Just a small trading note to say that I’ve traded out of most positions in stocks (I’m keeping my core investment holdings, this is the trade portfolio). So I thought I’d use this as a teaching moment on how to trade out.
Remember the mantra “In Late, Early Out”.
You get out when you’ve got a profit and things go flat. You don’t wait for a loss. I’m up about 5% from the entry a week and a half or so ago, and that makes the rent for the month along with a toy or two. Time to bank it.
Today, for the second day in a row, things are ‘flat’. I’ve got all the risk, but getting no gain. Better to just step aside. I can always buy in again on the next entry call. Furthermore, the cash raised takes 3 days to ‘settle’. Between now and then if I buy something and it’s a mistake, I’m married to it until those 3 days are up. Better to be in cash now, so that next Monday I can make a trade-in (potentially short or long) without that ‘problem’. Finally, I get to have a ‘no nerves’ Friday and weekend ;-)
This is the S&P 500. It is darned hard for me to find a stock not in that list of 500 that can fight the trend it sets, especially as lots of folks trade it as a block. So if it says “be out”, well, odd are every stock will have a bad day. On this chart I’ve got 3 indicators I’ve not talked about very much. They give a different insight to the “action”. Volume. Slow Stochastic. Rate Of Change.
The static chart as of this morning:
Doesn’t look like the usual charts I post. I’ll explain.
Volume. That’s pretty clear. How many shares changed hands. So we’re looking for trends vs price. What happens to volume as prices rise or drop. On major drops, we get volume spikes as folks panic out (and as short sellers cram the market down with massive shorts). On tops, the volume just thins out and fades. So what is happening here? Starting from the volume spike in May (“Sell In May And Go Away”) we’ve been rolling down hill. Not good. Volume was above the trend line at the entry at the end of August, but it’s running low now. So we have volume saying we’re in the party a bit late.
Look at the price itself. Generally rolling sideways. In those cases “range trading” is useful. How high did it go last time? Gee, about where it is now. Are things REALLY significantly different now than then? News flow has been mixed, at best… We also have 3 moving averages of price in three different time scales. 25 day, 50 day, and 75 day. During a trend, these are separated as the price runs away from the lagging averages. In a trendless market they ‘weave’ as the averages stay, well, average… and we’ve got a weave going on right now. No trend, and near the prior range high. Hmmm…. Maybe time to be out until a trend develops.
Bollinger Bands: These are markers for one standard deviation of price. In a trending market the price tends to run along the edge of a band (as the higher volatility moves the bands outward). In trendless rollers, price tends to reach a band and fall away as “reversion to the mean” hits. You can see the trend behaviour back in March and May. But now we’re trendless and you can see how price touches the edge and reversion happens in the last 3 months. Bollinger Bands say we have lots of risk and not much reward.
Now look at the price bars themselves. This is called a ‘candlestick’ display. Each bar is ‘fat’ between the open price and close price of the day. Inter-day prices beyond that sick out as fine lines. (You will need to click on the chart to make it bigger to see this well). At tops, the bars get short and more of them look like plus signs or even just a dash. At reversals, you will sometimes get a ‘kangaroo tail’ as the thin line pokes out in the direction the kangaroo will leave as it bounces off it’s tail. (Traders have an active imagination ;-) You can see the Kangaroo Tails in the last couple of bottoms and you can see the plus signs in our present top as well as the last top. Also note that at the bottoms the bars are very long while at the tops they are very short. We’ve got short bars. Daily price ranges are saying time to be out.
All of that alone can be enough to make the trade, but the other indicators are interesting too. But take just a moment to look back over those blocks of text. That’s a heck of a lot of information from one little panel of a chart. And that is what I look for in every chart. ALL those details matter. Every time. And that is why the chart “speaks to me” faster than text.
OK, Slow Stochastic, despite it’s name, is a fairly fast indicator. When a stock is not moving much, it gives fast trade in / out twitches on small changes. During a ‘running’ phase it will “peg” at an extreme. You can see both of those behaviours here. Notice we’ve been pegged at the top for ‘a while’. After a while, comes a reversion to the mean moment. Always. The hard bit is ‘how long is a while?’. I note that at the end it’s got a bit of a downturn. Not enough by itself, but in the context of the other indicators, enough to have a fast trade exit. For longer duration trades or investments, I’d just put a stop loss order behind my position on that kind of indication. (But given that my gain is all of 5%, having a 5% stop loss would just get me out flat and that doesn’t pay the rent, so I’ve traded out early).
Finally, we get to Rate Of Change. Unlike “Momentum” that tends to be balanced in both directions, ROC tends to be a bit more pessimistic. Notice that on some past decent runs it gave a tepid ‘good, be in’ indication (above the line), but on down runs it is shouting “BE OUT!!!” below the line. So right now it’s been saying ‘be in’, but the peak is past and it is headed down toward the zero line. If you look at the end of July, I would be leaving somewhat early based on that indication. What ought to happen is a dip below the line before the exit. So again, I’m probably a bit ‘early out’ but I’m OK with that. “Bulls make money, Bears make money, but Pigs get slaughtered.” So to some extent I’m “projecting” onto the ROC what it has not yet said. But the first derivative of those bars is headed toward a zero line…
Hopefully walking through those indicators gives you a better feel for how to read them.
So you can see if that was a ‘good call’ over time, here is a live chart that will change day to day:
And this is a version with the slower indicators that I typically talk about, the ones that give a longer term view. As of today, they don’t say ‘be out’, just ‘being in has been good so the future might be more of the same’. And the future might well have some more good days. I might well miss a bit of rise into a new run up. But I’ve made my lunch money for the month and, well, “early out”… Also, I have to add, the trader talk on various financial shows has been toward traders looking to take short positions in stocks that have run up a lot. While that does not make my decision, it does tell me to take a second look (with the faster indicators above) and ask if I’m being a bit piggy and overstaying the peak of the party.
On this chart I’ve changed from Bollinger Bands to PSAR (Parabolic Stop and Release – that some folks have called Parabolic Stop and Reversal). It’s an indicator that supposedly tells you where to place a stop loss or ‘buy if touched’. The little red dots accelerate toward the price. This encourages you to stay in early on, but closes the gap rapidly in a static price after a run. Right now the ‘dots’ are not near the price, so it too is saying ‘go ahead and stay in a few more days’ and says I’m being ‘early out’. It would argue for maybe taking only 1/2 my bets off the table rather than all. But due to some family demands and my desire for a ‘day off’ or two, I’m willing to ‘bank it’ now rather than play a more nervous game through Monday. I’m not using it to tell me what to do in this case, but to measure what I’m doing. How much early might I be? Look back at the last run up and you can see that PSAR runs a bit late. At the 90% of the run up point, it was still only about 1/2 way up the price run. About like it is now…
Now we look at RSI. In a running stock, we look for approach to 80 for a run ending to the top side or an approach to 20 for a run ending to the downside. Relative Strength Index tells you how much prices are changing relative to their past tendency to change. Is this a relatively strong run for this ticker? (And strong runs tend to eventually run out of steam). In a ‘flat roller’ it tends to wander each side of the 50 midline about 1/2 way to the 20 and 80. We know from the SMA ‘weave’ that we’re in a ‘flat roller’ for now, and we can see that RSI is about the same as the last top; about 65. Now we could be in the start of a new run (I note that the lows have been slowly rising from a 25-50 roll to more of a 35) but given the news flow being mediocre, that we have the “spooky” crash month of October coming, that historically September is a lousy month, the dropping volume… I’m just not seeing a great bull run right now. So RSI most likely is saying “near a rolling top”.
MACD is just Moving Average Convergence Divergence. If you look at those SMA lines you will notice that during a strong run the faster moving average pulls away from the slower one. A gap develops between them. By measuring that gap, you get a crude first derivative of price. The rate of change. As a run ends, the longer term moving averages catch up to the faster ones and the MACD shows a convergence (and eventually a cross over). At present, this says “be in” in that it is “blue on top” and “opening pointing up”. This is where Slow Stochastic has an advantage. MACD is a lagging indicator. Great for getting in a bit late, but it also gets you out a bit late. So I look at it with a ‘fast eye’. The blue line is going a bit flat. The red line will be catching up. Normally I’d wait a day or two for it to at least look alike a crossover setting up, but I want to hit the bank and take a break. At this point it mostly is saying ‘put a stop loss in place’. So I’m going a bit beyond what it says and doing it a bit early. The price bars and Bollinger bands give me some confidence that’s a good decision to over ride this indicator rather than wait.
Now look at those black bars in the histogram part of MACD. That’s the gap between the blue and red lines. It’s reached a max and is dropping. When it hits zero, you have crossover. So I’m choosing to leave a bit early, at the infection of slope of a tangent to those bars, rather than waiting for a zero.
Finally, we have ADX / DMI at the bottom. The ADX line tells you how strong the trend has been lately. We know that trend has been weak already from the ‘weave’ of the SMA stack that tells us the same thing. The ADX line (black) just does it as a number. Over 25 is fairly strong and MACD is best to use (as it keeps you in a strong trend rather than trading out at the wrong time). Below 20 it says to use Slow Stochastic as Slow Stoch will make the rapid trades that make you some money in a twitchy rolling going nowhere stock. Right now it is about 20 so confirms the decision to underweight the MACD ‘stay in’ suggestion. The red and blue lines just give separate ‘strength’ indications for up and down movements. Blue at 30 and red at 15 says a nice strong positive run has been going on. Normally I’d wait for the blue line to drop below the black line before I’d leave a trade (as the positive move days weaken in the context of the overall strength of movement in any direction). You want to see the blue line inflect down and the red up before you leave. I would expect that to happen in the next day or three as the up run falters and we get a couple of down days. That would be a good ‘trend following’ rule set. But I know ADX/DMI are lagging indicators. And at this moment I’m interested in being ‘out early’ rather than getting more gain for a couple of days and leaving my rent money at risk… If the other indicators were saying ‘be in’ I’d look at this and feel good. With the others saying ‘out sounds good’, this just tells me I might be a bit early, and that’s fine with me. So I’ll watch it for the next couple of days to see if it confirms my decision. If it doesn’t, and if a new run starts, it will chasten me about being too lazy to have made some more ;-)
Hopefully this lets folks see how a trader thinks, and how to read these charts. Yes, it’s a lot that goes on when I look at these charts. Trading is work. But you get good at it over time. Practice practice practice. It goes about 100 times faster with a glance at the chart, after some practice, than it does in words on a page, so don’t be discouraged by how long it took to read this and look at the chart. Eventually it’s just a ‘click click’ chart is up, golly, need to look at the fast chart ‘click click’ OK, I’m getting out… (or in).
So give it a try. It will pay off nicely.
This is one of my favorite “strange charts”. It has several ETFs (Exchange Traded Funds) on it that hold various indexes, or ‘inverse’ indexes (meaning it shorts that index). It looks a bit like a Rorschach blot. What I find interesting is the tendency for the S&P 500 (SPY the gold line) and it’s short SH (the blue line) to ‘kiss’ in the middle at buy points. Less reliable, but still of use as a ‘late call’ or confirmation is the tendency for a wide gap with lots of while space in the middle to presage a drop, especially when the slope of a tangent to price passes through zero and heads down (price inflection or derivative zero to negative). The highly leveraged products, like QLD, tend to give a fairly pronounced roll over at tops (like you can see now as QLD goes flat) and the inverse tickers spike up like TWM at the bottom of the graph; and give a pretty good warning to exit the less volatile tickers too.
Besides, It’s kind of cool to stare at it and imagine butterflies and ski runs and … ;-)
SPY - S&P 500 long SH - S&P 500 short IWM - Russell 2000 long RWM - Russel 2000 short UWM - Russel 2000 double long TWM - Russel 2000 double short QQQQ - Nasdaq 100 long PSQ - Nasdaq 100 short QLD - Nasdaq 100 double long QiD - Nasdaq 100 double short
Fun to watch if nothing else…
You could do the same thing with the Dow 30 industrials (DIA) or other indexes as well. I just usually trade these three.
SPY - S&P 500 long SH - S&P 500 short SSO - S&P 500 double long SDS - S&P 500 double short DIA - Dow 30 Industrials long DOG - Dow 30 Industrials short DDM - Dow 30 Industrials double long DXD - Dow 30 Industrials double short QLD - Nasdaq 100 double long QiD - Nasdaq 100 double short
You can see it looks pretty much the same.