Its Friday, and has become my wont, this is the day of the week I like to kick back, wax philosophical about various thoughts kicking about me noggin.

One of the things that I have been noticing of late is the way so many people seem to confuse facts with forecasts. Twitter is rife with this issue, as 140 characters plus a dearth of actual thinking turns out to be a bad combination.

Where I see this more than anywhere else in the investing world is the significant confusion between Possibility and Probability.

Stated differently, more than a few investors are lumping all of the non-zero probabilities together as one similar trade. They are not, for reasons I shall soon make clear, the same thing. This has been true with many people, from Pension fund managers down to mom & pop investors.

Let’s define the terms Possible and Probable so that we all begin with the same basic understanding of these words. Something that is Possible may — or may not — come to be. There is no statistical insight provided, it is merely an outcome that may or may not come to pass. It is possible that you could get hit by lightning, or win the lottery, or marry a supermodel. When we describe something as possible, we mean there is a non-zero likelihood of that outcome — it could happen; we just don’t know if it will or will not, but it might.

What seems to confuse investors about possible is the statistical likelihood of occurrence. Can a company on the verge of bankruptcy go on to become the biggest company in the world? It is possible — and Apple (AAPL) did just that over the course of a 15 year from 1998-2012. But think about all of the many tens of thousands of companies that have been on the verge of going belly up. Is it possible that they could do the same? Well, the answer is Yes, it is theoretically possible — but not very probable. And indeed, experience teaches us that most insolvent and near insolvent firms actually do go bankrupt eventually.

Probability is a the term for the branch of statistics dealing with chance and outcome.

Possibility is binary — can this happen or not?

Probable is more nuanced mathematics — there is a n% chance of a given outcome, where n = a number between 0-100.

Anything that is Probable must by definition be Possible; However, not everything that is Possible is going to be Probable.

And therein lay the rub.

Lately, I have seen this error come up in two areas — long term Pension Funds, and Venture Capital. The most common mistake is to conflate the two — mistake what is possible for something that is probable.

With Pension Funds, we have an issue of national import in that many of these defined benefit plans are significantly underfunded. To catch up, some of these have been directed to add more Hedge Fund exposure. There is an accounting reason — but not a good investing reason — for this based on idea of expected returns. While the tiny minority of funds that create Alpha means that there exists a theoretical possibility for Hedgies to help pension funds make up their shortfalls, the high fees and under-performance makes it rather improbable that this will occur. This is to say nothing of the long odds that Pension funds will pick the next Ray Dalio or Jim Simons.

Hence, the possibility of better performance perversely leads to the probability of worse performance.

In Venture Investing, the dream of 100X returns is the tantalizing possibility that keeps investors coming back to the VCs — despite the very long odds that your fund will find the next Instagram. The poor track record of the past decade, the limited ability to select VCs that will outperform, means that the dream is a possibility, but the odds are improbable.

Whenever asked about Angel or Venture investing, I say the same thing: Assume it goes to zero. That is the highest probability, most likely outcome. Invest no more money than what you are comfortable lighting up in flames (i.e., going to nothing).

Who should focus on the Possible? Writers, dreamers, artists, futurists, designers, technologists, entrepreneurs — the people who imagine the future. Nothing I wrote here should discourage any of these sorts of folks from pushing for the possible, (however improbable). It is how nearly all of human progress has been made, pursuing the improbable.

For investors, however, you are best served by sticking with the Probable. Understanding the difference will save you a lot of capital over the long run.

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

### 29 Responses to “Possible versus Probable”

1. rd says:

I am seeing lots of financial discussions these days talking about real long-term portfolio returns in the range of 4% or more. I have also seen recent discussions on the equity risk premium also in that range or higher. These are possible.

However, we have a world with 0% short interest rates, 2% long bond and equity dividend yields, and Shiller CAPE and Tobin Q values that are in the top quintile of their historic range. Is it probable that the historic mean or higher returns will occur over the next 20 years? I think that probability is very low. In particular, dividend yields on stocks were almost always higher than long bond interest rates until the 1940s. The switch to today’s condition where dividend yields are equal to or lower than the long bond rates has likely lowered the equity risk premium by 2% annually over the past 60 years.

It is possible, but improbable that most pension funds will be able to successfully bet that the financial condition of their company will improve so that they will have the ability to cover a significant amount of future pension benefits using current contributions at that time. I think these pension plans are making a huge bet that much of society is currently unaware of. Hopefully, the PBGC stays in good management hands because I think it will be needed frequently over the next couple of decades.

A retiring individual will have a very low probability of effectively implementing this approach as only a very small portion of the population will be able to continue to bring in income through work in their late retirement years. As a result, a decade ago with two decades to go before retirement, I elected to focus on maximizing retirement savings (@25% of gross income – up from 10%-15%) to the extent possible to cover the probable shortfall of investment returns below the long-term averages. However, I think the Millenial generation will see the long-term expected returns revert to the mean in their peak earning years a couple of decades from now, so I have been recommending 15% a year for a retirement savings rate for my kids within a year of them entering the work force. It is probable that this will work for them.

At this moment in time, both society-wide hyper-inflation and society-wide deflation remain on the table as possibilities, but I don’t see either as probable in the US over the next handful of years. But one must still keep an eye on those possiblities.

• willid3 says:

considering we can’t predict even what will happen tomorrow, let alone what will happen in the next hour or so, never could take longer term predictions (like x will happen in 10 years!) with a grain of salt (extra large size!). the best predictors we have today (with the best records too) practice with probabilities of (like it might rain, probability of 10%. what they dont tell you is that there is always a probability of rain its just not always very much),

2. financeboomer says:

We should keep these comments in mind when we start feeling greedy or desperate.

3. Tigerstripe says:

If you have a Cray with stacked multicore processors, then you can talk about probabilities in this “market.” Otherwise, you are fucked.

4. endorendil says:

Sorry, but this is plain wrong. Probable is not a term in the field of probability either.

Probabilties are, as you say, about numbers. Numbers that may be taken out of thin air, be founded on bad assumptions or whatnot, but they are numbers. Something is Probable when its probability crossed some threshold, say 50%, or 30%, or 80%. Without specifying the threshold, Probable is not a statistical statement, but a subjective evaluation of a probability.
The exact same holds true for Possible. Is it possible that a stamp jumps up 1 millimeter by collecting heat from its surroundings and converting it to potential energy? Yes it is (it’s a classic calculation in statistical studies), but the probability is very low. Very, very low. But without specifying the threshold, it’s as subjective an evaluation as “Probable”.

For an example of how statisticians deal with translation of hard numbers to fuzzy human language, you can check out the IPCC report (box 1.1 at the bottom of http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch1s1-6.html).

5. CANDollar says:

Regarding rd’s comments and Tigerstripes… the equity market can be disconnected from economic fundamentals for extended periods but Bill Gross reminds us that returns over the long run cannot exceed the return of the economy.

So given the indicators rd mentions, savings rates probably must rise… but people don’t like to save so it is probably in our interests to develop some sort of system of index based forced savings national pension program like the Australians have with an opt out clause, instead of an opt in, to promote higher participation rates.

• rd says:

I think we will see future consumption of the over-65 crowd fall instead. Saving takes discipline and foresight. Most people will just think that they will be fine until they discover after the fact that they are not. Most people can’t envision how much money it requires to provide a steady income stream from savings for 20-30 years.

In order to have our savings level, we did not follow Bush’s advice to just go shopping. Instead, we don’t spend money on things like Cable TV (\$1,500 pre-tax into tax-deferred account), fancy cell phones with data plans (2 x \$1,000 pre-tax into tax-deferred account), cruises etc. However, I expect that we will be able to provide an ongoing income tax stream to the government and steady consumption fueling the economy over the long-run as the money comes out of our tax-deferred retirement accounts.

A universal forced savings program, like universal single payer health insurance, is almost unimaginable in the current American mind-set. Mythic philosophies are much more preferable to practical actuarial type of analysis. I think those political changes will probably be driven by Millenials in their 40s and 50s when they reflect on what they saw their parents go through a couple of decades from now.

• willid3 says:

thinking that if we all cut spending at once, then we almost all we loose our jobs, and others will loose their businesses. the problem is how to balance the spending to what you earn. and most of the problem there has been the collapse in incomes over the last 30 or so years. had that collapse not happened, we wouldn’t likely be where were are because we wouldnt have had to go into debt just stay even. as it is, any inflation at eats away more and more of the income we do make.

• bonzo says:

Returns over the long run most definitely can exceed the return of the economy. Suppose the economy doesn’t grow at all. 0% GDP growth and share of economy that goes to corporate profits is also stable. Stock market can still grow very fast because shares outstanding shrinks (takeovers, buybacks). Nothing magical about this. Conversely, economy can grow very fast and stock market can decline, because of dilution (IPOs, secondary offering, options). Stock market is not the economy.

• Note that overseas profits are how this can occur. Short to medium term, markets have almost zero correlation with the economy

6. Bruce Bartlett says:

A variation of your point is the way people conflate the words “evidence”
and “proof.” I often hear people use the latter when they mean the former.

7. [...] –Possible vs. Probable: Barry Ritholtz explains the difference between possible and probable. “Anything that is Probable must by definition be Possible; However, not everything that is Possible is going to be Probable. And therein lay the rub. Lately, I have seen this error come up in two areas — long term Pension Funds, and Venture Capital. The most common mistake is to conflate the two — mistake what is possible for something that is probable. With Pension Funds, we have an issue of national import in that many of these defined benefit plans are significantly underfunded. To catch up, some of these have been directed to add more Hedge Fund exposure. There is an accounting reason — but not a good investing reason — for this based on idea of expected returns. While the tiny minority of funds that create Alpha means that there exists a theoretical possibility for Hedgies to help pension funds make up their shortfalls, the high fees and under-performance makes it rather improbable that this will occur. This is to say nothing of the long odds that Pension funds will pick the next Ray Dalio or Jim Simons. Hence, the possibility of better performance perversely leads to the probability of worse performance.” [...]

8. Bob is still unemployed   says:

“A probable impossibility is preferable to an improbable possibility.”
- Aristotle

I first heard that phrase on The West Wing, and it has stuck with me. More insight into Aristotle’s comment: The impossiblity of Aristotle

9. [...] The Blogfather, Barry Ritholtz, has written a great piece this morning over at The Big Picture about how people often mistake what is possible for something that is probable, and uses the examples of long-term pension funds and venture capital. You can read the whole article here: Possible versus Probable (The Big Picture) [...]

10. Larry R says:

Great article. Not great for the entrepreneurs. Hope all is well

Larry

• True — that’s why I mentioned “futurists, designers, technologists” — I’ll add above.

I suspect entrepreneurs should not totally ignore the “probable” . . .

11. nofoulsontheplayground says:

I think Jim Carey illustrates your point quite well in the movie, “Dumb and Dumber”

“So you’re telling me there’s chance. YEAH!!”

12. tightstop says:

Great post/topic – very important and timely given all the macro concerns always in the news. This reminds me of a blog post I saw on one of my favorite blogs the other day. http://www.wave2wavetrading.com/daily-outlook-611/
Thanks for the post, and the blog in general, Barry – always enlightening!

1. Anything possible is also probable.
2. Probability is a weighted chance of outcome based on some historical info, perceived ( projected data), derived data based on some complex formulations or cognition based on intuition /experience
3. Possibility is thus a special case of probability that is perceived to have very low chance of outcome
4. Elite traders mostly employ counter trend strategies which are generally seen as possible and “less” probable
5. Average traders employ trend following strategies which are seen as highly probable but less possible
6. It therefore means that the both concepts ( probability or possibility) are applicable in the market place
7. Possibility in the market place requires hard work and great skill while probability requires little or no work at all as in trend following systems.
8. However it is a fact that market ranges 80% of the time and trends 20%. Possible strategies are applied in ranging market while probable ones are suitable for trending markets. Because it is not possible ( in reality) for everyone to see a trend when it starts it becomes easy to see why such strategies produce mediocre results over time.
9. Conclusion: A trader is required to work extremely hard. Far harder than being the President of USA.

14. 4everirish says:

Interesting post….wonderful thought processes….i just read another blog two days ago that dealt with the same topic… talk about probability verses possibility…

15. VennData says:

My favorite were those GOP generated “Obama tripled the debt” back a few years ago, showing the future projections, based on a trend. When you point that out then explain that the first trillion dollar deficit (different then debt) was Bush’s last budget, they would become apoplectic.

It is just bad for the country that people believe this GOP Media Machine garbage. At least they’re followers missed the rally. Imagine if an investment adviser had done that to them. Oh well, they believe in the GOP, LOL.

16. [...] Possible versus Probable - understanding the difference is important. [...]

17. bear_in_mind says:

I think there’s a multi-factorial feedback loop between how behavioral expectations form and come to realization (i.e. possibility) and resides in different cognitive tissues versus those activated in analytical processes (i.e. probability).

Obviously, you’re familiar with Dan Kahneman’s work, but thinking more fluidly about these functions, I suspect that a so-called “median American” has come to grossly under-utilize their analytical faculties and have increasingly come to rely more heavily upon their limbic circuitry. After all, who doesn’t love a good adrenaline rush at a movie, sporting event, or winning hand at the casino?

Our popular media constantly feeds and stimulates that region with their advertisements featuring loud volume, pounding, throbbing music, stunning colors, sex, and swash-buckling action, action, action! Can you imagine the Baby Boomer generation sitting glued to their seats playing Halo? Fuggettaboutit.

Anyway, it just struck a chord that I think it’s very, very hard for most Americans to escape the seductive, omnipresent sway of advertisments (cognitive science is clear on this) and the brain can only attend to so many variables and functions at once. Distract and clutter-up neural circuitry enough in the constant drive to keep-up with the Joneses and one wonders if circuitry for higher-level analytics may suffer atrophy. Just a thought.

One other comment regarding cognition: an increasing body of neuroscience is suggesting that “…focal damage to the ventromedial prefrontal cortex (vmPFC) would cause a “doubt deficit”…” and “…damage to the vmPFC disrupts a “false tagging mechanism” which normally produces doubt and skepticism…” (see authors and URL below)

The research n is small and obviously requires broader participation, but the findings are startling.

I can sadly report, albeit anecdotally, that I’ve witnessed these type of deficits in seniors who’ve wound-up swindled by ruses they’d have NEVER fallen for a decade prior. But they experienced a gradual, silent atrophy in executive functioning that has rendered their BS-detector useless.

It’s easy for the untrained eye to explain it away as “confusion”, “gullibility” or “dementia” but it’s not that. They can display ZERO deficits in mood, social functioning, short-term recall, etc., so even law enforcement is befuddled how to proceed. Like most of us, we’ve been conditioned in the mindset of caveat emptor, but that principle assumes a level cognitive playing field. These new findings begin to challenge that assumption in ways that have pretty powerful ramifications.

I’ll finish with one such ramification which doesn’t seem to be on anyone’s radar; namely, frauds such as offshore 409 lottery scams targeting seniors are siphoning an ungodly amount of money from American households. Estimates are in the tens of billions.

I wouldn’t doubt it, as it’s unusual to find losses of \$30K to \$200K from people who lived frugally all their lives. Totally wiped out, some even getting cash-out reverse mortgages to send payments to the scammers. And law enforcement agencies throw their hands-up over jursidiction, manpower, the victim’s apparent complicity, etc.

I think I need to GMOFB…

= = = = = = = = =

A neuropsychological test of belief and doubt: damage to ventromedial prefrontal cortex increases credulity for misleading advertising

Division of Behavioral Neurology and Cognitive Neuroscience
Department of Neurology
University of Iowa
Asp E, Manzel K, Koestner B, Cole CA, Denburg NL, Tranel D.

= = = = = = = = =

• rd says:

I have seen some of the same research regarding seniors and financial acumen. One of my fundamental financial goals is to have a very simple portfolio with only enough accounts with different providers to provide protection against their default (especially simple income annuities) by the time we are 65. For drawdown investments, the goal would be to have them with as few decisions required as possible (all-in-one Vanguard LifeStrategy Funds are a distinct possiblity) with very simple rules on a single page regarding withdrawals and budgeted expenditures. Key major reserve accounts, like for healthcare costs, may have their own separate account to avoid accidentally depleting those reserves.

That way, the temptation should be greatly reduced to “play” with the money with declining faculties. It would also be easier for my spouse or kids to manage the accounts. If we can save enough money (well on the way), then that reduces the whole concern about trying to swing for the fences to make up for lost time or yield that seems to be one of the key drivers driving the seniors into silly decisions.

18. Invictus says:

Would submit to you the following:

Possible implies that a potential outcome is >0.
Probable implies that a potential outcome is >50%.

Probable, to me, implies “more than likely,” or better than 50-50. Possible is simply >0.

• rd says:

Time frames are also important.

What is possible at this moment may be probable to occur at some point.

Hurricane Sandy was a classic example of this. It is possible for a storm to hit NYC every year, but the likelihood is low for any given year. However, it is probable that a storm like this will hit at some point during NYC’s design life for infrastructure. For decision-making this is a huge difference as most individuals and policy makers will assume that the inevitable will occur on somebody else’s watch but at some point it will happen on somebody’s watch.

An event like this also illustrates the definition of fat-tail events. The water levels around NYC are constantly oscillating through a very large range due to tides and storms. However, it is only when a certain elevation in a given location is exceeded that the damage starts and the damage escalates very quickly with only small incremental water level increases, so the water level rising from 5 to 7 feet MSL is normal but going from 10 feet to 12 feet goes from causing major flooding to shutting down much of the city for a week or more and destroying much infrastructure. So virtually all of the damage comes from events that would be classified as improbable for any given year.

The patterns are also complex with tides varying in magnitude daily across a wide range over a year and different areas have high tides at different times of the day because of the complex hydraulics in the NYC area. Sandy happened to hit at a fairly high tide in The Battery-Sandy Hook, NJ area and so that area got whacked. If she had hit a couple of hours earlier or later, the damage patterns would have looked very different and different lessons may have been learned.

19. Iamthe50percent says:

Re pension funds investing in hedge funds – Look for the kickback here. No sane fiduciary would throw pension money into a hedge fund or any other trading strategy. Those directors are milking the beneficiaries. I’m from Chicago and I can smell kickback a mile away.

20. HM says:

in a word:

anything is possible but not everything is probable.

helene