Monday, February 05, 2007

Making Each Day a Learning Experience: Trading Lessons From the Morning Sessions

I've now been conducting the morning sessions, tracking the equity index markets in real time, for a while. My goal is to make every trading session a lesson that will move me forward in my craft. I decided to go back and pull out a few of the more important lessons from those morning sessions. These cement important trading ideas for me and, hopefully, for you as well as we start a new trading week:

Sunday, January 7th, 2007: In my personal site, I try to summarize the market data I'm looking at each day and how I'm putting the data together into an initial framework to start the day. This site summarizes some of the research I conduct to identify possible trading edges, mostly from 1-5 days out. The reality of trading, however, is that such preparation only gives you initial hypotheses and plans. The real skill of trading comes in when markets open and you have to analyze shifting patterns of supply and demand as they emerge. In my own trading, I follow several variables very closely (one-minute data):

* Whether volume is higher, lower, or equal to average volume for that time of day;
* How the most volatile market sectors (small caps, NASDAQ, semiconductors) are trading relative to the large cap indices;
* How interest rates, currencies, gold, and oil are trading during equity trading hours;
* NYSE TICK (number of stocks trading at offer vs. bid) and shifts in the distribution of the TICK;
* Volume of contracts executed at the market bid vs. ask for the ES futures;
* Whether a majority of sectors are participating in moves in the ES futures (I look at Spyder sector ETFs for much of this info);
* Value area (Market Profile) from the previous day's trade and how volume expands or contracts as we trade outside that area;
* Levels of support and resistance from the previous day as well as the current day, to identify potential trading ranges and levels we're likely to test.

If you were to watch me trade, you'd see me continuously shifting from one window/screen to another, monitoring these variables. At some point a pattern becomes clear and I get an idea of shifting demand/supply that will lead us to test a particular market level. That becomes the basis for a trade idea, particularly if it is in line with my prior research. I average 2 trades per day, mostly in the AM, and averaging 20 minutes in holding time.

Monday, January 8th, 2007: The main thing to take away from the AM trading session is the importance of flexible thinking. We started off with some research that suggested we were likely to take out the Friday lows. As the selling progressed early in the morning, however, it was apparent that many sectors were not participating. That suggested that it would be a mistake to chase those lows and, indeed, made sense to watch for reversals. Perhaps the most important mental shift was from "downtrend" thinking to "rangebound" thinking, with the AM lows and the 1420 resistance area forming the range. Having that range in mind enabled me to get on board to the long side when heavy selling could not bring us to new daily lows, but the range also alerted me to be aggressive in taking profits when the buying ran into a wall of sellers.

Tuesday, January 9th, 2007: That's a good lesson for the day: when you see the TICK hitting negative values, but the index can't make new price lows, it's generally an early sign that sellers can't move the market any further: that buyers are finding value at those levels. That doesn't necessarily mean that buyers will enter the market in force, but it's usually a good bet that you'll get enough buying to move the index back toward a prior high for a short term trade.

Monday, January 15th, 2007: It's not unusual to see a breakout move begin in one of the more volatile indices before strong buying shows up in the Dow or the S&P 500. I've found that helpful on occasions for timing. I also like to follow the DAX as a potential leader of the S&P 500 early in the AM. Most helpful of all is filtering the Market Delta charts so that they only post trades of a certain size or greater. You then can see how many large traders are in the market and whether they're predominantly buying or selling. That is *very* useful in handicapping the odds of reaching a pivot support or resistance level.

Tuesday, January 16th, 2007: Notice that measures such as TICK and volume at bid/offer tell us about supply and demand, but volume levels relative to average for the time of day tell us about volatility: how much movement we can expect from given supply or demand. Because we don't have strong buying or selling in the TICK (the Adjusted TICK is negative on the day, but not at extreme levels) and we don't have above average volume, I don't expect a large move on the day and, indeed, anticipate more of a rangebound market. The key is monitoring these variables in real time to identify as early as possible what type of market we're likely to be in and which price levels we're likely to hit.

Thursday, January 18th, 2007: I like to cement a lesson from each trading day. Perhaps the takeaway for today is that 85% of all days in ES are *not* inside days. That ratio is well over 90% when we have above avg volume. Once you know that, it's a matter of handicapping the odds of taking out either the prior day's high or low. Once we saw selling in those leading sectors and breaks to new lows, and once the TICK turned down and we got sellers hitting bids in size, we had a nice trade in ES toward Wednesday's lows and S2.

Friday, January 19th, 2007: The AM illustrated a few important things. First is the importance of flexibility. The odds of taking out yesterday's lows in ES were quite high, and that was my initial leaning. When selling dried up early in the AM with only a modestly negative TICK and not many stocks declining relative to advancers, I entertained the reverse hypothesis: that we were not getting selling and that leaning to the long side was the way to go. The second principle illustrated is the importance of patience. It was a choppy morning session, and it was easy to get scared out of a good position. But as long as the overall dynamics of supply and demand were not shifting, patience was indicated. Finally, you can see how a drying up of selling (today relative to yesterday) often precedes an influx of buying. That's a pattern that sets up intraday as well as on a swing basis.

Thursday, January 25th, 2007: Takeaway for today is this: transacting at the bid vs offer, which is what I'm tracking with the NYSE TICK, Dow TICK, Market Delta, etc, is a very short-term measure of trader sentiment. When we don't see buying sentiment following a nice up day, it makes sense to think about a transition to range bound trade and a reversion to the prior day's average prices. An absence of buyers (sellers) often precedes an influx of sellers (buyers), but it pays to wait for the latter, pick price levels, and take the high percentage trades.

Friday, January 26th, 2007: The lesson I want to stress for today's session is position sizing. We had huge odds of taking out yesterday's lows, creating a very favorable edge. Once selling started appearing in the market, you have to participate with all the size you can muster. You can lose money on 2 or 3 small trades, make money on one high odds trade with size, and wind up the day/week a solid winner. Your position size should reflect your confidence in the trade. Yesterday I mentioned putting small size on when you're feeling out a market. Today it was time to press the advantage. Those odds don't come along every day.

Tuesday, January 30th, 2007: In general, once you identify a candidate morning low, it makes sense to put in an initial position and use pullbacks in the TICK as opportunities to add to the position, as long as the TICK bursts take you to successive price highs. If you can't hit your price target and the buying sentiment (TICK, volume at offer) isn't moving you higher, then you have to entertain the hypothesis that you're in a trading range environment and take what you can from the market.

Wednesday, January 31st, 2007: As long as we can trade 15,000 contracts or so every five min in ES, there should be some movement for the short-term trader. It's when we get below 10K per five min period that things get deadly slow and I stop trading. I've found keeping those volume levels by my side to be helpful. Let's me know when it's worth playing, when it's not. The main thing is whether or not volume is above avg or not for that particular time of day. When well below avg, not worth trading.

Wednesday, December 6th, 2006: I would have loved to have seen a more exciting day for our morning session, but as always we take what the market gives us. My research suggested subnormal returns going forward and my read of the volume and TICK suggested a range bound day. That had me shorting early in the session when we traded above the prior day's average price. That was a very nice trade idea, but it paid out only moderately, which again clued me into a slow, rangebound day. When I saw institutions jump into the market and the TICK distribution go positive, I decided to enter in the direction of the market trend from the last two days as long as TICK stayed healthy. I knew that such a trade was flying in the face of the rangebound thesis, but my risk-reward was pretty good. As it happened, the market did slow down and I had to bail out with a small loss when the TICK no longer looked healthy to me. The main thing for today was using research and market data to figure out what kind of day it was likely to be and then to frame trades patiently rather than chase moves that reverse.

Wednesday, November 22nd, 2006: After each day of trading, I like to evaluate what I did right and wrong and use those to frame goals for the next session(s). What's clear to me from today is that I need to stick with my basic method of exiting positions: entering with enough size that I can take quick profits on one piece but let the other piece breathe and take advantage of a possibly longer move. I was too much in the mindset of "This is pre-holiday; just take what the market gives you". As a result, I didn't benefit as much from the patient, good entries as I should have. The point is to always be learning. Figure out what you do well and extend it. Adjust what you're not doing well. Today I did entries much better than exits.

Tuesday, December 19th, 2006: There's a difference between a good losing trade and a bad losing trade - A good losing trade provides you with information about the market. My initial short position was a good trade, riding the market's weakness in a short-term downtrend. When the trade reversed and took me out with a small loss, that was concrete evidence that buyers were attracted to value below 1430 in the ES. By waiting for the next round of selling in the TICK, I was able to ride this strength for a decent winner when the ES returned to the top of its preopening range. A bad losing trade results from a failure to take all the facts into account. The only information it provides is a heads-up to stay grounded in the market's volume flow before entering a trade. My last trade ignored solid buying in ES, even as the Russell was pulling back. That's not the kind of market weakness that should justify a short position. The large traders were not hitting bids in the most liquid of the indices. By jumping on a trade that had worked for the past two days before checking all the facts, I made a bad losing trade.