Saturday, February 15, 2020

Trading Hypotheses and Trading Ideas

As I'm writing this, it's 6:39 AM on a Sunday morning.  I've just finished an hour of market analysis and prep for the week.  Just a little example of the difference between dreaming about success versus waking up and prepping for success.  Recently I asked a trader to send me a review of his trading.  He explained that he was busy with personal things and would get that to me in a few days.  Right.  I'm still waiting for the review.  If you were to talk with this trader, he would sound convincing about his "passion" for trading.  Too bad he doesn't have a passion for working on his trading.

OK, off the soap box.  Above is an interesting chart of the ES futures, covering roughly the past month of trading.  One unique aspect of the chart is that each bar represents 300,000 contracts traded.  Below the chart, the green and red bars show whether buyers or sellers were dominant during that 300,000 contract period; e.g., whether more transactions were occurring at the current market offer vs. bid price.  The bottom oscillator captures a moving average of the volume at offer/bid data.

Ask any market participant about the stock index futures and they'll tell you that we're in a bull market of historic proportions.  And they wouldn't be wrong.  We've seen quite an upward trend recently.  Still, market cycles can occur within the context of market trends; few trends move higher with no corrections whatsoever.

A worthwhile heuristic I use in framing trading hypotheses is that the number of volume bars representing selling is roughly proportional to the number of bars of subsequent buying.  Another way of saying this is that the volume "trapped" by fading a trend is proportional to the amount of volume that ultimately has to cover their positions.  What is important in this is that we're framing a hypothesis, not jumping to a conclusion.  Before we can have a trading idea--and any "conviction" in that idea--we have to entertain a worthwhile hypothesis.  Worthwhile hypotheses come from looking at fresh market data or the usual data in fresh ways.  Creative perception precedes creative ideas: if we see the same things as other people, we'll likely trade how they trade and what they trade.  In other words, stale perception makes us part of the consensus, the herd.  It's seeing what others don't even look at that gives us hypotheses that can become promising ideas if subsequent order flow supports those hypotheses.

The chart shows the current proportionality between volume pushing the market lower and volume recently pushing us higher.  The question is whether the shorts have all been trapped and buyers exhausted, leading to a potential cyclical move to the downside.  Toward that end, I note that at the market peak this week we saw 496 stocks across the major exchanges register fresh 3-month new highs.  That compares with 751 at the January peak.  Similarly, we registered 182 new 52-week highs among NYSE stocks on Friday, with 60 stocks making new 52-week lows.  That compares with 226 new highs and 7 new lows at the January peak.  At the January peak, we saw the percentage of SPX stocks trading above their 50, 100, and 200-day moving averages at 79%, 84%, and 80%, respectively.  Most recently, those numbers are 65%, 72%, and 76%.  Small cap stocks are trading below their January peak, as are stocks outside the U.S.  In sum, as we've moved to new highs in February, the market has been losing strength relative to January, which is what we'd expect in a cyclical scenario.

To re-emphasize:  all this frames a hypothesis and the kind of unique information I look at in framing hypotheses.  If we get a fresh thrust to the upside this coming week with enhanced breadth and volume, that would clearly violate the hypothesis of a coming cyclical top.  If we see sellers dominating with weak breadth, that would support the hypothesis and could lead to an actual trade idea that I'd need to frame with sound risk/reward.  

This is a nice little example of what I do in my market preparation daily.  You may very well look at different markets in different ways.  That's great.  The idea is to have a framework for thinking about buying and selling and ways of measuring buying and selling to generate hypotheses and ideas that are more than mere subjective perceptions of chart patterns and indicator readings.  In generating ideas, I'm much more interested in understanding what markets *are* doing than in predicting what they will do.  

Oh, by the way, on the chart above, note that the point of control (the price with the greatest volume traded over the month) is 3325.  If we do see a failed upside break vis a vis the February rise, I'd expect that level to be tested, as buyers would become the ones trapped.  That's a different hypothesis...

Further Reading: