on to happen. That doesn't render this technical analysis tool any less useful, in my opinion. If I know the OEX might be vulnerable to a further drop and have bull put credit spreads that might be threatened if it dropped too far, I could make appropriate plans for hedging my bull put credit spreads. If I hadn't yet put on bull put credit spreads but was considering doing so, I'd want to be sure I was putting them below that next potential downside target. However, I wouldn't buy a long put position and hold on, no matter what, until that downside target was hit because a potential target was set at a certain level. That potential target might never be hit.

I included an annotated OEX daily chart in this article. That OEX daily chart's annotations mentioned that unless the OEX could produce sustained daily closes above a certain moving average, it was susceptible to further downside. Obviously we know what has happened since that chart was snapped, but let's look at what that daily chart was showing at the time it was snapped a week ago. What were the targets the nested Keltner channels were suggesting as possibilities? First, let's address the downside possibility. If the OEX could not maintain daily closes above 495.99, more downside potential existed. We know that the OEX did break out above that potential resistance on daily closes, but if it hadn't been able to do so, one potential downside target shown on the chart was at about 483.78. What about the case that did happen, a sustained breakout above that potential resistance on daily closes that was then at 495.99? The daily chart suggested a potential upside target in the 506-507 range, but it must be remembered that these Keltner channels are dynamic. They'll move with time in the direction of the price movement, so either a sharp sudden move or a sustained slow move toward a target would nudge it in the direction of the move. In that case, we would have expected the upside target to have been pushed higher this week as prices exploded upward. In fact, that did happen. As I type on the afternoon of February 19, 2010, that potential upside target (and potential resistance on daily closes) had been nudged up to 507.59, with the OEX at 507.92 as I type. It may be a will-it-or-won't-it end to the trading day. If the OEX now can't maintain daily closes above about 506.40, then it looks vulnerable to either a pullback toward 502.32 or a sideways move until the moving average currently at that level moves up under the OEX as it consolidates. This isn't a prediction but a demonstration of how I use these channels.

Again, none of this is exact or should be expected to be exact. However, these Keltner channels allow me to think in terms of if-then statements. If this happens, then that might be the result. I can make judgments, after years of watching these, as to whether support or resistance might look stronger, and can project potential price targets and next support or resistance levels if that support or resistance is broken.

For those who are interested, I use the following moving averages and offsets for my nested Keltner channels: 9-ema with 1.4 offset, 45-ema with 3.0 offset, and 120-ema with 7.2 offset. Different charting programs use different names for "offset." These were the settings suggested by the original two authors whose work pointed me toward Keltner channels, and they're the ones I still use.