Bulletproof Investing Performance Update: Week 90

Race car driver Mikaela Ahlin-Kottulinsky and her helmet (photo via Forbes).

Bulletproof Investing: Week 90 Performance

Each week, since the beginning of June 2017, I have presented at least two hedged portfolios created by Portfolio Armor to my Bulletproof Investing subscribers. This is an “investing with a helmet on” approach, and these portfolios are designed to last six months at most. As with any investment method, the returns with this approach will vary. But in the interests of transparency and accountability, I have promised to publicly share the final performance of everything I present, regardless of how it does.

Here, I update the final performance of the five hedged portfolios and the top 10 names (unhedged) that I presented in the 90th week I offered my service. Let’s look at what I presented in week 90 and how it did. I close by referring to recent changes we’ve made that should improve accuracy and increase performance in the future.

Portfolio 1

This was the $30,000 portfolio. The primary securities here were Array BioPharma (ARRY) and Crocs (CROX). They were selected because they had the highest potential return estimates, net of hedging costs when hedging against >13% declines, and they had share prices low enough that you could buy round lots of them for less than $10,000. Diodes (DIOD) was added in a fine-tuning step to absorb leftover cash from rounding down to round lots of the first two names.

The image above was generated by Portfolio Armor on Feb. 14 and presented in this Marketplace post at the time.

The worst-case scenario for this portfolio was a decline of 12.53% (the “max drawdown”), and the best-case scenario was a gain of 20.44% (the “Net Potential Return” or aggregate potential return net of hedging cost). The “Expected Return” of 8.25% was a ballpark estimate, taking into account the historic relationship between actual returns and Portfolio Armor’s potential return estimates.

Portfolio 1 Performance

Here’s how the portfolio did, net of hedging and trading costs.

This portfolio was up 3.52%, underperforming its expected return and underperforming the SPDR S&P 500 Trust ETF (SPY).

So far, we have six-month performance data for 38 portfolios I’ve presented that were hedged against >13% declines. Here’s how all of them have done. (Due to Seeking Alpha rules, I can no longer include tables with links to interactive charts, so I have included screen captures instead. You can find the interactive charts for every portfolio on performance section of the Portfolio Armor website).

Table via Portfolio Armor

Portfolio 2

This was the $100k portfolio. This one included Amarin (AMRN), ARRY, CROX, and Euronet Worldwide (EEFT) as primary securities. IQVIA Holdings (IQV) was added in the fine-tuning step again to absorb cash left over from the process of rounding down to round lots of the primary securities.

The image above was generated by Portfolio Armor on Feb. 14 and presented in this Marketplace post at the time.

The worst-case scenario for this one was a decline of 13.46%, the best-case scenario was a gain of 19.11%, and the ballpark estimate of an expected return was 7.78%.

Portfolio 2 Performance

Here’s how the portfolio did, net of hedging and trading costs.

This one was up 2.78%. So far, we have six-month performance data for 43 portfolios I’ve presented hedged against >14% declines. Here’s how all of them have done.

Table via Portfolio Armor.

Portfolio 3

This was the $1 million portfolio. It included AMRN, ARRY, Ciena (CIEN), CROX, EEFT, MarketAxess (MKTX), and Ulta Beauty (ULTA) as primary securities. Chart Industries (GTLS) was added in the fine-tuning step to absorb cash left over from the process of rounding down to round lots of the primary securities.

The image above was generated by Portfolio Armor on Feb. 14 and presented in this Marketplace post at the time.

The worst-case scenario here was a drawdown of 14.36%, the best-case scenario was a gain of 19.55% (the net potential return), and the expected return was 7.36%.

Portfolio 3 Performance

Here’s how the portfolio did, net of hedging and trading costs.

This one was up 5.14%, outperforming SPY, but not its expected return. So far, we have six-month performance data for 62 portfolios I’ve presented hedged against >15% declines. Here’s how all of them have done.

Table via Portfolio Armor

Portfolio 4

This was the $2 million aggressive portfolio. This one included AMRN, ARRY, CIEN, EEFT, MKTX, Tucows (TCX), and ULTA as primary securities. IQV was added to absorb leftover cash in the fine-tuning step.

The image above was generated by Portfolio Armor on Feb. 14 and presented in this Marketplace post at the time.

The worst-case scenario here was the max drawdown of 19.57%, the best-case scenario was the net potential return of 20.59%, and the expected return was 7.93%.

Portfolio 4 Performance

Here’s how the portfolio did, net of hedging and trading costs.

This one was up 7.8%, missing its expected return by 13 basis points, but outperforming SPY. So far, we have six-month performance data for 72 portfolios I’ve presented hedged against >20% declines. Here’s how all of them have done.

Table via Portfolio Armor

Portfolio 5

This was the $2 million top names portfolio. A name that appeared in this portfolio but not in the previous Feb. 14 portfolios was Xilinx (XLNX).

The image above was generated by Portfolio Armor on Feb. 14 and presented in this Marketplace post at the time.

The worst-case scenario was a drawdown of 8.43%, the best-case scenario was a gain of 17.32%, and the expected return was 5.95%.

Portfolio 5 Performance

Here’s how the portfolio did, net of hedging and trading costs.

This portfolio was up 6.22%, outperforming its expected return and SPY. So far, we have a full six-month performance for 88 portfolios I’ve presented hedged against >9% declines. Here’s how each of them did (click on a starting date to go to an interactive version of that chart).

Table via Portfolio Armor

One note about the table above: It includes both $100k portfolios and $1M portfolios. Starting with the May 24 cohort, I began presenting $100k portfolios hedged against >14% declines, so they appear in a different table from that point forward. My guess is that will slightly improve the average performance of the portfolios hedged against >9% declines.

Top Names

These were Portfolio Armor’s top 10 names as of Feb. 14. Names that didn’t appear in the portfolios above were The New York Times (NYT), AngloGold Ashanti (AU), VMware (VMW), Quaker Chemical (KWR), Starbucks (SBUX), and The Direxion Daily Brazil Bull 3x ETF (BRZU).

The image above was generated by Portfolio Armor on Feb. 14 and was included in the same Marketplace post as the top names portfolio above.

For this cohort, as of Feb. 14:

  • Average 36M Beta = 1.11 (the beta of BRZU was 3.46)
  • Average 20% threshold optimal put hedging cost: 2.61%

Top Names Performance

Here’s how the top names did:

The top names (unhedged) were up 8.71% on average vs. up 4.43% for SPY. So far, 50 top names cohorts have beaten SPY, one has tied SPY, and 38 have underperformed SPY over the next six months. You can see the performance for all of the top name cohorts I’ve presented so far in the table below.

Table via Portfolio Armor

So Portfolio Armor’s top 10 names averaged 7.18% over the average of these 89 six-month periods versus SPY’s average of 5.52%, an average outperformance of 1.66% over six months, or 3.32% annualized.

Top Names Time-Stamped

For a few months, in addition to posting those top names in my Seeking Alpha Marketplace service, I also time-stamped them on Twitter. If you click on the tweet shown below and scroll down, it will take you to a thread showing those time-stamped posts as well as charts of their subsequent performance.

Week 90 Assessment

The top 10 names (unhedged) outperformed SPY for the 50th time out of 89 weeks (we didn’t post the top 10 in week 1), and three out of the five hedged portfolios outperformed SPY, though only one outperformed its expected return. Overall, this cohort did decently – better than Week 89, but perhaps not as good as Week 88. Note that we’ve incorporated data from these 90 weeks of performance updates into recent algorithm changes that should boost performance and increase accuracy I described those here: When Strategy Meets Reality.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.