Option Investor
Trader's Corner

Early Birds Get the Worms

HAVING TROUBLE PRINTING?
Printer friendly version

When relatively new to trading, I once attended a conference and vowed to another attendee that I'd learned my lesson and would never trade amateur hour again. His face solemn, looking down his nose only a little, he intoned, "I make my best trades during amateur hour."

So do I, now. If I don't enter during the first hour, I often find a setup then.

Recently, a subscriber sent an instructive article on predicting strong closes, but the article by Brett Steenbarger focused on more than closes. It included calculations pinpointing the frequency with which the day's high or low was achieved within the first thirty minutes or hour of trading. Steenbarger referenced Mark Fisher's conclusion in THE LOGICAL TRADER that high and low prices for the day tended to congregate at the beginning or end of the market day.

That conclusion coincided with an observation I had made when studying breakout plays on Donchian channels. I noticed that Donchian channel breakouts that occurred at or near the open or close were often reversed. That would be true, wouldn't it, if the high or low of the day often occurred during the early morning period?

For those not familiar with Donchian channels, they're fairly simple. The charting service sets the top boundary at the highest high for a period set by the trader and the bottom boundary at the lowest low for a period set by the trader. Richard Donchian expounded on a breakout trading system in a booklet published in 1970, with the breakouts identified as closes outside the channels and with trades taken in the direction of the breakout. Donchian set his channels to show breakouts from a four-week period's highs and low, but I often use a 20-period setting on a 60-minute chart, so that I'm looking at breakouts above the highest high in the last twenty hours of trading or below the lowest low in the last 20 hours of trading. I offset the channels to the right by two periods, so that I see the breakouts sooner than I would otherwise. However, when backtesting the system, a system I eventually discarded as not suitable for my trading style, I found that my results suffered when including those early morning breakouts. They were too often reversed.

Early Morning Donchian-Channel Breakouts

Admittedly, the period depicted in this chart encompassed a time when the SPX was alternating bullish days with bearish days, so reversals occurred one after the other, but those reversals weren't as easy to predict as markets opened each day as they are to identify in retrospect. When markets rise or drop sharply in early trading, the atmosphere at the time feels extremely bullish or bearish. That's what it's called amateur hour, because professionals often are running markets up to sell into the strength or down to buy into the weakness.

The tendency I observed for those early morning breakouts to reverse helped to identify the next market action. The same tendency depicted on this chart showed up in my study of charts for a longer period of time a couple of years ago. It wasn't new to this period when markets zigged one direction one day and zagged the other the next.

Some of those early morning breakouts on that previous chart were followed by pullbacks and not full reversals. How would one set a downside target if the move might result in only a pullback? I use Keltner channels to set targets for all my trades, whether short-term or longer term.

Don't confuse the different channels. For the purposes of this article, the Donchian channels are identifying the breakouts. The Keltners identify the targets for a reversal play.

For example, the first early morning breakout pinpointed on that Donchian chart was on October 17. Although I use different time intervals and nest various Keltner channels to set targets, depending on whether the trade is a scalp or a position to be held for several weeks, the simplest setup for an early morning reversal can be viewed on a five-minute Keltner chart. This Keltner channel uses the following parameters: Length, 120; Source AvgHLC; Exponential; Multiplier, 7.2.

Five-Minute SPX Keltner Chart:

Partial profit could be taken as the central basis line was hit, with stops lowered to breakeven on the rest of the position. A conservative trader might take full profit, depending on trading style, internals at the time of the trade, outlook and other parameters. Note that RSI had signaled oversold conditions as the central channel line was approached, so caution and profit-protecting measures were suggested.

The second early morning reversal play suggested by that Donchian channel chart was a bullish one on October 19, and it resulted in a move from one side of the Keltner channel to the other.

Early Morning Breakout on October 19:

The period on the Donchian channel chart encompassed a time when markets were zigzagging back and forth, as had been previously mentioned. Early morning breakouts were regularly reversed. What about a time when markets trended?

Turns out, there aren't too many breakouts during the first sixty minutes of the day when the move is going to trend, although I'm not certain why that's true. Both in the previous study I conducted when first testing Donchian channels a couple of years ago and in a current cursory look at trending moves since January, most trending moves tended to break out about midmorning or mid-afternoon. While I cannot verify that this pattern always exists, anecdotally, it appeared to be mostly true over the periods I've studied.

Even when those trending moves did occasionally begin on an early morning breakout, as one did on September 6, other differences often appeared to alert traders that this was the beginning of a trending move, and a reversal should not be expected.

Early Morning Donchian Channel Breakout without Stochastic
Confirmation

The RSI reading also failed to show overbought conditions as the breakout occurred, but because RSI changes so quickly and has no crossing lines, that fact was not as easy to see for illustration purposes. For that reason, stochastics were used on this chart, although I prefer RSI in my actual trading. RSI should be signaling overbought before bearish entries are considered on an early morning Donchian channel breakout and oversold before bullish ones. An early morning breakout without those signals might not be reversed or even see much of a pullback.

Other tactics can be used to identify early morning breakouts that might be reversed. Donchian channels don't have to be used. Those proficient in the use of Keltner channels can use nested channels to both identify breakouts and set targets. In addition, those who read my Market Monitor commentary know that I watch the advdec line's actions, too. On September 6, that action showed the advdec line opening near what I deemed to be support and rising, not indicating a likelihood of an early morning reversal. On October 17, however, the advdec line reacted differently, giving more corroboration to the possibility of a reversal than it had on September 6.

Traders should use their favorite indicators and methodology to decide whether they'll consider a countertrend play after an early morning breakout. The purpose of this article was to show that traders can benefit from the knowledge that market highs or lows tend to congregate either in the early morning or late-day periods, not to promote Donchian or Keltner channels.

No methodology is foolproof, and that includes the tools provided in this article. Just last Monday, October 31, the SPX set up one of these trades by creating a Donchian channel breakout during the first sixty minutes of trade. A pullback did begin, but it did not retrace all the way to the targeted support. Thursday morning's first Donchian channel breakout did finally succeed in producing a pullback that performed as expected, although the breakdown did not begin until the 12:40 five-minute candle, when the OEX finally broke through the five-minute Keltner channel line depicted above, located at 562.56 at the time the OEX slipped back below it. By the 2:40 five-minute candle, the OEX had dropped to the basis line located at 560.25, eventually dropping a little below that before climbing into the close. That setup worked perfectly, but Monday's hadn't.

Account-appropriate stops are needed to guard against days such as Monday. So is an understanding of some realities of options pricing. Options prices tend to be plumped up during amateur hour. That's the reason that I'd been so leery of entering new trades during amateur hour, although that plumping-up tendency can benefit someone exiting a position during amateur hour. Benefiting from an early entry often requires some skill in getting between the bid and the ask and a quick move of the underlying in your direction in order to profit.

Still, if I were to sit next to that same man at another seminar, I'd be able to hold my chin high, perhaps even high enough to look down my nose, only slightly.
 

Trader's Corner Archives