I believe most folks look at a chart or a statistic and decide the market is overbought or oversold. "Look at that RSI at 80 (or 90 or whatever), we're overbought". How often do you hear that? My question is always: how do you know it's not going to 90 or 95? Or another question: how do you know it's not going to dip to 75 and then go up again?
I am not that familiar with the formula for the Relative Strength Index (RSI) in that I don't calculate it myself, but my impression is it is the indicator du jour for all those Technical Analysis Tourists out there. And let me point out one more thing: at the lows none of these tourists were saying 'oversold.'
I feel at this point I should disclose that I have no idea what the current RSI reading on any of the indexes is. But I do know how my own Oscillator and various other methodologies I use to calculate an overbought or oversold condition and yes we are short-term overbought. But there is more to that than meets the eye.
First of all, as I noted Friday, the Oscillator was set to climb over +500 on Friday if the market rallied and that it did. But what's the math behind it? The math behind it says likely to pullback and rally again. The math behind it says we'll be maximum (short term) overbought a week from Monday. Look at the chart and see how it rarely spikes and falls but rather spikes, dips and goes up again.
Then there is the McClellan Summation Index. As longtime readers know, I use the Summation Index to determine the direction of the majority of stocks. When it is heading up, the market is bullish. When it is heading down, I believe it is bearish.
I also use a calculation of what it will take to turn the Summation Index up or down, so since we're currently rallying, what will it take to halt the rise and then turn it down? Right now, it needs a net differential on the NYSE of -4,600 (advancers minus decliners). That's a lot as you can see from the chart.
But take a closer look and notice that the extreme 'what if' rarely marks the high. Rather we get an extreme reading and then we back off, digest, and rally again. Notice the peaks in the market were all when the 'what if' was less than the prior extreme reading.
Three weeks ago, I spent a great deal of time showing you the Russell 2000 Momentum Indicator. What I did here was plug in lower closes until we saw the indicator go up instead of down, showing a loss of downside momentum, thus telling us we were oversold.
For an overbought condition, it's no different. So now I have plugged in higher closes to walk the Russell up 100 points in the next week or so. The first little peak arrives midweek this week but the indicator doesn't dip down (and only a smidge right now) until next week. Thus we are getting short-term overbought sometime in the next week using this metric.
There is a certain rhythm to these oscillators, we tend to get a short-term overbought (or oversold) condition, followed by a dip or pullback or digestion period to relieve that initial overbought (oversold) condition, followed by another attempt higher (or lower). It's that second attempt that is the real test.
By then, we can see if momentum has gained or lost. Or we can see if sentiment has shifted sufficiently to determine if the market is no longer climbing a wall of worry. By then, we can determine if the intermediate term indicators are overbought yet, if there are negative divergences, or if there are confirmations.
So yes we are getting short-term overbought, but I still think after any digest or pullback we would rally again.
