Saturday, March 31, 2007

Trading Principles: Efficiency and Inefficiency

Here's a principle that has served me well over the course of my nearly 30 years of trading: Every day in the market teaches a lesson. There is always at least one pattern illustrated in a market day. It might be an intermarket pattern, a statistical one, or a relationship among indicators. My job at the end of the day is to ferret out what were or would have been the best trades of the day and identify clearly and explicitly the patterns that accompanied those trades. That means that, in a year of trading, you'll have at least 250 lessons: about as much "classroom" exposure as a college student would accumulate in two courses over the span of an academic year. Over time, those lessons outside of trading hours add up and amount to a graduate education in trading.

Indeed, thinking in principles is one feature that distinguishes experts from novices. Principles enable us to integrate many perceptual concretes into a single frame of reference, widening our cognitive grasp. Where the beginning trader sees bars on a chart, the expert perceives patterns. The expert, over the course of seeing many, many instances of a phenomenon, develops a concept--a principle--covering those instances. From that point forward, the expert's perception is guided by his or her conceptual lens.

With this post, I will begin a series of articles in which I share some of my "patterns of the day" with readers and identify the trading principle(s) behind the patterns. My hope is that this will help traders not just see markets, but *understand* them.

The pattern from Friday morning's trade in the ES futures is one of inefficiency. Efficiency is the term I use to denote the relationship between sentiment (a market input) and price change (a market output). When looking at efficiency, we're asking: How much input is it taking to generate a given amount of output?

When positive sentiment can no longer drive prices higher or negative sentiment cannot yield fresh price lows, we have evidence of inefficiency. Sentiment is no longer able to move price. Such inefficiency frequently precedes tradable price reversals.

In the chart above, we see sentiment represented by the gray bars (NYSE TICK) and price represented by the red candles (ES futures). Notice how we have net buying sentiment (positive TICK) early in the morning, but--after the initial upthrust--the bullish sentiment is no longer able to move price higher. Wave after wave of buying left us still below the price highs from around 9:45 AM ET.

Eventually all those buyers are going to have to give up and sell out of their positions. For that reason, the duration of the period of inefficiency generally correlates well with the extent of the subsequent price reversal.

We can see that reversal occur when declines in the NYSE TICK starting around 11:15 AM ET take us to new price lows. That tells us that we are *gaining* efficiency to the downside. We can also see that volume expands on the declines. That tells us that the longs are capitulating.

When you study enough efficiency/inefficiency patterns over varying time frames, you eventually gain the ability to see the patterns in real time as they unfold. But you need the principle to guide your perception, and you need to perceive before you trade.

Find at least one lesson from every day's trade, and you'll be amazed at the transformation of your perception.


AnaTrader said...


Your quote:
When positive sentiment can no longer drive prices higher or negative sentiment cannot yield fresh price lows, we have evidence of inefficiency. Sentiment is no longer able to move price. Such inefficiency frequently precedes tradable price reversals. Unquote

I wish to add to your remarks about Inefficiency on ES futures last Friday. On my study, there is a formation of Expanding Terminal Triangle, which indicated a breakout to the upside. (COM 1439).

However, in spite of positive seasonality, we need to see how markets will behave next week.

Hence, the need to identify and perceive patterns of the day, behind the patterns, as you put it so succinctly.

Yes, I am learning new lessons with each passing trade, and your new series of articles on patterns will add to my learning process.

Hsieh hsieh ni.

Brett Steenbarger, Ph.D. said...

Hi AnaTrader,

Thanks for the note. When a more extended time period has prices that are inside the prices for a less extended, prior time period, we have a consolidation range. By tracking demand and supply within that range and using pivot targets derived from the high/low/close of the entire formation, it is possible to generate very interesting, potentially high probability trade ideas. I will be posting on this shortly.


Yaser Anwar said...


I'm a little confused. Doesn't selling mean there is volume at the ask/offer price? You've indicated increased volume at the bid, isn't bid for the buyers? So in your example shouldn't it be increased volume at the ask/offer? tx

TradeR said...


Superb post, looking forward to the upcoming series. IMO traderfeed stands out a mile in the blogsphere for this content discipline - I'd join the other posts in thanking you for the level of contribution you make on a daily basis :-)

In terms of efficiency/inefficiency are you effectively looking at the level of correlation between movements in the NYSE TICK/Volume combination and the underlying movements in the ES contract price? Clearly real-time identification of this involves a high degree of assembling various pieces of quantitative data, but would it also be possible to calculate an efficiency coefficient in real-time to measure how sensitive price is being to sentiment/volume deltas?

Brett Steenbarger, Ph.D. said...

Hi Yaser,

Thanks for the opportunity to clarify. If sellers are willing to transact at the market bid, they are showing motivation to exit the market, as they're accepting the lower price. If buyers are willing to lift offers, they're motivated to own stocks and willing to accept the higher price. It's the principle behind Market Delta, and it's also what makes the NYSE TICK valuable.


Brett Steenbarger, Ph.D. said...

Hi Trader,

Yes, it is possible to quanitfy efficiency with an indicator by tracking how much price change you're getting per move in the NYSE TICK (or other indicator). Often the loss of efficiency can be viewed directly in a chart, when you see that an index can no longer make new highs or lows on buying or selling.


Ziad said...

Hi Dr. Steenbarger,

Great post on efficiency and infficiency, but I have a question for you. One pattern that you talk about is that of topping and bottoming processes. Much of the times the process manifests itself in a momentum extreme that is then followed by a price high/low. In such instances though we kind of have the reverse concept as effiency/inefficiency. In itself it makes sense as you are seeing new price highs or lows that are not longer being supported by excessive sentiment and are thus receiving less participation at the bid/offer. A reversal is logical. However, if you view it from the pricinciple of effiency, it can be construed as: new price highs or lows are being made without requiring excessive sentiment- which is actually efficient, and would call for continuation.

And actually the converse to this would also apply. For instance in your chart example, you illustrate the concept of inefficiency where TICK extremes couldnt take price up further. But looked at from the principle of topping processes, it could actually be seen as a continuation set-up and not a top, since lower highs are still attracting high sentiment and thus rises still getting good participation on the offer.

So it seems to me that the principles are the inverse of eachother in a sense, and could be contradictory. However, I have actually seen them both to work, and so I'm not trying to argue against them, but rather seeing if you could illucidate the concept in such a way as to make it possible to differentiate the two opposing ways of thinking so as not to get stuck being indecisive when action is needed. So is there a subtle difference that i'm missing, which would render them non-contradictory principles?

Thank you very much.

Brett Steenbarger, Ph.D. said...

Hi Ziad,

Thanks for the thoughtful questions. Think of topping and bottoming as a three-phase process: 1) momentum high/low; 2) price high/low; 3) further bounces/drops on positive/negative sentiment that can no longer make new price lows. The latter phase represents inefficiency. A head and shoulders pattern would be a common chart example.