Answer: To let a community of investment professionals, who have more up-to-date information on the 500 stocks than you do, and the resources to put that perspective together, do it for you.
This is not a new idea. Back in the 1980's before there were any ETFs, Professional Investment organizations wanted to have the means to invest in the index as an equity instrument. Major brokerage houses, predecessors to today's "investment banks," put their quant analysts to work to create "baskets" of specific stocks that had price movements highly correlated to the index. They created proportionally-engineered investments in fewer than 100 stocks that could proxy for the market average, instead of having to manage all 500.
It was a logical next step to try to forecast coming prices of the average by forecasting the basket's components. But no marvelous reputations were able to be built on this idea.
Now the challenge is easier. Just forecast the index-tracking ETF, SPDR S&P 500 (NYSEARCA:SPY).
Some of the same proportional-holdings "basket" approaches are in use today. Yet, we still don't see any "King of the SPDR-Price Forecasters" reputations being trumpeted.
There continues to be no end of relatively vague opinions about the S&P 500 index's coming price, based loosely on a plethora of S&P 500 EPS estimates. It seems the public's memory retention of EPS numbers is much less keen than their recollection of errant price forecasts.
Every professional investment manager feels he/she has to have a "macro" market forecast, but since it is so easy to take a market position these days by investing in ETFs, precise definitions of the outlook are easily deflected into far more interesting specific investment ideas. Still, the macro need continues to lurk in the background.
Instead of the notoriously fuzzy-result approach of guessing about an S&P 500 price based on a guess of an "appropriate" market P/E multiple, further based on an assembled aggregate of (basket-component) EPS guesses, let's look at a far easier to track approach, using "Intelligent Behavioral Analysis."
That is accomplished by looking to see what the market-making community is willing to pay for protection against unwanted market-average index price moves, and how they are structuring their hedges. That need arises continually in the course of putting firm capital at risk as they provide "market liquidity" to get client trade orders "filled." Correctly done, analysis of the hedges tells just how far up and how far down they think prices have a reasonable chance of going, in the next few weeks to months.
Currently there are several S&P 500 ETFs. If they don't all give the same forecast, which may be the best to use? Here are the choices: Grand-daddy of all is the SPDR S&P 500, the ETF that started that whole movement. Next came ProShares Ultra S&P 500 (NYSEARCA:SSO), which introduced the notion of 2x leveraged price motion into the game. A new competitor emerged on a see-you-and-raise-you-one basis with Direxion Daily S&P 500 Bull 3x Shares (NYSEARCA:SPXL), and ProShares countered with its own 3x Ultra S&P 500 (NYSEARCA:UPRO).
We exclude the inverse, or short-position structured S&P 500 ETFs since they have their own peculiar liabilities, and do not add anything useful to our exploration.
The following table provides a conventional analysis look at the choices, along with the index itself (SPX), the basis for comparison.
SPY, intended to be priced at 1/10th of the index, has achieved a sizable following since its introduction ten years ago, with nearly $150 billion invested. That compares with the largest Equity investments of Apple (NASDAQ:AAPL) and Exxon (NYSE:XOM) at $438 billion and $380 billion, respectively. But SPY's nature is quite different, with a daily trading value of $18 billion, meaning its entire holding can be turned over in 8 trading days, compared to 71 days for AAPL, and 388 for XOM. A market year consists of 252 trading days.
So SPY is a highly-liquid chip for portfolio managers seeking market exposure, influenced daily by quite short-term attentions. Its levered variants understandably have even shorter attention spans. And deservedly they have attracted less capital and fewer, if more intense, attendees. Their turnovers run only 2½ to 3 days.
But with all this liquidity, how well do they mirror the Index itself? That is revealed by the table's column of 3-year R2, or correlation of price change moves by each ETF with those of the index. A 100.00 correlation is perfect, and 99.+ is too near perfect to find any argument with. All track the daily directions extremely well, as the following picture illustrates. The yellow SPX line totally obscures the purple SPY line. And the others simply reflect the impact of leverage over time.
Indeed, this picture illustrates the desirability of using the increased focus of the leveraged ETFs as a more precise measure of the index's price change. That sensitivity is further heightened by the volume of daily trading relative to the capital committed to the instruments, the 2-3 day turnovers.
We will pursue that notion as we examine the forecasts of market professionals, as implied by Intelligent Behavioral Analysis. As suggested by the above picture, the different ETFs, and the Index itself, produce slightly different appraisals. Let's look at how those price range forecasts have trended in recent days for the Index, SPX.
The Index is not directly tradeable by the investing public as an equity investment. There are index futures that are tradeable as a commodity, and market professionals tend to focus on those because of their economy of capital commitment. Along with that economy comes an intense day-by-day management requirement that professionals are driven to accept, but is usually appraised by public investors to be too onerous and complicated, and thus dangerous, to be bothered by.
A more manageable instrument is SPX listed options, which are exchange traded. But lacking any direct access to the underlier, the index, they appear to most public investors as an unduly complicated avenue to pursue, particularly when SPY tracks the index essentially perfectly, and options on SPY are actively and simply traded.
The "inside" S&P market and forecasts
So there are really two markets of trading activity for the S&P 500, an "inside, wholesale" one used almost solely by (sell-side) market professionals on SPX, and a "public, retail" one used by investment management professionals and individual investors, operating around the SPY ETF.
Since SPY tracks the SPX so near-perfectly, this condition persists because differing operational considerations make it practical. For the sake of completeness, we will examine both markets and their forecasts.
Here is what the hedging activities in the SPX market imply as price range forecasts, day by day over the past 6 months:
(used with permission)
You will note that SPX's forecast ranges, the vertical lines, actively move as new influences are considered by pro market participants, always attempting to sense what may be coming next. The heavy dots are the end-of-day [eod] prices at the date the forecasts were made, and separate the forecasts into upside and downside potentials.
During this 6 months, the balance between upside and downside favors the upside, but this is not always the case. We measure this balance with a metric we call the Range Index, whose numeric value is that percentage of the whole range that currently lies below the market quote.
The Range Index, when a small number, indicates limited downside likelihood, and an expansive upside potential. The range size, or overall uncertainty from bottom price to top price, gets lost in the Range Index metric, but is picked up directly when examining the price gain prospect. There the top of the forecast range can be used as a sell target. (Can you say "return potential"?) Similarly, the downside complement ("risk exposure"?) is the price drawdown threat (to a long investor).
As market prospect perceptions evolve, the upside-to-downside balance may shift, producing Range Indexes of varying sizes. Over the past 5 years (1261 market days) this is what the distribution of SPX RIs (now at 18) has looked like:
The sell-side community is a far more tightly-knit body than are the buy-side professionals, and they tend not to have any meaningful presence of less-informed and well-capitalized individual investors. So their outlook may differ somewhat at times from the "other side of the street."
That side is best reflected by the past 6 month forecasts for SPY, seen here:
Its current Range Index, 43, is less optimistic than the SPX's 18, and its past 5-year distribution has looked like this:
The principal difference in the two outlooks lies in the downside, where a SPY downside of -5 ½% is being held possible, instead of the SPX -2.2% considered likely enough for pros to spend (otherwise profit) money protecting against.
The difference, of course, is that if you as an individual no longer have any influence over your government's actions and policies, it is easier to be more fearful than those who do.
What benefits can these forecasts produce?
This next table probes those differences of range size and current Range Indexes for the Market Index and the 4 ETFs in question. It also presents the history of outcomes of prior Range Index experiences, when put under a standardized portfolio management discipline as a comparative test.
The SSO's range size and Range Index are not significantly different from those dimensions in SPY. This is not comfortably logical, but serves to remind us that we are dealing with human perceptions (a "soft science") rather than with nature's physical outcomes (the "hard sciences"), where surprises are less frequent.
In contrast to SSO, range sizes for the (3x) ETFs are notably larger than SPX or SPY. But the downsides are quite similar in size to that of SPY. Appropriately, the upsides are more expanded, reflecting the leverage potentials. But again, there is no hard and fast "rule" that an ETF built to function with (3x) the price sensitivity of a single ETF, has to be viewed as a "full" (3x) larger variance, by observers whose minds are grappling at the same time with other mysteries.
Now let's look at the right-hand side of the above table. It portrays what has been the experience averages of that number of each ETF's prior Range Indexes (at the current level) when put in a test-tube where the top price of the forecast range at the time was set as a sell target, and each experience was limited to a maximum holding period of 63 market days from the date of the forecast.
The test's operating rules force a closeout of any position on the first day that its eod price is at or above the sell target, and failing that occurrence, closed at the end of the 63rd day. Win ratios measure the proportion of closeouts at prices higher than position inceptions at eod prices the day after the forecast. Annual rates of return are calculated as the nth root daily average of days held, of the geometric mean of all sample RI experiences, then raised to the 252-day market-year power. If the calculation specifics are puzzling, just accept that results have been correctly and carefully produced.
Going through SPXL's data as an illustration, its current Range Index of 32 has appeared 116 times in its available 1139 days of the past 5 years. Of that sample, 7 out of every 8 (87%) produced a gain at their closeout, and the average net gain of all 116 was 8.7%. The average number of days required for capital to be committed was 33, so an annual rate of gain of 90% was achieved.
Qualitative evaluations of these results include considering the win ratio (87%), the annual rate of net gain (+90%), its prospective reward-to-risk ratio by a comparison of the sell-target gain potential (11.7%), divided by the average worst-case price drawdown from position commitment price in all 116 sample experiences (-10.5%) or 1.1. And the history of average % payoff gain achieved (8.7%) divided by the promised sell target gain of 11.7%, or a figure of merit (or disappointment) of 74%.
But not all portfolios serve the same mission
Differing investment objectives can make these various qualitative measures more important or less significant, and varying thresholds of significance can enter into the investor's preference choices. So we try to provide as rich a description of what has happened in the past as is practical for comparative purposes.
One last dimension that is often helpful in developing perspective is to simply look at what has been the actual price performance experience over the last few years, following many levels of Range Indexes.
In illustration, here is the array of annual rates of price change for UPRO's 1065 available Range Indexes in the past 5 years. The table is set up from RI extremes at top and bottom, to cumulate to the mid-row blue line average of all experiences. The columns measure increasing-length weeks of 5 market days, from the date of each forecast.
The magenta total of 261 Range Indexes (at 35 or less) experienced 1-month (20 market days) price changes at an average annual rate of +123%, at two months after forecast at +68% and at 3 months of +85% or +83%. Its typical annual rate of gain over the 4+ years was around 42% from all Range Index experiences. Range Indexes over 40 experienced price gains at only 3/4ths (+29%) of the overall rate when held for 3 months.
Range Indexes lower than 38 tended to provide price change opportunities averaging significantly above the blue-line norm.
Comparable measurements for SPXL have not offered as attractive prospects in the 2 to 4 month holding period experiences at present RI levels, as have UPRO.
Intelligent Behavioral Analysis, relying on the informed judgments of economically highly motivated, experienced professionals, can and do lead to productive preferences between selection candidates for investment or reinvestment of capital in portfolios of equity and ETF holdings. The self-protective actions of these individuals provide insights into information often not available to "buy-side" investors. Important in that process is a sense of how "order flow" of price and volume emphasis by buyers and sellers of specific securities is trending. Aggregated, such senses in key "market-leader" names undoubtedly has future price influence on ETFs tracking various market indexes.
The implied forecasts from this style of analysis has over a decade of live-experience evidence that such insights can be predicatively productive.
Additional Disclosure: The author has an investment interest in the website blockdesk.com which, while not yet open to the public, is in conversion from being a delivery medium of information to institutional investors to a new life of providing similar help to do-it-yourself investors. Both brief and extended-time subscriptions for single or multiple issue inquiries should be at quite reasonable and manageable costs for individuals. Announcement of its opening is hoped for in the 4th quarter of this year.
Disclosure: I 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.