"what period when to when are you measuring with this recent big selloff as far as looking at retracement levels?


This week with my most recent Index Wrap measured the 'common' downside retracement levels (38, 50, 62%) by looking at the early-February low relative to the mid-September top in the S&P 500 (SPX) and the Dow 30 (INDU). With the S&P 100 (OEX) I used its (later) March low, as I did with the Nasdaq Composite (COMP) and the Nasdaq 100 (NDX).

For the sake of traders wondering what's up with this topic that they might want to know, I'll get into the potential VALUE of using the common so-called Fibonacci retracements in terms of highlighting potential support areas or possible downside targets a bit further on here. This part is the mechanics of WHAT low to what high is used to set up a retracement study which is a drawing tool in common charting applications.

Looking at the retracement levels from March to the mid-Sept high was appropriate in COMP and NDX as their early-March intraday lows were slightly below their February bottom so March was the prior lowest low of significance. In retrospect it would be good to now use the LOWER big cap S&P 100 low of February to be consistent with the bigger 500 (SPX) Index. I'll get to that shortly.

In the issue of what high to use after a sharp sell off, that's a no brainer as we use the prior intraday top. As to WHAT is the most significant prior LOW to use, that's not always clear.

I measure retracement levels on daily charts mostly. As to what is the prior most RELEVANT low to use, use of a weekly chart will sometimes clarify which prior low to use. An example is provided by looking first at the SPX weekly chart:

I'm looking for the most significant prior downswing low to measure from; i.e., the deepest PRIOR downswing, here relative to the SPX intraday top at 2019 in SPX in mid-September. This shows up for me at the 1738 SPX weekly low (the day was 2/5/14).

When we've established the prior low to use and we know the peak level, using the retracement study for a downside retracement is relatively simple: click on the low, click on the high and see the results. The 'default' values used will typically show what level 'retraces' 38.2% of 281 S&P points (2019 minus 1738), what level retraces 50% of that and what level retraces 61.8%. You can set various other retracement levels such as 25%, 75% and a personal favorite of mine, 2/3rds or 66%.

IF a 'minimal' 25-38% retracement has already been reached and exceeded, you can take that value out so as to not clutter up your chart(s). The same with 50% and what we often see as a 'normal' retracement for a stock or index. You can leave the 50% level displayed if you want it for reference or if the retracement level to date is still CLOSE to the one-half/50% level.

Last, but not least, I look at both the 62% (rounded from 61.8; as is 38% rounded from 38.2) AND 66% level, as they are close to each other and make a retracement zone which is a common 'deep' retracement. The deepest downside retracement is actually 100% or a round-trip back to the prior low, which is how double bottoms can set up.


A 25 to 38% retracement is common in a STRONG trend and is a fairly 'minimal' give-back. If a 38% retracement is exceeded, it's common to look for a 50% retracement as a potential next objective. This rule of thumb gives at least an idea of what might be a potential next move.

If a 50% retracement is exceeded, the next common, deeper, retracement level is 62% or (as I say) 'a little bit more' which is 66%. (If a retracement goes beyond 66% in a stock or index this pattern suggests potential back to the starting point of the rally or the starting point of a TOP in the case of measuring upside retracements in a bear rebound.)

Going back and using the February lows in OEX as my starting point to measure the common retracement levels, we see below that OEX has almost retraced 50% of its prior (Feb-Sept) advance. This gives me a better read of potential support.

If I continued to use OEX's March low as my starting point, OEX today would have exceeded (slightly) a 66% retracement. What might a 50% versus 66% retracement suggest? Well, the big cap S&P 100 has been in a strong advance for some time and, in strong trends, we often see a 50% retracement as the area where buying interest starts to come in again. That may be the case with OEX but stay tuned on that of course!

In the Dow 30 (INDU), the Average's retracement to date is holding in the area of a 50% retracement which is of course near to the level where INDU bottomed in early-August, setting up the potential for an approximate double bottom low. INDU is also holding near implied support by its longer-term up trendline. There's potential for the Dow to fall to the 16100-16000 area if we continue to see the Average slip below today's 16315 close.

This way of looking at downside potential by the use of measuring the common Fibonacci retracements gives an idea of further/lower downside objectives. Sometimes we don't have anything much to go on to even guess. With INDU of course, in the 16000 area, a 66% retracement, there are prior lows that would make it more compelling perhaps as a place to exit Dow Index puts and reverse into DJX calls. We'll see won't we!


It can be quite interesting in situations where a key retracement, among OTHER chart/technical factor, might point to a possible bottom. Whether that turns out to just be an interim low or a longer-term bottom. Keep in mind that a 62 to 66 percent retracement is a DEEP correction. Unless the sky is falling, instead of the computers piling on the sell side.

With the Nasdaq Composite (COMP) chart we can see potential support implied by the 62% Fibonacci retracement level at 4200, which is also where a longer-term up trendline intersects and where 4200 also represents possible support implied by an upside gap from May.

If you add to this that COMP has finally reached an 'oversold' extreme in terms of the 13-day Relative Strength Index (RSI), this mix of things doesn't add up to 'evidence' of a bottom (only price action can provide that) but it does get me focusing in what Nasdaq price action does from here that would suggest a potential bottom.