As is often the case, the simplest ideas can have the greatest impact. This article briefly describes a basic pattern in the S&P500 that I use to exit leveraged long trading positions, when the stock market changes from a risk-on regime to a risk-off environment, as defined by my Meta Strategy Model. It is one of the core concepts of my trading approach, which allows me to take outsized risks at opportune times and to exit with minimal impact to my trading capital as soon as things start to go wrong.

Profiting in a bull market

If anything can happen at any time in financial markets, why is it that one can take leveraged risk in a Quiet Bull Market Regime and have a very good chance to get away with it? It boils down to a very simple pattern that is reliable and tends to occur around the time the Meta Strategy indicators signal a regime change to risk-off. 

Human tendency to process and adjust to change slowly will lead to a predictable lag in behavior when the market regime has already shifted. The first, violent stock market decline near the end of a a bull market almost inevitably leads to a bounce that quickly retraces at least 50% of the drop before the market gets into serious trouble.

If one fearlessly scales into long positions during such a decline, the relief rally can be used to get out of leveraged long positions – usually with a profit, if one can stomach the temporary, deep drawdown. This requires a strong trust in the pattern, as well as a cap on maximum exposure that makes a pattern failure survivable.

In essence: Continue buying the first violent dip, then get out in the bounce, because the market regime has changed.

The pattern and the warning signal

First, we need to define the setup and the signal, which occurs rarely (in the meantime we can profit from a calm uptrend – in many cases for years):

a) Stocks are in a bullish trend: Price > 60 dma > 275 dma. Buy pullbacks systematically.
b) A sudden, violent drop occurs: It can be visualized and quantified by using Bollinger Bands of two standard deviations around the medium-term trend (60 dma). A quick decline exceeds the lower band – historically this translates to a sudden 8-20% crash in most instances. Continue to scale in (at set intervals planned in advance) until maximum, survivable exposure is reached.
c) A volatility warning signal triggers: A regime change is quantified by a significant inversion of the VIX Futures Term Structure (contango < -1,5%), which tells us there is a serious problem in the market. Often this roughly coincides with a break of the long-term trend (275 dma).
d) Exit all leveraged long positions at the 50% retracement of the initial decline. Depending on the magnitude of the drop this will result in a profit or a small loss.

Most pattern failures occurred, because bad things did not happen after all and stocks quickly regained their uptrend – it’s a benign failure, as it does not pose a risk of loss. This does not mean the setup won’t fail in the future (like it did in 1987 or 1990) and exceed prior drawdowns and / or fail to bounce sufficiently. For this eventuality I define my maximum loss in advance (my trading portfolio is limited to 20% of my capital). A big loss every other decade on average is well worth it for me, because I can generate much larger profits in the meantime by trading uptrends with leverage.

Here is how the idea played out in past decades:

Data Dump

All occurrences of the risk-off exit signal in the last 60 years including rare failures, as well as signals that resulted in an immediate resumption of the bull market trend without doing the trading portfolio harm.

Observations
A warning signal occurs roughly once a year (53 times in the past 60 years), with 82% of all instances leading to a benign, profitable resolution (the S&P500 drops 6-14% and quickly bounces to the mid-point of the decline; most declines < 8% lead to an immediate resumption of the previous uptrend).
A straight drop larger than 14% will feel very uncomfortable, trading positions will go into a deep drawdown and it usually takes longer to play out. However, this only happens once every five years on average. Included in those crashes, 8% of all instances (4 times in 60 years) exceed a 20% peak-to-trough decline, which will likely result in a loss after a long period of patience while being deep underwater.
Failures: In 1990 an entire shallow bear market played out, until 50% of the drop was finally retraced months later (any options would likely have expired worthless, but specialized instruments without built-in time decay may have succeeded). In 1987 stocks even dropped 34% straight (but did retrace 38% of the decline within days).

Scaling in and maximum exposure: A simple plan could be to cap maximum exposure at around 20% of capital and scale in with four separate positions during the decline (5% of capital at -4%; -9%; -14%; -19%). Maximum risk is a 100% loss of the combined trading position. In case this happens, I rebalance my trading portfolio back to 20% NAV using the remainder of the capital in my investment portfolio.

Charts
All instances since the 1960s. Red circles denote signals after which I continue to scale into long positions (rigorously following a plan set in advance) until maximum exposure is reached. The first strong bounce following an initial bottom is used to exit long positions at the 50% retracement at a profit or small loss in the vast majority of cases.

This article is based on the current market outlook in my weekly report. For more practical details on running a holistic investment & trading portfolio, and to support my research please consider subscribing. Thank you!

Good luck with your trading, and thank you for reading!

David

This is not financial advice.
These are my own views, as I may implement them in my own portfolio.
Please do your own due diligence!

Browse Recent Blog Posts