[pcabc]

Maroon alert!  20+ year plot!

Warning, very large image.

I hope no one minds the major departure from the usual limit of 800 pixel wide images on this site, as I have just plotted my work on a breadth indicator to warn of large drops in sentiment such as we have experienced now and in 1987, over a 20+ year period.  This chart encompasses both 1987 and 2008.  It is based on NYSE data.  Hopefully your browser will allow you to scroll though this image at full size, or you can download it into an application that can.  Or if you have a 4k display you might see it all at once, but might have to sit pretty close to see the detail.

The chart shows the breadth curves I usually plot for myself.  I have a colour coded bar plotted below the price which I described a few posts back.  Recently I have tried to plot major deteriorations in breadth to hopefully only trigger on major deterioration such as now or 1987, something to give me a clear hit to get out.

I have added 'ticks' to the bar so that it is easier to see these danger warnings, 'Maroon' and 'Black'.  'Maroon' means that there is serious market breadth deterioration but prices are still high whereas 'Black' means that the price has started to drop.  The decider between these two colours is whether price in the index used in the breadth chart is above or below the 50% retracement line.  The 50% retracement line is taken from the TrendAdvisor method, it is the line halfway between the top and bottom of a 60 day price channel.
   

I could have ignored splitting my indicator into Maroon and Black stages and just plotted the deterioration in breath.  However, I figured that this breadth indicator would be persistent in a stage 4 market and might trigger occasionally in stages 1 and 4.  So I set up the Maroon indication to show a breadth deterioration when price was strong, a rare but significant event.

The strongly deteriorating breadth is mainly based on the Weinstein Momentum and the percentage of stocks above their 200 day EMA.  These really are the significant triggers.  I don't see the faster percentage of stocks above the 50 day EMA being so useful.  New highs minus new lows seems to confirm current price action rather than predict.  Limited account is taken of the advance / decline line.  It is hard to do anything to simply spot divergences in price.  I have a scheme in place but with limited capability and it is weighted as such.  These lesser contributions have lead to me lightly greying out these curves.

At present I see no obvious contribution for the advancing / declining volume line.  So this is greyed out with a darker shade.  Note, I am looking for catching rapid deteriorations like now and 1987.  So comparing A/D and A/D volume against slower MAs has little or no value as slower MAs are crossed after the event and faster ones trigger all the time.

I believe I have given notice of deterioration, programmatically for current events and those in 1987.  For 2008 and the dotcom boom end I this is really unnecessary.  YOu were well into a clear stage 4 on the index charts before the worst happened.

What I need to do now is zoom in on the warnings raised not at major crashes.  Are they false alarms, or are they forgotten pullback events that people following stage analysis would have got out from?