Wednesday, February 28, 2007

Wednesday, February 28th Morning Market Comments

9:53 AM - OK, let's wrap up for the AM session. It was good that I blogged more actively this AM, despite my aching fingers, because it shows how *the same market principles apply* regardless of a market's trend or volatility. I started the day with a dual schema: some research suggested we tend to bounce after big days down, other research suggested follow through weakness after big downside momentum days. My leaning was to go with the trend and the idea of downside weakness to test Tuesday's lows. Well, we got the weakness and ER2 did break the lows, but not ES. Note my post on buying of large caps in the downward market. More of that seemed to be going on, as the TICK lows on the weakness were not extreme. When we broke below the overnight lows in ES and then moved back up into that range, it was a good sign that sellers just couldn't get it done. The TICK distribution gradually shifted upward and it was clear we'd see a test of those overnight highs. We broke the highs, but now note we're heading back into the range. Just as a move back into the range suggested the sellers couldn't get it done, a move back into the range now would indicate that buyers don't have the follow through. That would lead us to expect a move back to the day's average price. Just keep the principles in mind and size your positions so that any move against you in this heightened volatility won't blow you out. Hope this has been helpful. I'll update the market tonight on the Weblog; now I need to rest those fingers!! Have a great day.

9:45 AM - Sweet. Now we have to stay above that range.

9:40 AM - If we can stay in positive TICK on pullbacks, we should get a big breakout move.

9:36 AM - Note we touched the upper end of the overnight range and have pulled back a bit. We actually broke above this range in NQ and ER2 is now close. I would expect a significant move up if we can take out that range; that TICK distribution should tell the story. So far, it and ER2 remain strong.

9:29 AM - I'm getting a ton of emails from confused traders. Sorry I can't answer all in a timely fashion. I'm getting carpal tunnel syndrome as it is... :-) Note the upward shift in the TICK following the drying up of selling mentioned below. That has triggered the recent move.

9:19 AM - Wonderful illustration of the principle I mentioned a few minutes ago. Drying up of selling often precedes an influx of buying. Then, if we get subsequent selling that holds above prior lows on higher TICK lows, that becomes a secondary buying opportunity.

9:13 AM - Overall, we're seeing more volume at the bid than at the offer in ES, but not by much. I have to say that this market is behaving in a resilient way thus far. But drying up selling *often* leads to an influx of buyers--keep an eye on ER2 and TICK if that is to be the case--but we do need to see those buyers to be confident in any market rebound.

9:03 AM - Continued ER2 weakness. Still need to surmount the overnight range highs. That would turn me bullish.

8:54 AM - And we're pretty much back to where we opened! Note the breakout high in the TICK and the fact that, so far, we have not seen a TICK reading below -500. Quite a contrast to yesterday. Looks like this mkt is searching for a bottom--esp. if we get higher TICK lows on successive declines. Remember we have housing at 9:00 CT.

8:52 AM - Hope you can see that: the low of the overnight range is now overhead resistance.

8:50 AM - Note that we broke Tuesday's lows in ER2. Note also how that overnight range high couldn't be breached and led to the breakout below the overnight low. Let's now see if we stay beneath that range and how much volume buyers can sustain.

8:42 AM - The ER2 weakness was a good tell that we weren't going to break above that range this time around. I'll be watching ER2 for continued signs of strength or weakness.

8:40 AM - Note the trading range from 1402 to the 1413 area in the overnight mkt. That's the range to keep an eye on.

8:38 AM - Very high volume; ER2 showing some relative weakness. Keep an eye to see if it leads us down.

8:12 AM CT - Once again, I'll blog on this market only when I see something of real note. I've been getting record traffic on the blog--as have other bloggers--which attests to the heightened interest and uncertainty surrounding this market. My recent posts have tried to put the market decline into some kind of perspective. Bottom line is that we had an afternoon bounce to 1413.25 in ES and then a rise in overnight trading to 1412.50. We need to see the buyers take out those levels with fresh, solid buying volume to take the bull position. Note that we *did* take out the Tuesday afternoon highs in NQ and ER2 during overnight Globex trade, thanks to the bounce in the Shanghai market. So, in those markets, we need to see buyers sustain a move above those overnight highs. That would target the Tuesday average trading price (pivots posted to the Weblog). Failure to hold those highs would target a test of the overnight lows; then the Tuesday lows. We have Chicago PMI at 8:45 AM and new home sales at 9:00 AM. There are concerns regarding economic weakness that could affect this market. Also keep an eye on the Yen. We're getting a bit of a bounce from the recent drop, and it's Yen strength that has helped to hurt this market. Remember: volatility and movement in the overnight session tends to correlate well with volatility in the morning session. Position sizing and stops need to account for that shift. Back after the open.

How Did Money Flow During the Market Decline?

I mentioned recently that, throughout this bull market, it was rare to see a day in which net funds had flowed out of the Dow 30 stocks. As I was going through my market stats at the end of Tuesday's sharp decline, I noticed that the TIKI (the net number of stocks trading on upticks minus downticks for the Dow stocks) was actually positive on the day. This was in marked contrast to the NYSE TICK (which tracks the net upticks minus downticks among all NYSE-listed issues), which was quite negative on the day. Could it be, I wondered, that investors were putting money to work in safer, blue cap issues during the market decline and shunning more volatile, smaller stocks? That would certainly fit with the risk-aversion theme associated with the drop in China and the rise of the Yen.

I examined the Dollar Volume Flows into and out of each of the Dow stocks and indeed found that the Dow attracted net dollars on the trading session. Far and away the net money flow winner was XOM, followed by T, GE, and IBM. Stocks with net outflows on the day were paced by a stock that has seen outflows for a while: MSFT. Other stocks with outflows on the day were PFE, WMT, KO, JNJ, INTC, DIS, HON, GM, DD, CAT, C, and AIG. Of these, the first six have also had net outflows over the past several weeks.

What it looks like from this angle was that traders were putting their money into some old-line safe names--most notably an oil stock--and taking further money out of stocks that were already out of favor. Overall, however, there may have been less of an institutional exodus from large cap stocks than is apparent from the Dow chart.

High Momentum Stock Market Declines: What Comes Next

My recent post took a look at what typically follows a single day of large price declines. Here is a very different take on near-term prospects following a high momentum decline. Recall that one of my core trading principles is that a trending move with strong momentum tends to persist in the short run. Let's see if that might apply to the current market.

My best measure of stock momentum is the proprietary Demand/Supply index that I track daily on the Trading Psychology Weblog. Demand, you might recall, is an index of the number of stocks closing above the volatility envelopes surrounding their short- and intermediate-term moving averages. Supply is a similar index, tracking the number of issues closing below the volatility envelopes surrounding both their moving averages.

Demand and Supply follow every operating company trading on the NYSE, NASDAQ, and ASE. They tell you the relative number of issues closing either very strong or very weak in the equity markets. I have collected the Supply and Demand data since September, 2002. Tuesday's market hit the highest level of Supply recorded over that period: 420. We have only had seven other days with Supply levels above 200. The next day in SPY, five of the seven days finished lower in price, by an average of -.38%. Five days later, six of the seven occasions finished lower, by an average of -1.06%. These are much weaker results than the average market performance over that period.

Of the seven sessions in which we had very strong downside momentum, six traded below their day's lows during the following session. Indeed, four of the seven broke those lows by more than ten full ES points. Weakness, then, tends to carry over to the next day following a high momentum decline.

Interestingly, when we had days in which downside momentum was strong but not extreme (Supply greater than 1500, but less than 2000; N = 24), the next five days in SPY averaged a healthy gain of .67% (15 up, 9 down). By contrast, the average five-day gain over the entire period was .24%.

In sum, when we have large declines on downside momentum that is extremely broad--as we had Tuesday--we tend to see followthrough to the downside over the short term. When large declines are not so extreme in the breadth of their negative momentum, we tend to see reversals over the short term. It is for this reason, in spite of the small sample size from my data, that I will be open to selling strength early on Wednesday in anticipation of breaking Tuesday's lows. Note, however, that as I'm writing, we have bounced significantly higher in the overnight stock index futures. A recent comment on my previous post from Trading Nerd took the data back to 1930 (!) and found evidence of positive returns after a 3% decline, but that most of those occurred the next trading day. Such a pattern--sharp bounce, then some follow through decline--would help to connect the results from my previous post and this one.

Tuesday, February 27, 2007

Big Down Days: What Happens Next?


That seems to be the question everyone is asking following the meltdown on Tuesday. What we observed is a scenario in which everyone simultaneously runs for the exits. China's market was down over 9% in Shanghai prior to our open. The Yen soared to multiweek highs, a risk factor to the carry trade, as mentioned here a while back. As selling in the U.S. markets begat selling, we saw weeks' worth of profits eroded in a single session. Among money managers, who fight tooth and claw for each quarter's performance, such a drop was a major threat. Volume picked up dramatically, reflecting the exit of the institutional traders, and--with that--the decline accelerated, now wiping out months' worth of performance.

Such an occurrence is rare. I went back to 1990 (N = 4304 trading days) and could only find 25 occasions of a single day drop of 3% or more in the cash S&P 500 Index. What's even more rare is to have such a decline in the context of a bull market. How many times in the last 17+ years have we seen a 3+% decline in a market that had been up over the prior 20 and 60 days? It's only occurred three times out of all those trading days.

But shifts happen. And that, my friends, is why money management and risk control are all important. If you double your money, double it again, and then double it again, only to lose 90% in the next debacle, you wind up down 20% on your initial capital, needing a 25% gain just to return to where you were at the start.

So what happens after big down days? Let's start with the 25 occasions in which we've dropped 3% or more in a single day. At just about every time frame from one to twenty days out, returns following such a large single-day drop are quite bullish. One day later, the S&P averaged a gain of .47% (17 up, 8 down), much stronger than the average single day gain for the rest of the sample of .03% (2256 up, 2023 down). Twenty days later, the S&P was up by an average of a whopping 4.47% (20 up, 5 down), again much stronger than the average 20-day gain of .73% for the remainder of the sample (2641 up, 1638 down). In all, large down days have tended to represent buying opportunities since 1990.

Those findings may be a bit deceptive, however, because--of those 25 large down days--15 occurred in the context of a market that had *already fallen* 3% or more over the past 20 sessions. In other words, large down days have tended to occur toward the end of market downmoves--as a kind of washout. We don't typically see large down days following intermediate term strength, as noted above.

I relaxed my criteria a bit and found six occasions in which the market had not been down more than 2% on a 20 and 60 day basis prior to the large down day. All six occasions were up the next day, and all were up over the following 20 sessions. Indeed, the average gain over the next 20 trading days was an impressive 3.89%.

In short, big down days have tended to represent panic and, even when they've occurred in relatively strong markets, they have tended to occur nearer to market bottoms than tops. On average, traders have made money by buying into such panicky markets. While the current volatility may be more than many traders wish to tolerate, at the very least the pattern of superior returns following panicky declines should offer caution to those tempted to arrive late to the bear's party.

Tuesday, February 27th Morning Market Comments

9:36 AM - So far, note that the selling has come on higher TICK lows and the major indices have held their AM lows. If the TICK continues its positive shift, I'd expect us to take out the opening prices in ES. Note the upward shift of the TICK distribution thus far. As long as we get TICK pullbacks at higher lows and price stays above prior lows, I'll look to buy. Breakdown of those prior lows or breakdown in TICK would alter that, however. Hope that's helpful. Have a great remainder of your trading day.

9:14 AM - Note that we're hovering around the opening prices in ER2 and NQ. If you look at the TICK, really the selling is not as dramatic as the numbers would lead you to believe. A lot of the market's weakness was baked into the open. I'm focusing on how strong or weak the market is *since the open*. Not versus yesterday's close. That will tell us if we have underlying followthrough of supply or buyers laying in wait.

9:00 AM - Short covering rallies in declining markets can be quite sharp. Note we've seen some bargain hunting in NQ and ER2, which had been leading the downside. Seeing if we can put in a bottom this AM and get a real mother of all short coverings!

8:50 AM - I'm showing an equity put/call ratio of 1.5. That's as high as I can recall in a long time. Trin is close to 3.0. Declines lead advancers by well over 2000 issues. Lots and lots of bearishness.

8:43 AM - Keep an eye on EEM as a proxy for those emerging markets. Especially watch for strength.

8:27 AM - It's really funny and perhaps a sentiment indicator: there are a record number of visitors on the TraderFeed site right now. I'd like to think it's part of that flight to quality... :-)
More likely, people seek information when there is uncertainty...

8:24 AM - Thanks also to a reader and fine trader who points out that we're seeing multi-week highs in the Yen which, if you recall my earlier post, is associated with subnormal returns due to the dampening of the carry trade. Well, I don't think too many traders want to be carrying BRICs right now; we see a huge rally in the rates markets with the flight to quality. Look for a TraderFeed post coming up: what happens after we have large market declines and a flight to quality in bills/bonds.

8:13 AM CT - Well, I wasn't planning on continuing the comments so immediately, but it looks like a wild and wooly start to the morning, so I thought I'd add a little perspective. Bottom line is that we had a drop of over 9% in the Shanghai market overnight. That has spilled over to world markets, so look for a continued flight to safety (buying fixed income) and a retreat from emerging markets (big drop in EEM in premarket). As the Barchart newsletter reported earlier this AM, "The vice chairman of China's parliament said on Feb 6 that the government said that China's stock market was showing signs of a "bubble" and said that only 30% of the companies on the Shanghai Stock Exchange are "good to invest in by Western standards." I guess when you say those kinds of things about your own market, eventually people will listen! Does this mean the start of a bear market, or is this a correction that shakes out weak bulls? I try to address this in my latest post. We are trading some 15 ES points below yesterday's volume weighted average price. For two days, as noted in the Weblog, we've been in a short-term downtrend. This simply accelerates the momentum of the downtrend. In general, such momentum tends to persist in the short run. So I would caution against attempts to bottom pick this AM. The most important consideration, however, is that overnight range is correlated with volatility the next AM. We should see considerable market volume and movement this morning--much different from the slow, range bound, low VIX trade of recent weeks. Be very careful re: your placement of stops and your position sizing. The market is likely to move more than you are accustomed. If I notice anything major, I'll post observations. Have a good start to your morning trade!

Which Dow Stocks Are Hot, Which Are Not, and What It Might Mean

In my recent post and on my Trader Performance page, I introduced a measure that I called Relative Dollar Volume Flow. This indicator is designed to assess the sentiment of large traders in individual stocks by multiplying the price of each trade by its volume and either adding that amount to a cumulative total (if the trade occurs on an uptick) or subtracting it (if it occurs on a downtick). The daily total is then divided by the volume for that day to obtain a relative sense of the proportion of volume that was occurring at the market bid (on downticks) vs. the market offer (on upticks). If the Relative Dollar Volume Flow is positive, we have net dollars flowing into the stock for the day. If the Relative Dollar Volume Flow is negative, we have net dollars flowing out of the stock for that day.

As a whole, dollars have been flowing into the Dow. Since 2004 (N = 787 trading days), we have only had 57 occasions in which net money flowed out of the Dow 30 stocks as a whole. In other words, when we compute the Relative Dollar Volume Flow for each of the Dow 30 issues and sum these to obtain a total Flow for the Dow, rarely have we seen net selling of the Dow. More on that later.

This strength of money flow, however, belies large differences among the individual Dow components. Over the past 25 trading sessions, the Dow as a whole has been modestly higher. Let's look, though, at how many of those trading sessions saw net dollar inflows for each of the Dow issues:

AA - 22
AIG - 19
AXP - 17
BA - 24
C - 20
CAT - 24
DD - 25
DIS - 13
GE - 18
GM - 25
HD - 17
HON - 22
HPQ - 21
IBM - 24
INTC - 16
JNJ - 13
JPM - 18
KO - 11
MCD - 21
MMM - 19
MO - 23
MRK - 22
MSFT - 7
PFE - 8
PG - 23
T - 22
UTX - 22
VZ - 22
WMT - 14
XOM - 21

What we can see is consistent inflows into such issues as BA, CAT, DD, GM, and IBM and more consistent outflows from MSFT, PFE, KO, DIS, JNJ, and WMT. Overall, 17 of the Dow 30 have seen net inflows for 20 or more days in the last 25 trading sessions. That is not exactly a picture of broad weakness. Rather, it suggests that institutions have been moving money away from selected issues (many of which might be characterized as blue chip growth rather than value). I find it informative to go to a site such as Instant Bull and examine the news and the message board buzz about these stocks, as a way of seeing what the market is concerned about. In the case of the stocks with the most consistent outflows, MSFT and PFE, there are clear concerns surrounding Vista and health care costs, respectively. The Dollar Volume Flow figures suggest that money managers are truly acting upon those concerns.

Overall, since 2004, we have had corrections in the market, generally in the vicinity of 5-7%. On those occasions, we've seen a reduction of net inflows into the Dow 30 stocks, but no systematic liquidation. I went back to the 57 trading days in which we had net outflows from the Dow 30 as a whole. Three days later, the Dow (DIA) was up by an average of .35% (35 up, 22 down). That compares to the average three-day gain in DIA for the remainder of the sample of .06% (396 up, 334 down). In other words, investors have tended to use outflows from the Dow as short-term buying opportunities.

At some point we will have a bear market. Indeed, as I write, the preopening stock index futures are trading much lower. What will tell us whether such a move is a normal correction vs. a full-fledged start to a bear market? One thing we might look for is a broadening of dollar outflows from the Dow issues. As long as outflows are selective and money is moving into the Dow stocks as a whole, market declines are likely to be contained. That's been the pattern since 2004. To this point, I've seen no change to the pattern in the recent market data. The Dow has been down for three consecutive sessions, but only 9 of the Dow 30 have seen net outflows over this period.

Monday, February 26, 2007

Monday, February 26th Morning Market Comments

9:41 AM - Seeing a little relative strength in the semis; let's see if that continues. As long as we get more volume in ES hitting bids than offers, it's tough to pull off any turnaround in that index. Notice how the short-term trader breaks down the market into swings up and down and continuously compares the strength of the swings relative to one another to gauge underlying demand and supply. That's where the tools such as TICK and volume at bid and offer come in handy. We saw strength in the overnight market, but then buying soon after the open could not move us above the premarket highs and we saw selling in NQ and ER2, as well as semis. That set us up for a move back into the value area of Friday's trade.

Hope these morning sessions have sensitized you to some short-term patterns in the market. I will hold occasional morning sessions during March as the schedule allows. Thanks for your interest, and have a great day!

9:34 AM - Well, no sooner did I post that than we saw a bounce in TICK on low volume and then strong renewed selling. From the standpoint of execution, it makes sense to wait for those bounces to enter in the direction of the trend, not to get excited and sell weakness. I find it helpful to watch not only the net buying vs selling in the Market Delta bars, but also the high and low Delta values for the bar to see if we're getting stronger or weaker *within the bar period*. This very much helps for execution, as it keeps you out of buying a market in which price may be bouncing, but buying volume is tepid. Back in a bit.

9:28 AM - When you see selling in the TICK and volume hitting bids in ES but the indices can't make new price lows, it's telling you that the selling sentiment is still there, but that sentiment is having a harder time moving price. In other words, the market is becoming less efficient to the downside. Such shifts in efficiency are common around market turns, so I usually don't chase weakness if price is having a hard time budging.

9:12 AM - When you have downthrusts like that, what you want to do is wait for the buyers to take their turn. You'll see that by bounces in the NYSE TICK and by buying in the Market Delta (green). The important thing is to compare the volume on the bounces to the volume on the prior decline and to compare the level of the TICK on the bounces to the prior TICK levels. If the bounces are occurring at lower volume levels, with less buying at the offer, and at lower TICK levels, those become opportunities to ride the market for another move down. Conversely, if you get solid volume on the bounces and new TICK highs, that's when you look for a short-term reversal in the market. More on that in a bit.

8:55 AM - You can see how the ER2 and NQ weakness were great tells for the short-term breakdown in ES, which has now attracted increased selling volume. Nice example of how looking at the more volatile sectors/indices can alert you to what is likely to happen in the broader market.

8:49 AM - Seeing some profit taking in ER2 and NQ, as well as semis, despite positive TICK. We're getting more ES volume at offer than bid, but again not impressively so. I'd need to see greater TICK weakness before being aggressive on the downside, but neither is the buying follow through impressive so far.

8:40 AM CT - Watching some early relative weakness in semiconductors and NQ. Worth keeping an eye on. Volume in ES is decent, not extreme; with more volume transacted at offer than bid, but not by as much as you might have thought given the overnite strength.

8:17 AM CT - And a very good morning to you! I'm wrapping up my morning session blogging today, and today will be a very light day of blogging at that. Lots of new projects going on here, and those will be taking my time and attention. One of those projects is described in my latest post; some promising research for swing traders. See also my latest entry in the Trader Performance section of my personal site. Another project is learning online video production. Eventually I'll start posting some instructional videos that I think will be more informative in a how-to way than the morning session posts. I also talked with my editor at Wiley and we're investigating the possibility of putting together a combination book/video that focuses on some of the trading ideas from the blog and the morning sessions. And, finally, as we move into March and weather improves, I'll be back on the road and working with traders at different firms. It's all good...just a lot of balls to juggle at once! I know things are hectic when my email inbox reads over 900!

We have some good overnight buying in both Europe and in the U.S. futures. Interest rates are down, and we're back into the prior trading range in the equity indices. Note that Friday we broke the prior day's lows in ES, but not in ER2 and only barely in NQ. I'm looking at the possibility of this representing an intermediate-term bottom that will take us to new price highs. So my leaning is to be buying on selling that dries up. I had the right general idea on Friday, but jumped the gun on the buying. We need to see evidence of sellers being unable to push the market lower before committing to the long side. I'll be watching the semiconductors (which have been strong lately) and ER2 for signs of strength vs. reversal. NQ and ER2 are pretty close to their previous week's highs; keep an eye on those levels and whether or not they attract volume. Pivot targets are posted to the Weblog. Note that we are trading pretty much at the average level of the prior trading range at present. No economic releases today, but quite a few coming up this week. Let's see if trading slows ahead of those numbers. Back after the open; I will be posting only when I see something of significance.

Tracking the Dollar Volume Flows of Stocks

Over the weekend, I've been working on some fascinating research to answer a group of questions: How much money is actually flowing into or out of a particular stock on a given session? Are these dollar volume flows increasing or decreasing over time? Is the stock attracting more dollars than other stocks in the same sector? Than stocks in other sectors? And--most important of all--how is the amount of dollars flowing into a stock related to how that stock behaves going forward?

As I indicate in my latest Trader Performance entry, Relative Dollar Volume Flow appears to be a promising measure of underlying demand or supply for a stock. The idea is simple: take every trade that occurs in a stock during the day and multiply the price at which the trade was transacted by the volume of that trade. If the trade occurred on an uptick (or if the last price change was an uptick), you add this value to a running cumulative total for the day. If the trade occurred on a downtick (or if the last price change was a downtick), you subtract this value from the running cumulative total for the day. At the end of the day, the final cumulative total reflects the buying or selling of large market participants (institutions that transact large blocks, large portfolio managers), because their transactions account for most of total (given the large volume of their purchases or sales).

If you have institutions eager to get into a stock, they will buy large positions on upticks. If those institutions are eager to bail out of a stock, they will sell large positions on downticks. The logic is similar to Market Delta, only this is a measure intended for longer-term analysis. (It would be interesting to study it on an intraday basis for signals, but so far I have not done so).

I refer to this as *Relative* Dollar Volume Flow because I divide the cumulative total at the end of the day by that day's trading volume. This provides a better comparison of the money flowing into one stock vs. another: One million dollars flowing into a large cap stock that is an institutional favorite will reflect lower Relative Dollar Volume Flow than one million dollars flowing into a lightly traded small cap issue.

One application of Relative Dollar Volume Flow is to ETFs. Does the amount of money flowing into or out of an ETF tell us anything about the Fund's performance going forward?

I conducted a small investigation of the Dow Jones Industrial Average ETF (DIA) by tracking its daily Relative Dollar Volume Flow from the beginning of 2004 to the present (N = 771 trading days). When the 20 day Flow for DIA was positive (net money flowing into DIA; N = 442), the next 20 days in DIA averaged a gain of only .05% (245 up, 197 down). When the 20 day Flow was negative (net money flowing out of DIA; N = 329), the next 20 days in DIA averaged a very solid gain of 1.11% (227 up, 102 down). This is a pronounced reversal effect: when we see money flowing out of DIA over a 20-day period, returns have been superior. When DIA has attracted dollars over a 20-day period, returns have been subnormal.

Very interestingly, if we calculate the Dollar Volume Flow for the Dow by individually summing up the Flows for the 30 component stocks, this value correlates with the Flow for DIA by only .14. In other words, money flowing into and out of DIA is not especially reflective of money flowing in and out of the individual Dow stocks. Rather, DIA's flows (like those of other index ETFs) may be more skewed by the activity of hedgers and program traders who might not be making long-term portfolio adjustments.

Indeed, the Dollar Volume Flows for the summed 30 Dow stocks shows that money has been continuously flowing into those issues through the bull market and really accelerated after we made a bottom in July of 2006. Although there are individual issues with net dollar outflows (MSFT is notable in that regard), there is little evidence at present that money is systematically flowing out of the Dow. (Inflows have slowed over the last two weeks, but remain positive).

An especially promising strategy is to identify stocks with favorable expectations based on their Relative Dollar Volume Flows and those with subnormal expectations and create a long/short strategy of buying the former and selling the latter. This would create a diversified portfolio that would be relatively market neutral. I'm also beginning to take a look at falling stocks with rising Dollar Volume Flows and rising stocks with weakening flows to see if we can anticipate turns in those issues. Clearly, there's a lot of room for interesting research here. I hope to report results in the near future.

Sunday, February 25, 2007

Five Guiding Principles of Short-Term Trading

I recently posted five guiding principles of trading psychology. Those are big concepts that help me understand traders. Here are five guiding principles that I draw upon in understanding the psychology of the marketplace:

1) Markets are moved by their largest participants - I believe this is the single most important principle in short-term trading. Accordingly, I track the presence of large traders by determining how much volume is in the market and how that compares to average. Because volume correlates very highly with volatility, the market's relative volume helps you determine the amount of movement likely at any given time frame--and it helps you handicap the odds of trending vs. remaining slow and range bound.

2) Trends are created by shifts in supply/demand generated by global/macro relationships - The forces that ultimately move stocks are related to intermarket changes in supply and demand that result from interest rate shifts, movements among currencies, inflation and commodity prices, and profitability among companies in various international markets. This means that the short-term trader must keep one eye on the intraday shifts in buying and selling of large traders, but the other eye on the day to day and week to week changes that impact the demand for stocks. Knowing the big picture is crucial to trading in the right direction, even when the trading is short-term and your focus is on the initial minutes of trading.

3) Market moves on strong momentum tend to persist in the short run; moves on weak momentum tend to reverse - When the vast majority of stocks are participating in a market move, those moves tend to continue as short-term trends. When markets make new highs or lows and a large proportion of stocks and sectors do not follow, those moves are more likely to be reversed. This is one reason that it is important to watch the entire market, not just the stock or instrument that you happen to be trading. Because the major market indices tend to be capitalization weighted, they can make new highs or lows simply because a few stocks happen to be strong or weak. Tracking the breadth of moves tells you if there is an overall demand or supply for stocks.

4) There are only three kinds of trades: breakout trades, trend-following trades, and reversal trades. Breakouts and reversal typically occur in the context of bracketing markets. As we establish value in a relatively narrow range, the market probes the upper and lower bounds of the value range to test demand and supply. The breakout trades result when prices away from value (i.e., away from the average trading price within a range) attract fresh buying or selling interest. The reversal trades occur when prices away from value fail to attract buyers or sellers, leading to movement back toward value (the average trading price). The trend-following trade occurs after a breakout and consists of buying pullbacks in an uptrend or selling bounces in a decline. These trades last as long as markets continue to make new highs or lows on successive moves in the trend direction--and especially as the majority of stocks participate in those new highs or lows. Knowing the status of the market--bracketing or trending--tells you the type of trade likely to set up. Following Market Profile over multiple time frames is very helpful in determining market status.

5) Tracking market activity at the bid and offer is useful in detecting shifts in short-term demand and supply - Such tracking includes contract volume at the bid vs. offer, as charted in Market Delta, but also includes the number of stocks trading at their offer vs. bid price, as in the NYSE TICK. Over time, we commonly see transitional structures in which extremes of buying or selling interest bring the market to new highs or lows, resulting in reduced buying/selling, and eventual reversal. These transitions clue us into shifts in the psychology of the marketplace and are very helpful in framing trade ideas.

I stress the above five principles because they capture how I've found successful professional traders approach the markets. I'm sure there are highly profitable pros somewhere who trade wave patterns, moving averages, chart formations, and the like. In several years of working hands on with such traders, however, I have yet to meet one who uses these methods. The pros do, on the other hand, care very much about who is in the markets and why markets are moving. The principles provide a framework for making sense of market behavior, which then can be used to filter the setups provided by charts, cycles, and the like. The really good traders understand markets; they don't just predict them.

P&L Management: A Best Practice in Trading

From Brett: Our next best practice in trading comes from David Aferiat of Trade Ideas, who passes along a practice from some of the firms he works with. The idea is that we can become emotionally connected (or disconnected) to the specific stocks or instruments that are most recently contributing to our profit and loss (P & L). Such reactions take our attention away from sound money management. My comments follow David's submission.

From David: We work with many trading firms across the country that use Trade-Ideas to assist in their idea generation and risk management decisions. Dan Mirkin mentioned this in his recent appearance at the NY Traders Expo. Several firms teach their newer traders to remove a common heuristic that emphasizes previous winning trades over losing trades.

Traders tend to review their P&L and focus more on what symbols were winners for them in the current session. To reduce the tendency to get tangled up in winners and losers, these traders, who manage an average of 10 – 12 positions at any given time; minimize the ‘Symbol’ field in their open portfolio view. Only ‘Number of shares’, ‘Last Bid/Ask’, and ‘P&L’ fields are viewed.

When a trader does not see the symbol or name of the instrument, all the focus shifts to the P&L and the money lost or won in the position and making decisions based solely on money is a lot easier and quicker to process than a mini-internal debate about what the symbol is doing to your P&L.

From Brett: A similar practice that some traders found helpful at the proprietary trading firm where I worked was that they would cover up their day's P/L display so that they would not become overly focused on the money in making decisions. They knew that they had a risk manager at the firm who would call them if they hit a "warning level" of losses for the day. They didn't want to become too aggressive when they were winning, and they didn't want to become too risk averse if they were down money. The idea here is to separate out as much as possible your trading decisions from the ups and downs of what has happened most recently (much of which is random). Once you have your position sizing, price targets, and stops in place, you let the profits and losses take care of themselves based upon the edge you've cultivated. You want to be trading opportunity, not your mood of the moment. Tuning out data that would bias your trading--whether it's the TV, the "hot" or "cold" stock symbols, or a green/red P/L number--is helpful to emotional P&L management and a best practice. Thanks to David for passing this along.

Saturday, February 24, 2007

Book Review: Markets in Profile

From Brett: I recently mentioned the use of Market Profile as a trading best practice. Here is my full review of Markets in Profile: Profiting from the Auction Process by James Dalton, Robert Dalton, and Eric Jones (Wiley, 2007). I believe the underlying concept behind the book--understanding what is happening in the auction process at timeframes longer than your normal trading horizon--is an important one. "Which timeframe is in control of the market?" is a question that short-term traders rarely address, leaving them open to unexpected market shifts. The Markets in Profile book does a nice job of showing traders how to answer this question--and what to do about it. The following review was posted to the Amazon site earlier today.


When I initiated an internship program for new traders at a proprietary trading firm, I began by providing students with a conceptual overview of markets and the factors that make markets move. That overview drew heavily upon the book "Mind Over Markets" by James Dalton, Eric Jones, and Robert Dalton. Quite simply, that book is the most readable, lucid, and insightful work on the auction process underlying trading markets that I have yet encountered.

You can imagine, then, my delight when I had the opportunity to read an advance copy of the authors' new work, "Markets in Profile" and provide a review statement for the book's cover. I turn down these endorsement requests rather often. Frankly, many of them are simply favors being done by buddies and don't reflect an actual reading of the text. The least authors can do for their readers is keep it real.

There was no problem in my keeping it real when I wrote that "Good books teach, but the best books enlighten". Markets in Profile is one of those enlightening books, because it takes the conceptual framework from the first book and applies it to concrete trading time frames and trading scenarios. In that sense, I would say it is a worthy bridge between theory/understanding and practice.

A core theme of Markets in the Dalton, Dalton, and Jones text is the following nugget: "If you can correctly identify which timeframe is in control of market activity, and you have a good understanding of how the individual timeframes generally behave, then you are in a stronger position to trade, invest, and effectively control risk." The practical aspect of the book is that it illustrates how to read the Market Profile to grasp the long-, intermediate-, and short-term auctions and place current market activity into a proper context.

Time and again I have observed that a trading edge comes from understanding what is happening at the timeframes greater than the one you're trading. If you're in a trading range, you might want to be prepared for a breakout. Which way is it likely to break? Using price, time, and volume to see how value is being established at those larger timeframes provides invaluable clues.

As the authors also explain, observing volume patterns as the market moves away from value at one timeframe is very important to understanding whether or not we will return to value or reprice value (perhaps in line with the longer-term auctions). Seeing these dynamics evolve over time helps traders understand what is likely to occur, but also *why*.

I find the authors' use of examples, charts, and Market Profile diagrams to be quite helpful. This is not an encyclopedic work; it's a very readable 200 pages. By the end, however, readers will have a solid understanding of how to apply auction-based principles across timeframes and frame their trade ideas with that broader understanding. A particularly practical chapter is the one on day-trading, which shows how these principles are relevant to the active trader.

The authors make clear that Market Profile is not a trading system, but a framework from which trading ideas can be derived. My analogy isn't perfect, but perhaps it will make sense: Just as the periodic table helps chemists organize their understanding of elements and predict how elements are likely to interact with one another, the Market Profile organizes data on price, time, and volume to help traders anticipate market behavior. We have types of elements (metals, non-metals) and combinations of elements, and we have types of markets and combinations of markets: knowing type provides invaluable clues as to likely behavior in various situations.

It's rare that I don't have some reservations about a book. This one is clearly written, straightforward, and doesn't get away from its topic. There's no fluff, and (mercifully) there's no chest-beating self-congratulations from authors who try to make themselves out to be gurus. Markets in Profile is simply a guide for putting theory into practice at different time frames. That, in itself, is unique in the trading literature.

Detecting Participation in Breakout Moves: A Best Practice in Trading

From Brett: Our next Best Practice post comes to us from Aiva_Trader, who examined the last five months of data in the NASDAQ 100 Index (QQQQ) to establish the odds of continued upward movement once QQQQ breaks above its current day's high. Specifically, the study broke down occasions in which fewer than 30% of NDX stocks were participating in the move (i.e., not surpassing their current day's highs) vs. those occasions in which more than 70% of NDX were also participating in the breakout. Here are the findings:

From Aiva_Trader:

When QQQQ breaks its current day high (all occasions):

The close 1 minute later is up by an average of .026% and closed up 68% of the time.
The close 5 minutes later is up by an average of .030% and closed up 59% of the time.
The close 15 minutes later is up by an average of .043% and closed up 56% of the time.

When QQQQ breaks its current day high and fewer than 30% of index stocks participate:

The close 1 minute later is up by an average of .002% and closed up 68% of the time.
The close 5 minutes later is down by an average of -.028% and closed up 44% of the time.
The close 15 minutes later is down by an average of -.05% and closed up 39% of the time.

When QQQQ breaks its current day high and more than 70% of index stocks participate:

The close 1 minute later is up by an average of .05% and closed up 71% of the time.
The close 5 minutes later is up by an average of .05% and closed up 62.5% of the time.
The close 15 minutes later is up by an average of .07% and closed up 58% of the time.


From Brett: These results fit well with my own research. Breakout moves are more likely to reverse when a large proportion of sectors and individual stocks are not participating in the breakout. This helps to differentiate a market that is topping out (becoming more selective as it approaches highs) from a market that is breaking out (ready to establish fresh new highs). I suspect Aiva_Trader's observations could form a core concept for a trading system, particularly if the signals were filtered by such criteria as whether or not the market was overbought at the time of new highs, time of day of the breakout move, etc. It would also be fruitful to examine holding periods from 30-90 minutes out. My own recent research found that fading new 60 minute lows in SPY and IWM has been profitable at those holding periods. Many thanks to Aiva_Trader for the valuable observation in the NASDAQ market. Knowing the odds that a market has gone your way in the recent past is, for me, a best practice--and a helpful aid in avoiding bad trades.

Friday, February 23, 2007

Friday, February 23rd Morning Market Comments

9:27 AM - Clearly we didn't get the buying interest; once again stopped out. That TICK distribution is all important, and in retrospect I was bucking it and holding too long onto a strength scenario. Food for review and learning this weekend! Have a great weekend; wrap up on the Weblog tonite.

9:19 AM - Well, we'll either hit my target or my stop. I'd be stopped out on new lows for the day and the initial target would be that average price range for ES.

9:17 AM - Once again, I tried a little buying as we failed to make lows on the recent selling, but stops are tight and I'll need to see solid buying coming in. My basic idea is to buy as long as we stay above yesterday's lows; I'm not impressed with the breadth of selling today relative to yesterday, so have been leaning toward buying weakness. We'll see if that works out. The key is to keep position sizing and stops reasonable to survive being wrong!

9:10 AM - Note semi and NQ strength here.

9:05 AM - Need to see more strength in ER2 and TICK and ability to hold lows before retaking long side. Looking at bounces as potential selling areas, particularly if TICK and volume are weak on the bounces.

9:02 AM - Stopped out as ER2 and TICK weakened. Watching those bounces carefully now.

8:55 AM - I'm going to have to wrap up shortly, but my basic strategy will be to buy pullbacks in the TICK as long as those occur at higher TICK lows and price stays above its lows. Otherwise I'm out.

8:52 AM - The nice thing about entering positions small and then adding to them if conditions are right is that you can more easily ride initial shakeouts. We need to see a positive shift in the TICK and we need to see ER2 strength, or I'll be out of that initial position quickly.

8:48 AM - Bought a little ES on that last drop, but will have tight stops; seeing if support from yesterday PM will hold.

8:42 AM - Note some relative strength in the semis early on; keep an eye open.

8:38 AM - I'm looking to see if we can put in a bottom and get selling to dry up above the Thurs afternoon lows. That would make me a buyer for a move back to that average price we've been oscillating around. If we break those lows, esp on strong volume, I'm in the mode of selling TICK bounces. During early trade, I try to frame these what-if scenarios to guide my strategy in advance. That helps me pounce more quickly if opportunity arises.

8:35 AM - Volume pretty tepid so far in ES; nothing to write home about, and nothing to suggest thus far an active institutional presence for a trending trade. Volume skewed toward the hitting of bids; weakness in ER2 has led ES in opening minutes.

8:32 AM - Note that we've already broken overnight lows in NQ and ER2. Need to see them acting stronger before I'd take the long side.

8:19 AM - So far we're looking at 1459.25 as near-term resistance in the pre-opening market and 1456.50 as near-term support. I was impressed, given the intensity of selling, that the market held its AM lows during the afternoon. I'm looking for those afternoon lows to hold in today's trade. If they don't, that clearly represents fresh selling interest and would be a bearish indication. If they do hold, I'd expect us to take out the pre-opening highs and the afternoon high of 1459.75 as initial targets. Morning weakness and afternoon strength has been a common pattern in the past several weeks. This suggests that selling is finding buyers: there's been an underlying bid to this market. Note that semis were strong yesterday; let's see if this is something more than a one-day wonder. This is often a leading sector. Note also that we held above Wednesday's lows in ER2 and NQ, even as we made a day-over-day low in ES. That also suggests some underlying strength to the broad market. Back after the open.

8:09 AM CT - Good Friday morning to you. This will be an abbreviated morning session for my blogging, as I have a 9:30 AM appt. No economic releases scheduled for today; that, on top of a Friday trade, may keep us in the recent multi-day trading range, as prices have pretty much oscillated around the 1458-1459 level in the ES futures for over a week. In addition to the day's pivots, the Weblog also summarizes an intraday trading pattern that has worked well over the last few weeks; do check out. Many thanks to those who participated in my online chat session hosted by John Forman; the audio archive is now up. My most recent blog post summarizes a few of the points from that session. More in a few.

Five Guiding Principles of Trading Psychology

When I recently participated in an online chat presentation for John Forman, I assembled my ideas into ten basic principles that have guided my thinking about the psychology of traders and the psychology of markets. In the very near future, if my testing continues to be promising, I hope to present a market indicator for swing traders that rests firmly upon these principles. Stay tuned! In the interim, here are the five principles that pertain specifically to trading psychology. Next up will be five principles for trading the markets.

Principle #1: Trading is a performance activity - This is the core idea behind my most recent book. Like the playing of a concert instrument or the playing of a sport, trading entails the application of knowledge and skills to real time performances. Success at trading, as with other performances, depends upon a developmental process in which intensive, structured practice and experience over an extended time yield competence and expertise. Many trading problems are attributable to attempts to succeed at trading prior to undergoing this learning process. My research suggests that professional traders account for well over three-quarters of all share and futures contract volume. It is impossible to sustain success against these professionals without honing one's performance--and by making sure that you don't lose your capital in the learning process. Confidence in one's trading comes from the mastery conferred by one's learning and development, not from psychological exercises or insights.

Principle #2: Success in trading is a function of talents and skills - Trading, in this sense, is no different from chess, Olympic events, or acting. Inborn abilities (talents) and developed competencies (skills) determine one's level of success. From rock bands to ballet dancers and golfers, only a small percentage of participants in any performance activity are good enough to sustain a living from their performances. The key to success is finding a seamless fit between one's talents/skills and the specific opportunities available in a performance field. For traders, this means finding a superior fit between your abilities and the specific markets and strategies you will be trading. Many performance problems are the result of a suboptimal fit between what the trader is good at and how the trader is trading.

Principle #3: The core skill of trading is pattern recognition - Whether the trader is visually inspecting charts or analyzing signals statistically, pattern recognition lies at the heart of trading. The trader is trying to identify shifts in demand and supply in real time and is responding to patterns that are indicative of such shifts. Most of the different approaches to trading--technical and fundamental analysis, cycles, econometrics, quantitative historical analysis, Market Profile--are simply methods for conceptualizing patterns at different time frames. Traders will benefit most from those methods that fit well with their cognitive styles and strengths. A person adept at visual processing, with superior visual memory, might benefit from the use of charts in framing patterns. Someone who is highly analytical might benefit from statistical studies and mechanical signals.

Principle #4: Much pattern recognition is based on implicit learning - Implicit learning occurs when people are repeatedly exposed to complex patterns and eventually internalize those, even though they cannot verbalize the rules underlying those patterns. This is how children learn language and grammar, and it is how we learn to navigate our way through complex social interactions. Implicit learning manifests itself as a "feel" for a performance activity and facilitates a rapidity of pattern recognition that would not be possible through ordinary analysis. Even system developers, who rely upon explicit signals for trading, report that their frequent exposure to data gives them a feel for which variables will be promising and which will not during their testing. Research tells us that implicit learning only occurs after we have undergone thousands of learning trials. This is why trading competence--like competence at other performance activities such as piloting a fighter jet and chess--requires considerable practice and exposure to realistic scenarios. Without such immersive exposure, traders never truly internalize the patterns in their markets and time frames.

Principle #5: Emotional, cognitive, and physical factors disrupt access to patterns we have acquired implicitly - Once a performer has developed skills and moved along the path toward competence and expertise, psychology becomes important in sustaining consistency of performance. Many performance disruptions are caused when shifts in our cognitive, emotional, and/or physical states obscure the felt tendencies and intuitions that lie at the heart of implicit learning. This most commonly occurs as a result of performance anxiety--our fears about the outcome of our performance interfere with the access to the knowledge and skills needed to facilitate that performance. Such performance disruptions also commonly occur when traders trade positions that are too large for their accounts and/or do not maintain sound risk management with their positions. The large P/L swings cause shifts in emotional states that interfere with the (implicit) processing of market data. Cognitive, behavioral, and biofeedback methods can be very useful in teaching traders skills for maintaining the "Yoda state" of calm concentration needed to access implicit knowledge.

The most important question I can ask an aspiring trader is: Are you engaged in a structured training process? Education--simply reading articles in magazines, websites, blogs, and books--is important, but it is not training. Training is the systematic work on oneself to build skills and hone performance. It requires constant feedback about your performance--what is working and what isn't--and it requires a steady process of drilling skills until they become automatic. No amount of talking with a coach or counselor will substitute for the training process: not in trading, not in athletics, and not in the dramatic arts. Training yourself to proficiency is the path to a positive psychology.

Thursday, February 22, 2007

Profiting From Short-Term Momentum: A Best Practice in Trading

From Brett: Today's best practice comes from reader Steve, who generously shares one of his setups in the Russell index market. His methodology is based on trading in the direction of short-term momentum, as long as this is aligned within a greater trend. Recall that Lord Tedders and Henry Carstens recently offered their perspectives on testing trading ideas. This would be an excellent core idea to test out and tweak with filters, stops, etc. Many thanks to Steve for passing this along. My reaction comments follow at the end of his post.

From Steve:

I call this best practice "Trading Short Term Momentum"...

I suggest that for futures trading, a setup geared to trading short term momentum can be profitable.

My chart setup for the Mini-Russell (ER contract) would include two (2) simple moving averages

1. Three (3) period sma with two (2) period offset (+2 offset)
2. Eighty (80) period sma

System rules

No trades within the first half hour.

On a close above the three (3) period sma, enter long on a stop at the next open

On a close below the three (3) period sma, enter short on a stop at the next open

Long only above the eighty (80) period sma (Enter long positions only on a close above the 80 sma)

Short only below the eighty (80) period sma (Enter short positions only on a close below the 80 sma)

Stop loss of one (1) Russell point

Min profit target of two (2) Russell points
(Once price "touches" two (2) points profit, trader must close at least 1 contract)

Maximum three (3) losing trades/session

Maximum three (3) consecutive losing days/week...stop trading for 7 trading days.

Minimum position size two (2) contracts

Chart time frames 3-7 minute candles/bars


From Brett: There are several things I like about Steve's idea. First, he is selective. He will only take trades after the first half-hour, and he will only take trades when the 3 SMA and 80 SMA are aligned. Note that he is using 3-7 minute bars. That means that the 80 period SMA will put him in trades in which the market has been strong or weak for a day or so. My best hunch is that this would work best in an environment of above average volume/volatility. I also like the money management--taking quick profits with half the position--and I like the discipline of limiting the number of losing trades per day and week to avoid catastrophic drawdowns during periods in which there is poor follow through on short-term trends. All in all, it's a great example of what I've called "rule-governed trading". By formulating rules for good trading practice, Steve is able to capitalize on short moves that go with the trend. By avoiding the first half hour of trading (during which we sometimes get economic releases), he lets the trend show itself before he commits capital. Very nice window on some best trading practices; thanks, Steve.

Thursday, February 22nd Morning Market Comments

9:44 AM - Some days you see the market well, other days forget about it! Today was one of the better days. I hope these sessions help you frame hypotheses about how the market is trading by focusing on the balance of buying and selling within defined ranges and in the context of concrete price targets. These ways of thinking can be of help regardless of the concrete entry signals you might happen to use. Once you see how to identify ranges, read buying/selling pressure, and identify how large traders are leaning, the rest is practice and skill development. I'll be doing my phone interview tonite (link below); wrap up tonite on the Weblog.

9:37 AM - I exited a bit early; we did touch that 1458.75 target, but I wasn't seeing big sellers hit the market the way they had done earlier. I'd be on the lookout for putting in a bottom, which would establish a new range bound market. Wrap up in a bit.

9:35 AM - Clear downward shift of TICK over the course of the AM. Those TICK bounces are providing excellent entries.

9:33 AM - Took a little profit here. Watching to see if selling dries up, but ready to get back in if TICK expands to downside.

9:23 AM - We barely took out the overnight low; 1458.75 pivot is next downside target.

9:19 AM - Note how volume really picked up on the decline as the longs liquidated. Excellent market sequence. Textbook. Those TICK bounces are great short entries once sellers start to dominate.

9:15 AM - Great example of a false breakout and reversal. Worth reviewing at the end of trading.

9:12 AM - My leaning is to sell TICK bounces that come at successively lower price levels, esp in ER2. Note how we're back into the trading ranges for the 3 indices.

9:09 AM - In general, I try to follow the large traders. Over 30% of all stock market volume (and futures) is due to program trading. So when I saw a buy program hit the broad market--and especially ER2--with an upside breakout in TICK, I knew that the large traders had shown their hand, at least in the short run. I waited for the first pullback following the buy program to make sure no huge sellers slammed us down, and that was my entry. R1 was my target and when we didn't move promptly toward R2, I took what the market gave me. Very typical trade for me, combining awareness of large traders, TICK, price ranges, and execution on pullbacks from the anticipated move.

9:04 AM - Sure enough: ER2 back into its overnite range.

9:01 AM - Keep an eye on ER2. It led the upside, but may also lead the downside if buying is not sustained.

8:58 AM - Note that we've broken above our trading range. We want to see buyers sustain the rally, however, and *keep* the market above that range. Otherwise, I'll be looking to fade the breakout.

8:55 AM - Took some profits here after we surpassed R1; I'll be explaining the trade in a bit.

8:48 AM - I went long on the pullback following the program buy; will explain later.

8:45 AM - Program buy

8:41 AM - Reminiscent of yesterday's trade, with lots of little runs and reversals; very choppy. Makes it dangerous to buy strength or sell weakness. I become very patient in these kinds of markets. Until I see a clear trend of buying or selling interest, I stay on the sidelines.

8:36 AM - 1459.75 represents the overnight lows; I'd want to see that level hold on selling before I buy. We're very near overnight lows in ER2. I need to see some selling dry up in ER2 before I'd take the long side for a breakout trade.

8:35 AM - Moderately high volume; decent participation in ES, with slant toward early buying. Strong semis; watching ER2 closely for selling. If we're going to take out the upside, I want to see the indices and sectors in gear.

8:27 AM - I've been contacted by a couple of professional trading organizations to work with their traders starting in March, so I will be wrapping up these daily morning market updates at the end of February. It's been great being able to do these; February has been a nice free time for me, and it's given me a lot of time to work on my own trading. If you have questions about any of the topics touched upon in the morning sessions, feel free to leave a comment on the blog or email me directly. During March I'll conduct occasional morning sessions as my travel and work schedule permit and will announce those in advance. Also in March, I'll begin work on a video training project for traders that will tie together many of the themes from these morning sessions. Many thanks for your interest and support.

8:22 AM - After the close today at 3:30 CT (4:30 Eastern time), I'll be doing a free Web talk with John Forman, author of The Essentials of Trading. Here's the link for registration. I'll be taking calls/questions and talking about the psychology of traders and trading. Some of my talk will touch upon the theme of the most recent blog post on somatic markers. It's a fascinating topic and area of research. Also, later today I'll put out the group email to help programmers and traders get connected. Do email me if you are interested and haven't already gotten in touch. Thanks!

8:18 AM CT - Good Thursday morning! We're knocking on the door of bull market highs in the equity indices; a bit of a pop in interest rates. Initial claims came in a tad higher than estimates; no major move in the dollar vs. Euro. We're hovering near lows in the Yen. As noted in the Weblog, we're operating in a five-day trading range, so the big issue is whether these current price highs attract sufficient buying interest to launch us to new levels of valuation. If we see a drying up of buying in early trade, a logical price target would be the Wednesday pivot level and volume-weighted average price, as posted to the Weblog. If we see firmness in the NYSE TICK, I will expect a move to new price highs. Back in a few.

The Role of Somatic Markers in Trading Decisions

I think this is one of my more important articles: a perspective on trading psychology from the field of cognitive neuroscience.

In past posts, I have emphasized the importance of controlling the body's level of arousal as a fundamental trading skill, because so many psychological disruptions of trading are state-dependent. There are cognitive and behavioral methods for achieving such control; these include simple exercises that traders can perform on their own.

There is far more to trading psychology, however, than simply taming emotions. The idea that emotions are the root of trading problems--and that successful traders eliminate their emotions--is one of the great myths in the field. Indeed, research in cognitive neuroscience--particularly the work pioneered by Antonio Damasio--suggests that emotion is absolutely crucial in decision making.

Damasio's seminal contribution in this area of research is his somatic marker hypothesis. An overview of this hypothesis and relevant research is available in this review article. According to that article, the gist of the somatic marker hypothesis is that:

"Structures in ventromedial prefrontal cortex provide the substrate for learning an association between certain classes of complex situation, on the one hand, and the type of bioregulatory state (including emotional state) usually associated with that class of situation in past individual experience. The ventromedial sector holds linkages between the facts that compose a given situation, and the emotion previously paired with it in an individual's contingent experience. The linkages are ‘dispositional’ in the sense that they do not hold the representation of the facts or of the emotional state explicitly, but hold rather the potential to reactivate an emotion by acting on the appropriate cortical or subcortical structures..."

Let's paraphrase that in plain English! A section of the brain's frontal cortex is responsible for storing and processing associations between situations and the emotions that had been activated by those situations. When we store these associations, we are not storing conscious memories. Instead, we create links between situations and how those situations make us feel. Those linkages are "dispositional", which means that they lead us to act in one way or another.

Here's a simple example: I have many fond memories of spending time in cafes in the U.S. and Europe. If I pass a cafe on the streets of Seattle and smell the aroma of coffee, that perception activates the feeling-states associated with my prior experience. Thanks to those feeling states, I am disposed to enter the cafe and spend some time there. My decision to enter the establishment is not a simple, rational process of toting up pros and cons. Rather, the feeling state mixed with the perception of the cafe--a link embedded in the ventromedial prefrontal cortex--leads to the decision.

So it is with all decisions, Damasio argues. Feelings--from explicit emotions to felt body states--are joined with perceptions to guide our behavior.

The research cited in the review article suggests that, when people experience damage to their ventromedial prefrontal cortex, they are no longer able to make sound decisions. One especially fascinating lines of research involves a risk-taking game that requires decision-making similar to that involved in trading. This "gambling task" involves choosing cards from four decks. Two of the decks have higher payoffs on wins, but also larger penalties for losses, so that the overall returns are negative. Two of the decks have smaller payoffs for wins, but smaller penalties, and an overall positive expectation of gain.

At the start of the gambling task, subjects don't know the odds for the piles. Over time, however, normal subjects learn to prefer the piles with the positive odds for winning. When they are hooked up to biofeedback equipment, they display clear skin conductance responses ahead of making their choices. These stress-related feeling states appear to help guide the choices of piles. Conversely, subjects with damage to their ventromedial prefrontal cortices do not display such anticipatory responses. They lack the feelings that guide good decision-making. As a result, they display a sustained preference for the high-return, high-penalty, negative expectation decks. Without access to their encoded feeling states, they cannot make good decisions under conditions of risk and uncertainty.

Now here's the really interesting part. Normal adults who describe themselves as risk-takers do not have damage to their ventromedial prefrontal cortex regions, but they do display a similar tendency to persistently choose cards from the risky, low-return piles. When connected to biofeedback equipment, the risk-takers *do* display emotional signals prior to making their choices, not unlike the normal subjects. They tend, however, to override those signals with their explicit thought processes. The brain-damaged subjects, on the other hand, never receive the emotional signals from the ventromedial prefrontal cortex in the first place.

The implications for trading are significant. The idea that successful traders overcome or eliminate their emotions is completely off-base. If that were to truly happen, the trader would behave like a brain-damaged patient. Rather, it is the overriding of emotional signals and signals from felt bodily experience that facilitates poor decision-making. This overriding can occur because of anxiety, greed, or myriad cognitive rationalizations. The implication, however, is that--at some level--experienced traders receive valuable physical and emotional signals to guide their decision-making. It is the temporary lack of access to these signals that leads traders to behave like the risk-takers in Damasio's experiments.

In a recent post, I emphasized the value of biofeedback in achieving self control. It may well be, however, that the greatest value of such disciplines as meditation, relaxation training, and biofeedback is to help people maintain a clear mind so that they do not override and obscure the somatic markers that are necessary for sound decision making. I strongly suspect that traders would be better served by disciplines that enable them to be accurate observers of their feelings than prescriptions for squelching those feelings. Choice lies at the heart of trading, and if Damasio is correct, our felt experience is a necessary substrate of choice. The somatic markers are there; it's simply a matter of keeping our access open to them, even as we're surrounded by risk, reward, and uncertainty.

Wednesday, February 21, 2007

Designing a Robust Mechanical Trading Strategy: A Best Practice in Trading

From Brett: This best practice post comes to us from Edward Heming, who is the author of the Lord Tedders trading blog. He discusses a few aspects of developing a reliable mechanical trading strategy and also covers the pros and cons of mechanical trading. Note that Henry Carstens has also made available a series of articles on the topic of developing trading systems. What I most like about Lord Tedders' article is the insight that researching system ideas is a great way to gain a feel for the market. For that reason, it can even benefit the discretionary trader. Those wanting to gain some of the benefits of system testing without the challenges of programming can look to the Odds Maker program developed by Trade Ideas or can follow the advice of Bonnie Lee Hill and utilize the drop-down menu testing platform available through Ensign Software. With such tools, it is easier than ever to truly determine if your ideas are providing you with a performance edge. Thanks to Edward for the insightful post.


One of the questions that I often get asked about strategy design is, “how do you design a robust mechanical trading strategy?”

To understand how to build a robust mechanical strategy it is important to understand what a robust mechanical strategy is. A mechanical strategy is simply a quantified decision stream that leads either a “trading robot” or the trader himself to determine position size, entries, exits and stops all in a completely hands off fashion – in other words if you have a working mechanical system your input is not needed (or if so to a very limited degree). Additionally, for a mechanical strategy to be robust, it must capitalize on a “trading edge”. This can be anything from a statistical edge (trending) to an execution edge (arbitrage). Furthermore, this strategy must hold up over an extensive period of trades historically (at least several hundred) and must hold up in future trading (which can be simulated).

A mechanical system has several advantages that discretionary traders do not, such as the ability to perform quantitative and data mining analysis quickly and over extended historical periods. Additionally, mechanical systems can alleviate some of the emotional distress that accompanies discretionary trading – particularly among new traders.

However, it is important to recognize that mechanical trading has several disadvantages as well. The first being that you must be able to quantify each and every trading decision that the system will make, secondly the mechanical system will have to be periodically adjusted (just like a discretionary trader adjusts their methods) either through inherent adaptivity, optimization, or diversification. Lastly, mechanical systems only work if one puts in the tremendous amount of time and effort required to program, test, debug, and continually adjust it.

To design any mechanical strategy it is important to consider three things before anything else: 1) your objective for that system, 2) your market, 3) your timeframe. Once you have determined this, it is easy to find your essential methodology because there are only 4 ways to trade any market: 1) trend trading, 2) momentum trading, 3) reversion to the mean trading, 4) and fundamental trading. Once you have determined your objective, market, timeframe and method you are ready to attempt to put together your first strategy. Many of you are probably thinking at this point, “what if I don’t know any of that stuff?”

If you are already an experienced discretionary trader this should not prove to be overly difficult. However, if you do not have extensive experience you will have to find a method that works. This method can be as simple as a moving average cross long/short to as complicated as a continually adjusting collaborative neural network that is genetically re-optimized daily. The very best way for the inexperienced trader to build a new system is to test ideas. This can be done in two ways – visually or programmatically. For someone without extensive programming experience, the best would be to start with what I call “candle by candle” back testing. This is performed by taking an idea (such as a moving average crossover) and testing it with historical data on the given market and time frame by moving your charts forward from the past into the future and trading the way the system would – without future knowledge of the markets.

This method is how I tested my first ten “strategies”, four of which I still continue to trade today (including two that were designed by Phil McGrew which I tested using this method and still trade today). However, I had to test nearly fifty or sixty ideas to get down to those ten strategies that work, and finally refine the process until I had found four of those ten systems that I found tradable. To give you an example of how time consuming this process is, I tested these ten strategies extensively often looking at over 2 years of 15 minute bars and “executing” hundreds of trades. I spent nearly 700 real hours doing this testing (and I’m pretty quick with a chart and excel). It sounds like a lot of work right? Well it was, but it also gave me a feel for those markets that is nearly as good as having traded those markets in real time.

After doing this for some time, I felt that there had to be a more effective way to test ideas. And there is – programmatic testing. Programmatic testing again can be very easy – a simple moving average cross is a simple thing to program in nearly any programming language. However, the difficulties that can destroy the beginning programmatic trader are nearly endless. Many popular trading packages do not trace your equity position tick by tick, rather it is tracked bar by bar (and if you’re trading daily bars you can imagine the problems). Also, ideas that I had tested extensively by hand sometimes were difficult to program. I have had so many experiences where I miscoded a critical concept (even by a slight degree) and this ended up giving drastically different results than my hand testing. Without the knowledge that it was the code that was incorrect, I might have falsely dismissed many trading ideas that were in fact valid.

Additionally, at this level of programmatic trading it is very important to consider factors of minimizing inputs (degrees of freedom) and utilizing flexible inputs. An example of this would be to utilize a 3 ATR stop instead of a 60 pip stop so that as the prices and volatility of the market fluctuate your stop is not being taken out because of random noise. Other ways that you can improve the robustness of your strategy include utilizing realistic fills and commissions and ensuring that your limit orders would have actually been filled (this is not as easy to test in some software as it should be).

Optimization is another useful tool to consider at this point in your strategy testing career. This is a powerful but two edged sword. Utilization of genetic algorithms and similar “hill climbing” techniques are a common way to ensure that your optimization does not give you a single point anomaly, but rather that there are similar input values surrounding your inputs that give similar equity graphs. Walk forward testing is another useful tool that can help you achieve realistic results and see for yourself whether a strategy would have been successful on data that was not optimized (similar to the future).

Going further into programmatic trading, after having experienced many pitfalls, I feel that I ought to be able to test more than one idea at a time. In fact, ideally I would like to test many ideas, over multiple time frames and multiple markets. Right now this is the work that I am involved in designing and I feel that this will help me analyze the markets with the speed and precision that will take my trading to the next level. This is the arena of the best strategy designers, where statistical data mining, market analysis, timeframe analysis, technical analysis, fundamental analysis, and money management are combined with realistic evolutionary testing into a single package.

As you can see, advanced programmatic testing and trading is a complex arena. I myself am still learning and by no means consider myself an expert. The good news is that successful robust mechanical strategy creation and implementation can be done in as simple or as complex a manner as you choose. After all, the very simple strategies tested and/or designed with candle by candle backtesting are still a cornerstone of my trading methodology.

From Brett: Notice Edward's advice: start small, keep it doable, and then build your skills. Your best ideas will come from intensive observation, but some of the best ideas are the simplest and most straightforward. I've recently posted a call for traders and programmers who would like to collaborate; this could be one promising way of getting started!

Wednesday, February 21st Morning Market Comments

9:52 AM - Back to the mode of looking at bounces for possible selling. I will be on the phone conferencing with traders so need to wrap up. Two important items for today: patience--let the market show its hand. We were getting a drying up of selling and only then did it make sense to go long. I nibbled long, the buying began to wane, and then we retraced. Second important item: don't let the winners turn into losers. I thought we had more upside potential, it didn't emerge, and I got out. That avoided significant losses. It's been a whippy trading session--we're not so far from where we opened. The important thing is to watch for those shifts in buying/selling to gauge likelihood of followthrough on moves. Have a great rest of the day; wrap up tonite on the Weblog.

9:41 AM - You can see how the shift from absence of buyers to the emergence of sellers occurred. That's why it's so important to scratch positions; not hold and turn gains into losses. Clearly market momentum has shifted back to sellers.

9:33 AM - Waited, waited, and buying never came. Scratched out.

9:12 AM - I continue to buy those pullbacks, but need to see stronger TICK to stay long.

9:05 AM - I'm nibbling long on TICK pullbacks that stay above the prior TICK and price lows.

8:59 AM - I'm considering the market's ability to hold above its Tuesday lows a positive for bulls here. Need to see expansion of buying (TICK), however, to return to the Tuesday highs.

8:55 AM - A lot of oscillation ahead of LEI, TICK distribution weak, but not extreme in either direction. Volume moderately high. My leaning is to be patient and let the market show its hand by either breaking out of its AM range or failing to sustain such a breakout.

8:49 AM - Need to stay above overnight lows in ER2 to get me buying.

8:45 AM - I'm actually now leaning toward buying if we get pullbacks that stay above the overnight lows. Basically letting the market show its hand.

8:38 AM - Volume is moderately high, got some bounce early. Need to see that dry up before I would go short.

8:26 AM - Oh yes. We have Leading Economic Indicators reported at 9:00 AM CT and FOMC minutes released at 13:00 CT. Clearly the equity index markets haven't liked the higher than expected core CPI; a strong LEI might also stoke inflation fears.

8:22 AM CT - Well, we moved above the 3-day trading range on Tuesday after overcoming early weakness, but now we're right back into that range on the response to the CPI data. Interest rates are up, dollar stronger vs. Euro, and--surprise--the dollar is stronger vs. the Yen despite an interest rate hike by BOJ. Not much in that response that would threaten the carry trade in the near term! Longer term, the BOJ may be forced into additional action if selling in the Yen persists, especially vs. the Euro. Check out this morning's post regarding persistent market strength; there's a very interesting trading pattern that's worked out 10 of the last 10 occasions over the past 3 weeks. Pivot-based price targets and some evidence of continued selectivity in the recent market runup are posted to the Weblog. My leaning will be to sell bounces that stay below yesterday's pivot, with an eye toward S1 and Tuesday's lows as downside targets. I am, however, also keeping an eye on the pattern mentioned in the blog post, in which early selling has been an opportunity to buy. We saw that in spades yesterday. Back after the open. I have some phone conferences with traders this AM, so my posts will be less frequent and focused more on general market conditions and ideas than specific trades that I'm placing.

Persistent Strong Closes: The Personality of This Stock Market

Here's a little piece of trivia to chew on: Out of the last 15 trading sessions, we've traded below the prior day's closing price 10 times. On 13 of those 15 occasions, however, we've closed closer to the high of the day than to the low. In other words, we've typically had some degree of weakness during the market day, but--as on Tuesday--buyers have tended to step in and use such weakness as an opportunity to accumulate stocks.

How rare is this? Going back to 1990 (N = 4305 trading days), I could only find 34 occasions in which 13 or more days out of 15 finished closer to their highs than to their lows. Such persistent strong closes are rare.

One stretch of persistent daily strength occurred between October 20 and October 31, 2006. We had one occasion on June 6, 2005, two on February 15 and 16, 2005, and one on September 4, 2003. We don't see another such string until September, 1996. Sixteen of the occasions occurred between 1991 and 1993; none before that. If you recall your market history, you'll see that, over this period, strings of 13 or more days out of 15 closing nearer to their highs than lows have occurred only during bull markets. They have not occurred during 2000-2002, 1998, 1994, or 1990, for example. Indeed, we might even conjecture that part of what makes a bull market is not an absence of daily weakness, but the tendency for investors to pounce upon such weakness as value and an opportunity to buy.

This pattern of pouncing on weakness as an opportunity to buy has been occurring at even very short time frames. I took a look at the Odds Maker program from Trade Ideas and asked the question: What has happened over the past three weeks when you have bought a breakdown from the opening 15-minute range in the S&P 500 Index (SPY)? Such a break below the range of the first 15 minutes of trade has occurred on 10 of those days; buying those occasions and holding for the next 30 minutes resulted in 10 winners. The average win size was the equivalent of a little over 1 full S&P point, so that the trade would have netted about 11 ES points over that period.

So what has happened after periods of persistent strength when we've had 13 or more days out of 15 in which we've closed closer to the day's high than to the low? Ten days later in the S&P 500 Index, we've averaged a gain of 1.17% (23 up, 11 down). That is much stronger than the average ten-day gain of .37% (2508 up, 1797 down) for the entire sample. It would appear in this case that strength begets strength. One of the reasons for this is that the occasions of persistent strong closes have tended to cluster. We had a string of eight occasions in October, 2006; four in August/September, 1993; four in January, 1992; and six in January/February, 1991. Because such strings are common, we cannot assume that, because the market has been persistently strong on a daily basis, it is now due for weakness.

Every market, like every person, has a personality: a set of traits that define behavioral tendencies. This has been one facet of the current market's personality: persistence of strength. This investigation suggests that such persistence is rare, but--at least in the short run--it, too, tends to persist.