Lots of hoopla around the "longest bull market ever" this week. But it requires some fuzzy math around the -19% drops that happened in 1997, 1998 and 2011. See this tweet for the details.

Other usually smart people have made the argument to start counting from the recovery. Um, right. So NDX didn't have a bull market after 2000 until late 2016? What about the ~300+% rally off the 2009 low?

There are plenty of other incidences like this: Nikkei rally from 2012, Shanghai Comp from 2014, EEM from 2016, etc. Plenty of money to made on those three rallies, so saying the bull "doesn't count" because it isn't above a prior high made years or decades ago doesn't work if you are about making money.

What really matters is whether you should have allocation to an asset class. Let's think about it the other way, when you shouldn't have exposure to an asset class.

Weekly close below YP* (*fractional closes are judgment calls; clear rejection is a better signal)
Weekly close under W50MA
W50MA is falling

Paying attention to these three conditions will get you out of the asset classes that are really in trouble. Here are a few examples this year.

People who want to do a little more fine turning can include HPs as well.

EWZ clear rejection of YP (and HP), below W50; at the time W50 was probably nearly flat. First long term sell / avoid signal since recovery above YP in 2016.

GLD first below W50 and HP, and then soon after broke YP and went a lot lower after that.

Gold miners GDX have been more significantly bearish for most of the year, and had more damage on the drop.

Remember to use total return technicals on any higher yielding asset. Here's TLT from stockcharts.com, showing a longer term avoid in January, several holds of YS1, and a recent recovery of OK to allocate status (above YP, above 50MA, yet 50MA is still falling somewhat).