A few readers have contacted me asking about the extreme spike in TRIN or Arms Index. Bespoke featured this chart of the 10 day moving average of TRIN with the implication that it was an early warning for the impending bear market that started in late 2007:
Source: TRIN Index Shows Extreme Selling
Bespoke doesn’t come out and say that explicitly but it isn’t too difficult to imagine most people interpreting the chart that way. To play Devil’s advocate, let’s take another look at this from a few different perspectives.
First, let’s expand our time horizon to more than just 8 years. Here’s a chart of the NYSE TRIN (10 day moving average) from 1992:
Pretend that you’re trading in 1998 when the short term smoothing average of the TRIN spiked to a record high (above 1.4). What would you have thought? I’m sure many were rending their garments and pulling their hair out.
What if we pretend it is 2003? It was over 1.8 by then! Oh, the humanity!
I think my point is clear. It is obvious that when we look at the long term history of this indicator, we have to treat spikes with great caution because they may simply be resetting the definition of “extreme”. In other words, it isn’t so much an early warning signal as a shifting signpost.
In case you haven’t noticed, this isn’t your grandfather’s stock market. Volatility is consistently high and volume is acting very weird. Maybe it is the fact that retail traders have taken their marbles and gone home. Maybe it is the fact that high frequency algorithms vie with each other in picoseconds and produce a mirage of liquidity. Maybe it is something else. By nature, the market is an ever evolving thing.
Another way to get some perspective on this issue is to switch markets and look at the Nasdaq TRIN instead of the usual NYSE variant. If you’ve read the blog for some time, you will already be familiar with my criticism of NYSE breadth data. The Nasdaq isn’t polluted with non-operating company stocks so that itself is one advantage.
Looking at the long term chart of the Nasdaq TRIN we see that it looks very different from the NYSE TRIN. For one, there was no corresponding spike in early 2007 to act as any sort of “signal”. Second, it seems to be a more stable with more constant highs and lows.
If you recall your high school chemistry, it is important to have a null hypothesis. As humans we have a natural tendency to seek out information that will confirm our established conclusions and our built-in bias. But the scientific process demands that we seek out information that could negate our hypothesis.
Bespoke’s blog post didn’t get any comments but the same article republished on seekingalpha did. As a reader recently noticed, public comments left on this blog and the like are a good source of anecdotal evidence. While qualitative, they proved to be a helpful guide during the last bear market. Whenever I wrote anything positive (and I did more than a few times) readers would react with a mixture of derision, hostility and skepticism. I’ll leave you to decide where things stand these days.
If you really don’t feel like thinking for yourself on this, you could do worse that defer to the man who created the index, Dick Arms and his take on the recent extreme TRIN spike.
Enjoyed this? Don't miss the next one, grab the feed or