MARKET FORECASTS = EMPTY NOISE

Manage
1

I was inspired by a comment from Brett regarding his giving up on market predictions in the “2017 was a great year!” thread started by Andreas. However, I got so wound up while writing my reply that I thought it would be better to start a new thread, rather than hijack Andrea’s thread.

Regarding Market Forecasting

I couldn’t agree more, Brett, and well said!

If you bear with me, I’d like to relate my background briefly and how I came to the same attitude as you, Brett: I started investing while still in high school (all the way back in 1974), so I have been at this game for 43 long, but fulfilling years. I went on to get an undergraduate degree in both physics and finance from Illinois and then an MBA from Princeton. I started my career taking a job on Wall Street, then out to Beverly Hills on the opposite coast as an Investment Analyst for Drexel Burnham’s junk-bond machine. Before Drexel crashed and burned, I moved to Merrill Lynch as a Senior Equity Analyst, which ultimately became a management role. I received an invitation to become a founding partner in a successful hedge fund, followed by founding several other investment-related enterprises.

So my point is, I’ve been around the ‘investing block’ and wanted to share some insights I learned in all those years about what I discovered (from the inside) about what the industry considers its most valuable product. Throughout all of those years of interesting trials and tribulations in the investment industry, I realized that the one over-arching theme of the investment world is PREDICTING THE FUTURE. It is the bread and butter upon which the industry thrives! Estimates, forecasts, predictions, prognostications, etc. - it’s a co-dependent relationship because the industry wouldn’t spend so much time, energy and effort producing it if investors didn’t think they really needed it. Likewise, investors wouldn’t believe they need the best, highest quality, most accurate estimates if the industry hadn’t trained them to desire it.

As an Equity Analyst for those big investment firms, one of the tasks I always hated (but perhaps the most important to my bosses) was that I had to create forecasts of each company’s sales, earnings, cash flow, and every other possible numerical derivation. They taught me that the best approach to making financial projections was to merely extrapolate current conditions, but that didn’t seem appropriate to me because there was no value added (and it may not be what I honestly thought). I knew that anyone could extrapolate a series of numbers, so what was I doing to justify my six-figure pay at age 24-25?

My manager told me, "Well if you think the fundamentals might justify a little more, add a bit to your forecast - or vice-versa. Just don’t get too enthusiastic or negative about a company - perhaps a leeway of 5% here or there is all. You can’t go wrong with middle-of-the-road. Just don’t submit a negative forecast, even if you really think it’s true unless the company is already on our “Slam List (of companies the firm covers that were already in trouble).” Wow! This conversation was a wake-up call that freaked out my young brain about how the industry actually worked. Sales were more important than honesty.

I couldn’t agree to that, so I decided to up my game with the future in mind so that in the future I would know how to get an accurate estimate. I spent a lot of time with several of the best in-house stock & market analysts and at industry conferences, met many famous investors. I tried and tried, and failed and failed to improve my forecasting technique. I learned all I could from the so-called ‘experts,’ but I saw even the best in the business get it wrong time and again.

How could this be? So I started asking them directly, and I soon realized it didn’t faze them at all that they were consistently wrong in their estimates – they knew the whole thing was just a game they played with the investing public. This game served to help them extract outsized compensation for their ability to offer pure, unadulterated BS professionally. This BS-factor is why almost all forecasts are bullish. Negative projections don’t sell anything!

MY INVESTMENT-LIFE CHANGING REALIZATION

Ultimately, the conclusion I came to was professionally LIFE-CHANGING: The average investor believes they will make big money if they know “what’s coming.” – That is, what’s coming in the very short-term (next week’s employment report), intermediate-term (next quarter’s earnings release by XYZ Tech Corp), or long-term (next year’s S&P Earnings Estimate). This need to know “what’s coming” has been ingrained in the psyche of virtually every investor who hopes to “win” at investing. Most think that figuring out the best prediction is the ‘holy grail,’ and if they can just decipher it, they’ll be rich. To serve this need to know “what’s coming,” there are dozens of magazines, websites, blogs, and TV stations that spend 24 hrs-a-day presenting people who tell you with confidence, impressive terminology, complicated charts and an expensive suit what’s coming tomorrow.

BOTTOM LINE: It’s all a bunch of white-collar, fabricated EMPTY AIR. It’s a scam, a con, a hustle, and a very lucrative one at that for the button-down grifters who have mastered it. Unfortunately, it lowers a genuinely challenging and worthwhile intellectual pursuit into something like an upscale Tupperware party. Even worse, it includes investing in the spectrum of gambling activities. The average investor hears a forecast on CNBC about the Materials sector getting hot, he reads about ABC Technology, Inc. is coming out with a new, breakthrough product that is predicted to change its industry, or Barrons lists Brazil as the #1 country for investment in 2018. As a result, Mr. Average Investor “makes a bet” (literally, those are the words used) on equities in those areas that are forecast to be a good money-maker.

And worse – it takes advantage of a public that is just hoping to get a leg up in saving well for retirement.

Like you Brett, I left it all behind, and my investing world changed dramatically for the better. I haven’t seen a show on CNBC, Bloomberg or any of the others (whatever they are) in decades. I have no desire to pick up a Wall Street Journal, Barrons, or Forbes magazine.

I focus all my attention on researching, testing and applying quantitative tools, such as those offered on P123, to the challenges of identifying real-time signals of relevant changes in a sector’s Earnings Yield, an economic measure – such as unemployment, market internals – such as New High vs. New Lows, and a plethora of other, quantifiable econometric and market measurements.

When I decided that I wanted nothing more to do with predictions of anything having to do with investing, it was a real breakthrough for me. I honestly believe that the FIRST RECOMMENDATION FOR SUCCESS I would make to a young, beginning investor (or even an old one who wants to turbocharge his/her investing game), would be to completely ignore the narrative STORIES that are told about companies, people, events, or markets and also to disregard all forecasts, estimates, and prognostications. I would recommend that the investor pay attention to FACTS (such as current Price/Cash Flow Ratio, or Trailing Twelve Months (ttm) Income), and to NOT base their investing decisions on other people’s speculations about the future (i.e., market forecasts, earnings estimates, etc.).

Quitting the industry’s stories and forecasts cold turkey is not easy to do because we are all subconsciously drawn to guiding narratives. Stories give us a feeling that helps everything ‘make sense’ in this crazy world and provides explanations in a way that allows things to fit into our world-view. The problem with stories is this exact issue: we can twist or color a story in our minds into whatever we want it to be. Not so much with facts, such as a company’s 3rd Quarter 2017 Gross Profit.

Stories provide a sense of comfort to everyone, but in my opinion, you’ll never be a truly successful investor until you abandon the narrative stories (that don’t really explain) and the endless stream of predictions (that are rarely accurate). Instead, master the identification of meaningful quantitative market signals and develop an effective way to convert those signals into algorithms that a computer can understand. A young person armed with these skills can truly take on the investment world of the future and come out a winner!

Chris

(P.S. - Does anyone want to bet against me in saying that Warren Buffett never uses market predictions, business TV shows, or analyst’s earnings estimates to pick a stock?)

2

Christopher, thank you for sharing your experiences,I could not agree more :slight_smile:

3

Hi Chris, Great post and thanks for the perspective. In particular this line resonated with me: “I haven’t seen a show on CNBC, Bloomberg or any of the others (whatever they are) in decades. I have no desire to pick up a Wall Street Journal, Barrons, or Forbes magazine.”

I struggle to curate what news I hear and pay attention to, but am getting better at tuning things out. I still enjoy reading and learning about companies I have interest in - and I think I sometimes find useful discussions - but it is effortful to avoid going down the rabbit hole and being distracted by so many stories and comments designed to pull my attention. I find it helpful to remind myself of Tristan Harris’s warning that behind all screens on the internet are thousands of the smartest people in the world with incredible resources, behavioral data, and laser focus on grabbing and keeping my attention - however they have to do it. I’m manipulable, and “they” know the buttons to push. The only solution I’ve found is to limit my exposure and to constantly be on guard (which is a hard way to go about the day).

Interestingly, I do find sentiment data useful in modeling, so I hope all the analysts don’t go away :wink:

4

Very interesting insights indeed, which are confirming all my own experiences and suspicions. Very often nothing more than modern snake oil salesmen in expensive suits, trying to exploit an unsuspecting public by selling false hope and cures.

Many thanks for sharing, Chris!

5

Good post, Chris. I agree there is a lot of noise out there about “the future.” However, let’s not get carried away . . .

“Don’t get me wrong I am a strong proponent of market timing, I just let the markets tell me when it is time to do it.”

Anyone who engages in market timing makes top-down predictions about the future of the market. If I am 100% long and then decide to put a hedge on because of some breadth or unemployment or yield curve signal, I am predicting the market will go down. If I take a hedge off and go back to 100% long, I am predicting the market will go up. I make money if these predictions are accurate, and lose money if they are inaccurate.

“When I decided that I wanted nothing more to do with predictions of anything having to do with investing, it was a real breakthrough for me.”

Any “active” stock investor (i.e. hoping to beat a benchmark as opposed to passively track one) who buys a stock is making a prediction (based on some type of information) that the stock will outperform. The ability to beat the market depends on the quality of the information upon which this prediction is based (and the quality of the execution of the trading plan). This is why we have the “information ratio.”

Any “tactical” asset allocator (i.e. hoping to produce better risk adjusted returns than say some static 60-40 stock-bond benchmark) who moves out of one asset class into another is making a prediction (based on some type of information) that the new asset class will outperform the replaced asset class.

The whole P123 website is designed to provide informational tools so we can make these bottoms up “predictions” on stocks and asset classes, not to mention top down predictions on market timing.

6

Hi Parker,
Thank you for your post you make some interesting points.

“Anyone who engages in market timing makes top-down predictions about the future of the market. If I am 100% long and then decide to put a hedge on because of some breadth or unemployment or yield curve signal, I am predicting the market will go down. If I take a hedge off and go back to 100% long, I am predicting the market will go up. I make money if these predictions are accurate, and lose money if they are inaccurate.”
We may be parsing words here but I would make the differentiation between predicting (i.e forecasting) and hedging(decreasing risk exposure). I do not pretend to know the future nor am I predicting that the market will go down. If my perceived risk has gone up(based on historical probabilities not certainties or opinions of others) I may wish to decrease my risk exposure.(ex. when markets become more volatile or begin to roll over). That is very different from predicting market direction.

I think we can all agree that the way forward is doing our own research and not listening to the Noise.

7

Grinold & Kahn make the the point in Active Portfolio Management. Paraphrasing at P. 261:

Active management is forecasting. The consensus forecasts of expected returns lead to the market or benchmark portfolio. Active managers earn their title by investing in portfolios that differ from the benchmark (i.e. differ from the consensus) which implies forecasting different expected returns based on some sort of information.

It’s a great book in case anyone is interested.

8

But I hear what Chris is saying. He is contrasting coming up with a price or numerical based forecast (e.g. stock XYZ has a $100 price target or will earn $5.00 per share next year) vs. having a bias and positioning oneself for it (e.g. I like stock XYZ based on these 7 factors and will go long while I have no idea where it will go on an absolute level).

A factor model is built because we hope we can find out-performance with less risk. This might simply be dropping less than the market in a correction. But I think the type of forecast Chris is talking about would be if you said, “My factor-based model is set to do 50% returns in 2018 based on these fundamentals and complex forward looking calculations.”

9

Noise is an interesting thing. There are 2 kinds for us.

There is random noise and “Deterministic” noise. The difference is important.

For us, deterministic noise is the fluctuation in stock prices that have reasons but these reasons are not built into the model yet. We (I at least) always want to get a better model that takes into account the fluctuations that have reasons and that can be modeled at P123.

But when I try to model this deterministic noise I could get it wrong. My model for stock prices related to unemployment (say) may be just plain wrong: maybe I use the wrong moving average or unemployment does not matter unless the economy is also crashing. Because my model is not correct yet the determinist noise remains along with the random noise. Noise I keep trying to fit.

I can fit some of this noise with a bad model. But this noise changes going forward. The bad model–with a poor fit to the things that are really driving the market—remains. This is overfitting.

I believe some people can time the markets. If I try (or when I have tried) it has ended up in an exercise in overfitting. It is probably going to require someone smarter than me or someone with better data and/or a better computer than me to get something that is not overfitted and harmful. Some of these smart people are probably at P123. I’m just not one of them.

-Jim

10

Prediction, outlook or “fatidic speculation”, what’s the difference?

Chris, when you posted - not that long ago - that "SMALL-CAP STOCKS ARE GOING TO PLUMMET STARTING MONDAY (https://www.portfolio123.com/mvnforum/viewthread_thread,10144_offset,10), was that an example of the predicting you warned against?

In any case, I do think predictions made in the popular press should be ignored since they generally lack context and follow up. With experience, I’m finding it easier and easier to ignore what’s published in the press and to rely on the data found on P123 to refine my own understanding of the economy and stock market.

Walter

11

My large cap models performance largely this year was based on some brick and mortar retail chains ($ROST, $BKE) that had the audacity to exist in a world with Amazon. These are not overly complex models, just basic quality and value. Absolutely no way I would have been invested in these names if I made stock picking based on listening to financial media narratives (and there is no good media narrative around a company that sees a huge upward price trend based on the fact that it was previously priced for an awful outlook and merely turned out to be just slightly bad). Last year my returns were largely based on some regional banks who saw a huge spike upwards after Trump won, as no one was predicting it. This makes me appreciate that I’m not in the stock prediction picking game anymore.

The only thing that drives me more crazy than the prediction game is the CEO Celebrity game. I’m not saying that management or a competent CEO doesn’t matter, but the way the media turn these people into human interest stories where they’re some all knowing oracle of business is kinda funny. I remember when Netflix’s CEO yo-yo’ed from visionary to idiot genius in the span of about 6 months. 1.5 years ago the Chipotle CEO was a genius, and now most of the financial media claims he should be replaced by someone who knows what they’re doing. It’s largely nonsense and Cult of Personality based, but I’ve heard many people say they would never trust a quantitative approach because it doesn’t properly take into account management and leadership. I’m not Warren Buffett. As a retail investor, I don’t have access to any of these CEOs and even if I did I probably wouldn’t be able to tell anything about them in any meaningful way. I would probably be subjecting myself to about umpteen potential biases that would cloud my decision if I did.

12

I have been on what I think is a pretty steep learning curve, being only actively investing/trading for about 2.5 years. Looked at ALL types of trading but I’m sure there are plenty of other techniques out there that may or may not work that I haven’t even seen. I will say that in that time instead of trying to predict if the market will be up or down, or listening to “what 2018 will bring for the market”, I have made and backtested a set of rules for trading. So far they have proved to be profitable. No one here or anywhere else in my opinion can catch every little up and down the market makes. Just have a set of rules on when to get out (or be comfortable for a wild ride at least) and be sure you can live with the “what if” scenario. However, if someone does have a strategy that wins each and every single day, I’d like to hear it!

13

So Chris, were you behind the scenes re: any of that garbage Alan Flans pushed at me?
:slight_smile:

14

Chris,

Excellent post. Agree 100%.

Leda Braga, a quantitive hedge fund manager, made a great observation about our industry, namely that quantitive investing is all about changing the emphasis from forecasting to risk management. That is a very powerful insight. “Fundamental” investors who justifiably find fault with quantitative investing fail to see the serious limitations with their primary tool, i.e., forecasting. How can one accurately forecast stock prices, on a reliable timetable, when one has to rely on other investors whose actions are so fickle and unpredictable? At the end of the day, it’s a matter of choosing your poison. For my money, a quantitative approach that, as much as possible, avoids the sins of data mining, over-fitting, etc., is far preferable to a narrative-based methodology.

And about Warren Buffett and Charlie Munger? They avoid forecasts like the plague. They know that the secret to accumulating wealth is through a single-minded focus on buying great businesses at good prices, and tuning out everything else. This interview of Charlie Munger says it all: https://www.youtube.com/watch?v=WlC40B9qZ20

Ed

15

Hi Marc,

Sorry, but I’m not sure what you are referring to Marc, but I am intrigued. I had left well before the Federal lawsuits scooped up Flans and all the other Drexel honchos. But that was a loooooong time ago and these days I can barely remember what I had for breakfast. :wink: To what are you referring, exactly?

Chris

16

Thank you for all the interesting commentary on my post. I wrote it with the hope that it might make some think twice before relying too heavily on the industry’s ubiquitous prediction machine. Here are my responses to some of the comments:

So true, Michael. The prediction and ‘story’ machine is well financed and employs marketing and human behavioral experts who know exactly how to attract viewers/readers and get them to focus, worry, engage, buy/sell, view ads or any other objectives they may have. One of the most common purposes is simply to generate activity, i.e., buying and selling stocks, which produces commissions and fees, which drives the industry’s money machine. Much of the investment world is focused on producing activity, as opposed to cultivating wealth for investors. The Obama administration passed the Fiduciary Responsibility rules for stockbrokers and other professions, which for the first time in history, required brokers and advisors to have the client’s best interests in mind when making recommendations. For the last century, brokers and advisors were far more interested in generating trading activity and commissions for themselves than helping the client save for retirement. However, the last I heard, the “business-friendly” Trump administration canceled that rule before it went into effect on Jan. 1.

Kurtis posted an excellent response that speaks for me on this one:

Exactly, Kurtis! I have found that the most effective approach for consistent profits is to arbitrage an equity’s current price with what analysis shows is its current ‘fair’ price. This is a classic value-investing approach based on Graham and Dodd’s principles laid out more than 80 years ago, in which we can determine the reasonable value of an equity based on a number of approaches, such as the company’s liquidation value, its fair value in an acquisition, its value by a number of measures when compared to its peers, etc. Estimates of the future never enter the picture for me. This value-approach is also what has created the wealthiest man in the world from investing, Warren Buffett, and a long list of investing legends, including Seth Klarman, Whitney Tilson, Joel Greenblatt, James O’Shaughnessy, and many more.

When there is a distinct difference between a stock’s current price and its current value, there is an opportunity to close that margin, and the investor/arbitrageur can reap the gain. I also apply a particular momentum requirement that ensures the stock’s undervaluation is beginning to be recognized by the rest of the world, and I review its technical chart (using Bollinger Bands, CCI, PMO, and other indicators) to make sure its price is at a spot that is opportunistic for near-term gains before I make a purchase.

Actually no, Walter. If I recall correctly, that post was made based on my observations of technical charts that showed current over-extended, technical price conditions in the small-cap universe (distinct from ‘overvalued’) that suggested that a pullback was likely. I didn’t use forecasts, earnings or other estimates, or any other prognostications, (which, I know from experience are just extrapolations of past trends, often with a favorable boost added).

So true. This is one of the techniques that the TV business news channels use to attract viewers. People are naturally drawn to stories about famous people, and CEO celebrities are no exception when it comes to pulling in business viewership. Another well-worn technique is to use negative stories to attract viewers. Humans have a peculiar attraction to ‘end-of-the-world’ (or in this case, ‘end-of-the-market’) stories. Worry is something that aided in survival as we evolved over the eons, and stories that worry people draw greater attention and thereby, sell more ads.

I read that during the Financial Crisis, CNBC viewership skyrocketed more than 1,000%. Another example is that ratings for cable news went through the roof the first months after Trump took office, ostensibly because a large percentage of American citizens were positively worried about what was going to happen. While those numbers have somewhat declined now that people have caught on to Trump’s reality-show antics, he knows how to push people’s buttons and get more attention whenever he wants it. The point is that all of these attention-grabbing methods, when applied to investing, are intended to manipulate the viewer/reader and serve the needs of the investment industry rather than the investor. I find it’s best to shut them out of my life so they can have no subconscious influence on me.

Exactly! Only I would add that in addition to changing the emphasis to risk management, quantitive investing can also identify opportunities in a new way that can seperate us from the average investor who largely lives in a herd mentality. Rules-based investing does not rely on other people’s biased opinions (that is, if you leave out the factors under the ‘Estimates’ heading in the P123 factor menu).

EVIDENCE: If you want some indisputable evidence about just how inaccurate and worthless analyst’s estimates are, just visit FinViz.com. Just below the section showing fundamentals, the site lists the ratings and forecasts that a smorgasbord of highly paid analysts published for the company’s stock price. Each recommendation has the date it was issued, so you can see what their forecast was on, say, June 2017, and then see how the stock performed in the subsequent six months. You’ll find that there is rarely a correlation between forecast and performance, often with a stock doing the opposite of what was projected by the analyst.

Occasionally a forecast will be accurate, especially in the case of a company or industry that is heading down the toilet (think Energy sector in mid-2014 to Jan. 2016). Those should be easy ones, but even so, most analysts can’t get it right if conditions change (because they are extrapolating the past). For example, see the page for Valero Energy (VLO) in FinViz. In October-November 2017, there were four out of five analysts who were downgrading their forecasts for the company – even though a rally in the stock began in June. (My guess is they were still extrapolating from previous year’s numbers.) My P123 valuation-based S&P 500 trading system recommended the company on September 4. I analyzed VLO’s chart, saw that it was on the cusp of a significant gain and bought it the next day at $68.01. The stock has profited 37% for me in four months – not as incredible as some of my other trades, but not too bad at all.

I am winding down my long career in the individual stock investment world and thought I might, from time to time, pass on a few of the (sometimes expensive) lessons I have learned over those years. I spent quite a bit of time on the inside of the investment industry and discovered, somewhat uniquely, how the estimates and forecasts are derived. Brett’s comment in another post about how he had profitably abandoned projections prompted this one from me, which I hoped might contribute to someone’s investing success. I’m glad to see that it seemed to be of benefit for some.

(By the way, if there is a desire in the community for me to share what I analyze in a stock’s chart before purchase, I would be glad to do that. As I said, I am winding down my stock-based investing and have no problem sharing, in this case, the specifics I look for that dramatically reduce the chances of picking money-losers.)

Chris

17

Chris, thanks for sharing. Would love to hear more about reducing the chances of picking money-losers.

18

Chris, I’m very much interested in ideas on technical rules. I keep trying different ideas but have quite a bit of trouble (“futility” might be the word) finding technical rules that help much at all in the context of fully developed models. I can find technical rules that “work” well independently, but in the context of my ranking system once all factors are applied I can mostly remove them from the model without issue and tend to do so. I’ve noticed I think for shorter < 4wk trading that some of the technical rules related to mean reversion including BBand, relationships to MAs seem to be helpful, but when I push holding periods out to 4-13 wks the advantages fade or even start hurting a system. I’ll keep looking though as it seems there should be a way to get advantage from price wobbles given the mean reversion tendencies. I’ve started wondering if maybe putting the technical rules in the screen (and checking for buy signals on top ranked stocks daily) instead of in the ranking system itself might be a more fruitful approach?

But anyhow, I guess short answer is “yes” I’m interested in the topic.

19

Great insights Chris. You mentioned being in Beverly Hills for awhile - if you’re ever in the LA area and have some free time on your hands, let me know and we could grab some coffee, talk investing.

20

Anyone interested in a SoCal P123 meet up?