THREAD:
Many times, we hear that even if we know there's a bubble, it's difficult to anticipate the top. That is true, but some elements can help us to measure how risky the market has become.

1/ I like to use logarithmic charts. The straight line means that the index has moved exponentially since 2010, which is quite crazy. But it has even managed to rise faster after March 2020, implying monster moves since the beginning of COVID-19 pandemic.

2/ Note that a similar thing occurred during the dotcom bubble, with a breakout above the straight line. Not saying that it signals an imminent rupture, but one should wonder whether such trend is sustainable. Remember that during a bubble the trend becomes its own fundamental.

3/ The LPPLS model helped to identify the September 3 rupture. Even if, the index has move higher after November, the uptrend has been less and less powerful. Which brings us to the question of bearish divergences.

4/ Indeed, we can see a form of MACD bearish divergence over a multi-month period for both the S&P 500 and the Nasdaq.

5/ Interestingly, the LT picture also signals a bearish divergence of the RSI, meaning that even if the Nasdaq has gone higher and higher, the trend is getting weaker and weaker.

6/ From a very ST perspective, it's important to note that all indices are trading at (or close to) ATH, while the volumes have significantly decline (on both futures & cash markets). Low volumes are generally the sign of a coming correction.

7/ Last, an important contributor to the performance of the past months may be disappearing. I’m talking about the retail army and their YOLO trade. First, the volume on equity calls may have topped, limiting the impact of the famous gamma squeeze.

8/ Secondly, recent IPOs and SPACs have shown signs of weakness. Maybe too many issues to digest? The same thing has happened on the less famous convertible bonds market.

9/ Despite the decline of the retail whale, sentiment indicators are still in the greed zone (PMs and allocators are still massively OW US equities), and a simple look at tweets and answers confirm the fact that we've just reached the "we're invincible" moment again.

10/ What could go wrong? First, leverage is a big issue, as the Fed has encouraged extreme risk-taking behavior for years. The failure of Archegos is a sound reminder of what happened when levered agent collapse.

11/ Besides, contrary to what people think, there's no guarantee of liquidity in the market when things go wrong (i.e. when instantaneous volatility spikes). I recommend the excellent thread by @FadingRallies to understand the problem of the convexity of the order book.

12/ Last, individuals tend to underestimate the probability of occurrence of fat-tail events. Something that has been well document by Taleb long ago. Meanwhile, volatility spikes display forms of power-tailed distributions, meaning that large spikes will continue to occur.

13/ In other words, people are structurally short downside volatility, as they think that the market cannot drop more than 5/10%, and that there will always be someone to bid. As more people share that opinion, the VIX offers interesting opportunity for contrarian investors.

14/ I like those compression patterns of instantaneous volatility, as this chart by @NorthmanTrader, with interesting track records over the past year. As Buffet says: "Be fearful when others are greedy and be greedy when others are fearful."

15/ I'm not making investment recommendation, but I do think that there's been too much complacency in this market while fat-tail risks are seriously underestimated. You never know exactly when a market starts to be derailed, but at least you know there is no margin safety today.

If you liked this thread, I recommend my recent posts:
https://theswarmblog.com/index.php/2021/04/07/into-the-swarm-1-archegos-in-the-coal-mine/
https://theswarmblog.com/index.php/2021/03/25/whale-beaching/
And that one about the LPPLS model:
https://theswarmblog.com/index.php/2020/10/16/it-is-all-about-fractals-tech-stocks-and-the-lppls-model-remastered/