Completed SweetSpot Trades 1999-2016
The gross returns of all SweetSpot trades completed to date are presented below. Performance is expressed as the dividend-adjusted percentage returns an investor would have seen if she had bought at the beginning of each trade period and exited at the end, neither adding nor selling any shares along the way. Returns are measured against the corresponding dividend-adjusted returns of SweetSpot’s most-representative market-index ETF(s) (see “Benchmarks” below).
Past performance is not an assurance of similar future results.
*”Market” is the S&P 500 (SPY) from 1999-2007, and global stocks from 2008 onward (see “Benchmarks” below).
**Average and aggregated returns are presented to illustrate SweetSpot’s history of profitable, market-beating trades. Due to concurrent and overlapping trade periods, the effects of compounding, and other factors, they do not represent actual annualized returns that an investor who traded the strategy would have realized. See the two charts below for gross and net cumulative annualized returns since inception and for the most recent one, three, five, ten and 15 years.
2016 in Review*
“When everybody thinks alike, everyone is likely to be wrong.”
~ Humphrey Bancroft Neill
After starting the year at deeply depressed levels, prices for commodities and other hard assets rallied strongly in 2016. The resource-rich SweetSpot portfolio benefited from the recovery, gaining +19.6 percent.* This result was gratifying, both in absolute terms and relative to the +8.5-percent return of the All Country World Index ETF (ACWI). Returns were consistent across the portfolio, as 21 of 23 holdings saw gains, and all outliers were to the upside. Portfolio holdings were paced all year by two investments that at the beginning of the year, just about everyone thought should be avoided: [One gained +107.8 percent for the year and the other +64.9 percent. (Proprietary information that would identify portfolio holdings has been redacted.)]
SweetSpot’s relative returns in 2016 aren’t the only indication that contrarian investing is alive and well. For example, in August Bloomberg posted links to a pair of articles:
>>Take more risks, says hedge fund manager Paul Tudor Jones.
…Even while one economist says the future of investing is in following the herd.<<
Never mind the substance (but if you care, you can find the articles here and here). What’s telling is Bloomberg’s use of the words “Even while…” meaning that the two articles were being contrasted. That is, after the first article called for investors to take on more risk, Bloomberg’s editor seemed to think that the second article’s call to follow the herd would spell less risk. On the contrary (pun intended), the riskiest trades also tend to be the most crowded ones. Investors follow the herd at their peril. Considering that Bloomberg is a thought leader, this kind of gaffe bodes well for the persistence of a herd mentality among the investing public, a mentality that SweetSpot investors rely upon to provide them with favorable trades year after year.
“You can see a lot just by looking.”
~ Yogi Berra
Annual Returns 1999-2015*
SSI has long maintained that all that matters is our buy and our sell — everything in between is just so much noise. For years, the absolute and relative returns of SweetSpot’s multi-year trades presented above told us all we needed to know as long-term investors. We didn’t even bother to calculate annual returns until 2005 when we wondered how they would compare with others’ track records. They compared favorably.
SweetSpot’s long-term advantage has persisted as we have added years of real-time trading to the program’s cumulative record. Still, in contrast to the outstanding track record seen above for completed SweetSpot trades, annual returns in recent years have been unremarkable:
Up-to-date real-time annual returns
Presented below are the model SweetSpot portfolio’s annual gross** and net*** returns since real-time trading began in 1999, alongside the corresponding returns of the portfolio’s market benchmark. Returns for 1999-2010 assume an equal investment in all positions held each year. Beginning in 2011, the model portfolio was rebalanced each year to take advantage of the superior historical returns delivered when trades were nearest to completion. Reported returns for those years reflect recommended weightings. Performance figures assume the reinvestment of all dividends and other distributions.
Source of dividend-adjusted price data: http://finance.yahoo.com except where noted.
*Benchmark is the S&P 500 (SPY) from 1999-2007, and global stocks from 2008 onward (see “Benchmarks” below).
**Gross returns account for the fees and expenses that fund sponsors charged to fund assets, but they do not reflect:
Commissions and trading fees. From 1998-2006, SweetSpot investors traded commission-free, no-load mutual funds exclusively. When ETFs were added to the mix in 2007, commissions and trading fees as a percentage of trade value were negligible because each position gave rise to just two transactions — one buy and one sell.
Investment advisory fees. Before 2008, SweetSpot was traded by a private family office and others who paid no advisory fees. (Indeed, until SweetSpot Investments LLC registered as an investment adviser in 2008, the firm was prohibited from charging for its services.) After the registered firm opened for business, most investors subscribed to the information needed to trade the strategy in their own accounts. Subscription fees represented a small fraction of one percent of most subscribers’ invested assets.
In 2016, SSI declined to renew its registration as an investment adviser and began to offer SweetSpot to self-directed investors on a subscription basis exclusively. Subscription fees continue to represent a small fraction of one percent of most subscribers’ invested assets.
***Net returns assume that an investor paid a one-percent annual advisory fee, plus trading fees and commissions on a $200,000 portfolio.
Refining our Method
Beginning with trades entered in 2011 — and for the first time since real-time trading began in 1999 — SweetSpot investors saw lackluster relative returns over a multi-year period. By January 2014, four of the five holdings due to be sold trailed their market benchmark. A question arose:
If our strategy is to invest on the cheap, why would we want to sell something that’s cheaper at the end of our holding period than it was at the beginning?
A quick-and-dirty test gave us a clear answer: we wouldn’t. This is easy enough to verify: In the chart above, simply find the eight completed trades that underperformed the market (out of 39 total trades through 2013). What would have happened if we had not sold when we did? Here’s what we found:
- Three of the eight trades needed just one year to transform themselves into market-beaters;
- Two trades needed two years;
- One trade needed four years;
- One trade made weak progress after two years;
- One trade made good progress in the one year since we sold; and
- All three unprofitable trades became profitable.
Most significantly, the average return on a total of 14 hypothetical one-year holds was +25.8 percent, or 12.2 points better than the market’s +13.6-percent corresponding gain.
A cursory look at historical data from 1989-1999 — before real-time trading began — yielded similar findings.** Holding onto underperforming trades would have produced average annual returns of +13.0 percent, more than double the market’s average +5.7-percent gain. Results were consistent, as average returns in every trade year were positive and exceeded market returns.
These initial findings convinced us to hold onto our underperformers, which is why only one position entered in 2011 is shown among the completed trades listed above. We also felt motivated to dig deeper into the data, and what we found felt like a breakthrough of sorts.
“The things we are doing will not go away. We may have bad years, we may have a terrible year sometimes. But the principles we’ve discovered are valid.”
James Simons, Ph.D, mathematician and hedge-fund manager
In 2014 SSI conducted a comprehensive review of its investment process, drawing on current literature, 25 years of historical fund data, and 15 years of experience trading SweetSpot in real time. We looked at 77 completed multi-year trades that either were, or would have been, entered between 1989 and 2011 using SSI’s established method.** Each trade was held for an initial three-year period, and longer if: 1) it was renewed (that is, identified as a new buy in subsequent years); or 2) it underperformed the market in its first three years. Between renewals and underperformers, 41 of 77 completed trades were held for longer than three years.
The numbers were clear: Most underperformers became market-beaters overall within one to six years (or four to nine years after they were first bought). Average annual trade returns in years four through nine would have been +18.0 percent, doubling the market’s corresponding returns (+8.9 percent).*** These numbers far exceeded what we saw from positions in their initial three years, averaging +13.7 percent (or +5.7 percent better than the market). Aggregating the returns for all 77 trades, we see:
Backtest: Completed SweetSpot Trades, 1989-2014***
- Average annual return per trade: +15.1 percent
- Corresponding market return: +8.3 percent
- Excess annual return: +6.8 percentage points
- Sixty-nine of 77 completed trades outperformed the market. That’s a win rate of 89.6 percent, meaning the odds that any given trade would outperform were 9-1.
- Excess returns would have been achieved while taking on a small fraction of market risk [see “The Risks of Investing (in SweetSpot and Otherwise)“].
- Results were consistent throughout the test period. Average returns in all trade years (one through nine) were positive and exceeded market returns.
Implications for the SweetSpot method
“Price is what you pay. Value is what you get.”
The decision we made in 2014 to treat underperforming trades as deserving of special handling is an elegant refinement of SSI’s method. The reality is that every trade is subject to its own unique circumstances, and its own optimal holding period. Granted, only in hindsight can we know for certain what was optimal. But we must act (or not) in the present. Returns relative to the market are an intuitive, vetted metric that can inform our sell decisions and improve our ability to account for trade-specific factors.
Past performance is not an assurance of similar future returns. Nonetheless, the results of SSI’s backtest are significant, providing historical evidence that:
- A contrarian investment philosophy that favors investing in undervalued assets can deliver superior absolute and relative returns over time;
- SSI’s reliance on negative fund flows as a means to identify undervalued assets has been effective;
- Optimal holding periods — that allow cheap assets to recover their value — often vary from trade to trade; and
- Once an asset has been identified as undervalued, relative returns can inform an estimation of the asset’s optimal holding period.
SSI’s review in 2014 was prompted largely by SweetSpot’s mediocre returns in the past few years. We sought to answer the questions: Has the strategy lost its edge? What has changed?
The short answer is that nothing has changed but the times, and the times haven’t changed in a way that calls into question SweetSpot’s long-term prospects. Indeed, SweetSpot suffered more than one multi-year period of weakness in SSI’s 25-year backtest, yet we still saw outstanding overall results.
A key finding: Any strategy that seeks to outperform the market — even the best long-term performers — will experience occasional extended periods of underperformance. It is fortuitous that SSI has refined its method in a way that will better accommodate such periods. SSI’s research — on which this refinement was based — didn’t consider annual portfolio returns, but instead focused on individual trades. Addressing the latter just happened to improve our perspective on the former.
SweetSpot’s historical advantage is rooted in the foibles of human nature — specifically, the predictable irrationality of investors that is most pronounced as down markets approach a bottom.
“The irrationality of a thing is not an argument against its existence; rather a condition of it.”
SSI sought to determine if investors somehow learned from their mistakes and corrected them, neutralizing our trading edge. A survey of the relevant literature uncovered no reason to believe that they have. On the contrary, not only has the human species not evolved in recent years, there are indications of devolution.
*Except as noted, all reported returns are gross and assume the reinvestment of dividends and other fund distributions.
**Disclaimer: Backtests do not represent actual trading; they may not reflect the impact that material economic and market factors might have had on your adviser’s decisions if your adviser were actually managing the tested strategy during the backtest period. No representation is made that investors will see profits similar to hypothetical past results.
***In SSI’s backtest, yearly data refers to calendar years. Reported returns represent total returns. “Market Return” is: S&P 500 index, 1989-’90; Fidelity S&P 500 Index Fund (FSMKX), 1991-2007 (verified numbers provided by Investors FastTrack); and the All-Country World Index ETF (ACWI) thereafter. (Domestic stocks were SweetSpot’s most-representative market benchmark until 2008, when a majority of portfolio positions was first drawn from a global universe.)
Independent Third-Party Verification
All completed trades through 2011 were verified by an independent, third-party auditor here. All trades occurred in real time and are documented by account statements showing the settlement date, price, and amount of each trade; as well as records of email transmittals to investors (going back to 2003) recommending each trade.
Rules of Trading, 1999-Present
 Each January, buy the top-ranked sectors (or lowest ranked by the investing public) according to a proprietary formula applied to end-of-year fund data. With negative fund flows as its key metric, the formula seeks to identify the sectors that investors have abandoned, making them more likely than others to be at or near the bottom of their market cycle and thus ripe for purchase.
 Sell after 36 months, or longer if a position’s returns trail its benchmark.
i. Positions may be sold early under extraordinary circumstances. This has occurred once, in 2004: Insurance (Trade 10) was sold 2.5 months early after widespread fraud in the insurance industry was revealed.
ii. Until 2005, a position was sold after 12 or 24 months if it landed near the bottom of the annual rankings. This sell signal was abandoned when both real-time and backtested results showed that its use did not materially improve performance, and thus it increased turnover needlessly. The following trades were exited according to this now-defunct signal:
- Pacific Basin (Trade 1) was held for 12 months.
- Medical Delivery (Trade 4) was held for 12 months.
- Materials (Trade 9) was held for 12 months.
- Chemicals (Trade 12) was held for 24 months.
- Transportation (Trade 13) was held for 24 months.
iii. Before 2014, positions were sold on schedule even if they trailed their benchmark. Research performed in 2014 showed that such positions should not be sold, but instead should be given additional time to achieve excess returns. When real-time and backtested trades from 1989-2013 were combined and sorted by trade year, we saw that, historically, underperforming positions have been among SweetSpot’s strongest performers in their later years:
 Before 2014, if a current Portfolio holding was renewed as a buy, it was sold 36 months after renewal whether or not it underperformed its market benchmark.
- Leisure was initially bought in December 2000; renewed in January 2003; and ultimately sold in January 2006.
- Communications equipment and networking (Trades 21 and 22) were initially bought in January 2006; renewed in January 2008; and ultimately sold in January 2011.
- Wireless (Trade 25) was initially bought in January 2007; renewed in January 2009; and ultimately sold in January 2012.
- Europe SmallCap (Trade 36) was initially bought in January 2009, renewed in January 2010; and ultimately sold in January 2013.
Beginning in 2014, renewed positions were sold three years from when they were initially bought (effectively ignoring the renewal) if they had outperformed their benchmark in that period.
- Automotive (Trade 45) was initially bought in January 2012; renewed in January 2013; and sold in January 2015.
By definition, renewals have seen excessive selling while we held them. Usually they are poor performers at the time of renewal, deserving of extra time to work out their issues. Autos were an exception, a stellar performer despite its renewal status. A corollary to the wisdom of holding onto underperforming positions is that we should sell outperformers on schedule, meaning three years after they were first bought. In other words, letting performance inform our judgment of optimal holding periods cuts both ways. This view was borne out regarding Autos. The auto industry saw record-setting sales in 2015, but high expectations must have been baked into our sell price as Fidelity Select Autos gained just +0.2 percent that year.
 When two or more funds represent the same trade, combine their performance as follows:
i. Calculate the average returns (during the entire trade period) of funds that are held simultaneously or that significantly overlap:
- Fidelity Japan (FJPNX) and Fidelity Japan Smaller Companies (FJSCX) together represented Trades 6 and 31.
- Fidelity Select Retailing (FSRPX) and the PowerShares Retail ETF (PMR) overlapped as portfolio positions representing Trade 20.
- Fidelity Select Medical Delivery (FSHCX) and PowerShares Healthcare Services ETF (PTJ) overlapped as portfolio positions representing Trade 23.
ii. Calculate the sequential performance of funds with mutually exclusive holding periods:
- Fidelity Select Materials (FSDPX) initially represented Trade 19. Two years later it was sold in its entirety and immediately replaced with the PowerShares Basic Materials ETF (PYZ).
- Fidelity Select Networking (FNINX, now defunct) initially represented Trade 22. One year later it was sold in its entirety and immediately replaced with the PowerShares Networking ETF (PXQ).
- SweetSpot’s initial position representing Trade 24 was Fidelity Select Paper & Forest Products (FSPFX). FSPFX was sold shortly before Fidelity merged it out of existence in 2009; and was immediately replaced with the Guggenheim Timber ETF (CUT).
 Fidelity Select Consumer Staples (FDFAX) (Trade 5) was formerly Fidelity Select Food & Agriculture.
 Fidelity Select Materials (FSDPX) (Trades 9 and 19) was formerly Fidelity Select Industrial Materials.
 Fidelity Select Environment & Alternative Energy (FSLEX) (Trade 16) was formerly Fidelity Select Environmental Services.
 Fidelity Select Communications Equipment (FSDCX) (Trade 21) was formerly Fidelity Select Developing Communications.
 Historical price data for Fidelity Select Medical Delivery (FSHCX) (Trades 4 and 11) — unavailable at http://finance.yahoo.com — was obtained from http://www.fasttrack.net.
How Trades Were Identified
Trades 1-3 were based on information provided by Morningstar and Lipper, as reported in the Wall St. Journal in December 1998.* Trades 4-22 were based on a proprietary formula applied to data for all non-diversified Fidelity stock funds. Collectively, those funds served as a proxy for the broad stock market. Trade 23 and later trades were identified by applying the same formula to data for the entire known universe of non-diversified stock mutual funds and exchange-traded funds (ETFs), thus increasing the likelihood that market behavior would be accurately gauged. This change was made to accommodate:
- economic globalization that blurred the line between domestic and international sectors; and
- the proliferation of sector ETFs that compete with Fidelity, making Fidelity’s funds a less-reliable proxy for market behavior at the sector level.
*Damato, Karen, “Emerging Markets Trail Rally but May Be Bargains for the Intrepid,” Wall Street Journal; New York; by Karen Damato (Dec. 7, 1998). Start page: A11; ISSN: 00999660.
The most meaningful test of an investment’s past performance is the corresponding performance of its most-representative market index. Here, market-index benchmarks are represented here by corresponding index ETFs — and not the actual indexes — for two reasons:
- A performance benchmark enables one to compare a managed investment with a passive, market “buy-and-hold” approach. Yet market indexes as such are not available as investments, whereas index ETFs are. Fees that investors pay to index-fund sponsors are negligible, averaging about one-tenth of one percent annually.
- Investors care about total returns, which consist of share-price appreciation plus dividends paid. Yet market indexes only track changes in price; they don’t account for dividends, effectively understating their returns. Index funds, on the other hand, distribute all dividends to investors, and their returns are adjusted accordingly. It is ironic, then, that index funds provide a more accurate picture of an index’s actual returns than the index itself!
From 1998-2006, SweetSpot trades were selected from a universe of mostly domestic Fidelity sector funds, making the U.S. stock market — as represented by the S&P 500 SPDR ETF (symbol: SPY) — the appropriate benchmark for those trades. When SweetSpot converted to a global program in 2007, trades were benchmarked against the global stock market. The model SweetSpot portfolio made the switch to a global benchmark in 2008, the first year that a majority of portfolio holdings were drawn from a global universe.
Calculating Global Returns
Before 2008 there was no ETF available to investors that tracked the global stock market. Therefore, global market returns corresponding to SweetSpot trades entered in 2007 and 2008 are represented by the combined performance of the three pre-existing ETFs that together comprised the global stock market, in proportion to the size of their constituent markets:
- S&P 500 SPDR ETF (SPY) (U.S. stocks);
- iShares MSCI EAFE Index ETF (EFA) (foreign developed-market stocks); and
- iShares MSCI Emerging Markets Index ETF (EEM) (emerging-market stocks).
Beginning with trades entered in 2009, SweetSpot’s global benchmark became the iShares MSCI All Country World Index ETF (ACWI). Annual returns for both global and domestic stock markets since 2007:
- 2007: global +12.9%; SPY +5.1%
- 2008: global -42.2%; SPY -36.8%
- 2009: global +32.4%; SPY +26.4%
- 2010: global +11.8%; SPY +14.6%
- 2011: global -7.8%; SPY +1.9%
- 2012: global +16.8%; SPY +16.0%
- 2013: global +22.4%; SPY +32.3%
- 2014: global +3.8%; SPY +13.8%
- 2015: global -2.3%; SPY +1.2%
The identity of current Portfolio holdings is proprietary information available only to subscribers.
New SweetSpot investors are generally advised to invest in most or all current positions. Accounts should be invested in line with the stated risk tolerances and other individual preferences. More-conservative investors may be inclined to hold a percentage of cash in reserve, both to reduce volatility and to take advantage of potential opportunities to add cheaper shares. More-aggressive investors often prefer to remain fully invested at all times, and to let positions run.
How can one know that the returns presented above weren’t produced by one of many different investment strategies that ran concurrently? Couldn’t SweetSpot simply be a lucky contest winner?
Indeed, when SweetSpot began trading in 1998, the strategy had not proven itself in real time. SweetSpot’s manager first learned of the strategy in the course of conducting general research on the effectiveness of contrarian investing. SweetSpot was the specific contrarian approach that he was looking for, and he added it to his pre-existing mix of stock and bond “buckets” with the modest goal of diversifying his investment styles.
SweetSpot quickly revealed itself to be something special (see Trades 1-8 above). Not only did it shine in 1999 — the blow-off year of the 1990s bull run — but it weathered the ensuing three-year bear market without suffering a single losing trade. SweetSpot’s manager gradually shifted more and more of his assets to the strategy, and shared what he was doing with an ever-growing number of friends and family members who traded alongside him. By 2005 he no longer could justify investing stock assets in anything other than SweetSpot. Since then he has had most or all of his stock allocation invested in SweetSpot.
Disclaimer: SweetSpot Investments (SSI) is a publisher of investment data and research, and is not a registered investment adviser. Subscribers are responsible for their own investment decisions. Information that SSI provides to subscribers does not constitute a recommendation that any particular security, portfolio of securities, transaction, or investment strategy is suitable for any specific person. To the extent that any such information may be deemed to be investment advice, it is impersonal and not tailored to the investment needs of any individual. See more-detailed disclosures and disclaimers here.
Past performance is not an assurance of similar future returns.