The Perils of Shorting: A Real Life Example

Gordon Gekko said:
I genuinely think that over a longer period you will do spectacularly worse than the market.
Click to expand...

Hi Gordon

If I trade shares regularly, I will do worse than the market because the costs will eat into my profits.

But if I buy and hold a portfolio of 10 shares, I will do differently from the market.

I might perform better and I might perform worse. But I can't see why I should expect to do worse? I would have thought it would be 50/50.

The only reason I can think of might be the argument that all of the returns come from 2% of the shares. So it's less likely that if I have 10 shares I will have one of them. Is that your argument?

Brendan
 
zephyro said:
for example what are your 2025 projections for total electric car sales, average sale price, Tesla market share %, net profit margin, etc. etc.
Click to expand...
I'm showing my age, but it reminds me of when I last had a full-time job, now over a decade ago. I had to certify whether companies I was advising would be able to stay in business for the next 30, 40, 50 years to meet their liabilities. My young colleagues did myriads of spreadsheet calculations, Monte Carlo simulations, the lot, projecting all sorts of possible futures for the company. It was my job to integrate their calculations with what I knew of the business from my interactions with the Board and senior managers. I invariably found that the insights gained from understanding the mind of the CEO, the continuity and quality of the top management team, the consistency of the strategy and its execution from quarter to quarter, year to year, were far more important than all the spreadsheets and financial projections in the world in determining the company's ability to survive in the long-term.
Looking at Tesla, we know that Elon Musk is a genius, but can he lead a complex car manufacturing company? What about his state of mind, given some of the things he posts on Twitter and his 'lifestyle'? What about his ability to retain good people? What about consistency of approach to strategy? In a July 'news' update (here) I quoted from the Q2 2019 Update letter to investors, where he wrote, in relation to FSD (full self-driving): "We are making progress towards stopping at stop signs and traffic lights”. What a statement to include in a quarterly update! Was there anything in the latest quarterly update on whether they've managed that feat? Probably not, given the R&D cutbacks.
 
Last edited:
Sarenco said:
According to Vanguard, the probability of a 10-stock portfolio outperforming the broader market is 34.5%.

You would have to hold over 500 stocks to get to a 50/50 coin toss.
https://www.vanguardinvestments.se/documents/institutional/increase-odds-of-owning-less-stock-that-drive-returns.pdf

To be fair, the Vanguard simulations ignore transaction costs and management fees.
Click to expand...
Sarenco I have before rubbished this Vanguard analysis but let me try again.
All models of the stockmarket are that prices are log symmetrical.
So let us take a very simple model of the market. It consists of only 2 stocks. Each have the following distribution.
Chances are 50/50 of a 50% fall or a 200% rise.
A single stock portfolio will have a 50% chance of losing 50% and 50% chance of doubling. That is an expectation of 125%: (50% +100%)/2
A full double stock portfolio will have the following possibilities:
25% that there is 50% on both i.e. an overall return of 50%
50% that there is 50% on one and 200% on the other, a return of 125%
25% that there is a 200% return on both, an overall return of 200%
The expected return is 125% the exact same as a the single stock portfolio, but there is a 75% chance of a positive return versus a 50% on the single stock portfolio.
So yes, the odds are that the single stock portfolio will underperform the full portfolio but because the odds of the single stock portfolio shooting the lights out are higher it balances out.
But I agree that the single stock portfolio is riskier.
 
Colm Fagan said:
Do you know something about my state of senility that my nearest and dearest are afraid to tell me? I presume you know the theory that, if I just sit on my hands and do nothing (which is my default mode) then the expectation is that I will do exactly the same as the market. I would 'genuinely' love to know why you think I will do 'spectacularly worse than the market'.
Click to expand...

Why do you believe that your concentrated portfolio of single stocks would do the same as the market?

Presumably you do what you do because you think you can generate alpha? Or is it to minimise fees?

My fear is based on a view that too many of your researched stocks will turn out to be duds and too few will turn out to be star performers.

This is partly based on my own view that we are living through a period of enormous disruption and technological change where the “winners” are even harder to identify. As a result, I believe that the DIY investor is even more of a lamb to the slaughter than ever.

If I were you, and I’m not, I’d take 80% of my money and put it into an MSCI World ETF or cheap global equity fund, I’d self-manage the other 20% of it as you’re currently doing, and I’d have fun writing about and comparing the performance of the two pots.
 
Brendan Burgess said:
Hi Gordon

If I trade shares regularly, I will do worse than the market because the costs will eat into my profits.

But if I buy and hold a portfolio of 10 shares, I will do differently from the market.

I might perform better and I might perform worse. But I can't see why I should expect to do worse? I would have thought it would be 50/50.

The only reason I can think of might be the argument that all of the returns come from 2% of the shares. So it's less likely that if I have 10 shares I will have one of them. Is that your argument?

Brendan
Click to expand...

My argument is that none of us have the resources to pick the right 10. It’s not just blind luck (i.e. pick any 10). We’re picking them, and we just don’t have the skill to do so. And yes, much of the return comes from a smaller number of star performers. I don’t have to nous to identify the next Amazon; so I outsource the management of my money.
 
Hi Gordon

I agree that none of us can pick systematically pick winners. But by that criterion, surely Colm can't systematically pick losers either? (If you think he can, then you should short his buys and buy his shorts)

So the expected return should be the same as the market. He is as likely to outperform as to underperform.

Brendan
 
Brendan Burgess said:
Hi Gordon

I agree that none of us can pick systematically pick winners. But by that criterion, surely Colm can't systematically pick losers either? (If you think he can, then you should short his buys and buy his shorts)

So the expected return should be the same as the market. He is as likely to outperform as to underperform.

Brendan
Click to expand...

But aren’t you then assuming that nobody can pick winners and that it’s all a game of chance?

How much of the return from global markets is coming from the US, from tech, and from certain specific companies?

Apple is up by nearly 80% this year. It’s in my global equity fund and it’s in my ETF; it might not be in my concentrated stock portfolio.

In this era of technological change and disruption, I fear that someone like me trying to pick stocks will be left owning this generation’s Kodak, maybe a couple of okay performers, and seemingly good companies that get blown away by technological advancement that I don’t understand.
 
Last edited:
I'm really not equipped to debate actuarial models, assumptions, etc.

So I will rely on common sense.

It would be obvious to a blind man that the bulk of historic stock market returns are attributable to a very small minority of stocks (Exxon, Apple, etc).

So the likelihood of picking 10 winning stocks at random from a menu of more than 10,000 publicly traded stocks globally seems pretty small. On the other hand, the likelihood of picking 10 losers seems pretty high.

Once you accept that a small minority of stocks are "super performers" then it seems obvious to me that your chances of picking "winners" increases as you add stocks to your portfolio. Conversely, the more concentrated the portfolio, the greater the odds of missing out on the winners.

In other words, the distribution of returns is not symmetrical - there are far more "losers" than "winners". It's not a 50/50 bet.

Of course, market cap matters. Pick the 10 biggest stocks in the U.S. market by market cap and you can be pretty sure that you won't be a million miles off the S&P500 after 10 years.
 
Sarenco said:
So the likelihood of picking 10 winning stocks at random from a menu of more than 10,000 publicly traded stocks globally seems pretty small. On the other hand, the likelihood of picking 10 losers seems pretty high.
Click to expand...

but as the majority of stocks have positive beta and generally rise in price over time, surely the probablity even by chance of picking 10 winners is higher than picking 10 losers
 
Its like prize bonds the market total % return is the same , you could buy 10 prize bonds or 10,000 prize bonds (well neither of them is not enough to get winners regularly ) but because majority of the market % return is made up of bigger prizes the chances of a higher return in prize bonds increase as you buy more bonds .
 
Sarenco said:
I'm really not equipped to debate actuarial models, assumptions, etc.

So I will rely on common sense.

It would be obvious to a blind man that the bulk of historic stock market returns are attributable to a very small minority of stocks (Exxon, Apple, etc).

So the likelihood of picking 10 winning stocks at random from a menu of more than 10,000 publicly traded stocks globally seems pretty small. On the other hand, the likelihood of picking 10 losers seems pretty high.

Once you accept that a small minority of stocks are "super performers" then it seems obvious to me that your chances of picking "winners" increases as you add stocks to your portfolio. Conversely, the more concentrated the portfolio, the greater the odds of missing out on the winners.

In other words, the distribution of returns is not symmetrical - there are far more "losers" than "winners". It's not a 50/50 bet.

Of course, market cap matters. Pick the 10 biggest stocks in the U.S. market by market cap and you can be pretty sure that you won't be a million miles off the S&P500 after 10 years.
Click to expand...
If you pick a stock at random you have the exact same expectation as the whole market. Of course if you pick the whole market your returns will vary between -10% and + 20% whilst if you pick 1 stock it will be between -30% and + 40%.
But you are right, you are more likely to lose with a small portfolio but this is compensated by having a higher chance of shooting the lights out.
 
Last edited:
Sarenco said:
Could you rephrase that in English that we can all understand?

In other words, what do you mean by "majority", "beta" and "generally"?
Click to expand...
It certainly needs some defining since the average beta is 1 by definition, and a negative beta would be truly bizarre
 
Duke of Marmalade said:
If you pick a stock at random you have the exact same expectation as the whole market. Of course if you pick the whole market your returns will vary between -10% and + 20% whilst if you pick 1 stock it will be between -30% and + 40%.
Click to expand...

That makes no sense at all.

Call me old fashioned, but I reckon if you pick the whole market, your portfolio’s performance is likely to mirror that of the market.
 
Last edited:
Gordon Gekko said:
That makes no sense at all.
Click to expand...

What the Duke is saying is...

If the market rises between -10% and +20%, an individual stock might rise between -30% and +40%.

But the expected outcome is the same.

Consider a coin toss.

If I toss 100 fair coins and bet €1 on heads every time - my expected return will be zero. My actual return will probably be somewhere between -€10 & +€10 or between - 10 cents per toss and + 10 cents per toss.

If I toss only one coin, I will either win €1 or lose €1. But my expected return will still be zero.

Which comes back to your statement that you expect Colm to lose a lot of money.

His expected return will be the same as the market. But it could be significantly lower or significantly higher.

Brendan
 
But picking a single stock at random is not a simple coin toss. That's really the point.

The probability that you will pick a loser is far higher than the probability that you will pick a winner.

Again, the distribution of stock returns is not symmetrical - there are far more losers than winners. It's not a 50/50 bet.
 
Sarenco said:
the distribution of stock returns is not symmetrical - there are far more losers than winners.
Click to expand...

Hi Sarenco

Is this definitely true?

I think it might be true for all shares from the initial flotation.

But is it true for long established companies with long records of profits?

It might be, but it surprises me.

My gut feeling, and it's only a gut feeling, is that, over the long term, out of 10 shares, 2 bomb, 2 give fantastic returns, and 6 give moderate returns.

Brendan
 
Has this not gone a bit off topic into financial theory. Colm doesn't randomly select his shares using his trusted Ouija board (or maybe he is!) so is this not all a bit moot talking about returns of randomly selected stocks. People mightn't agree with his approach but there is no financial or statistical rule that says he is going to be wrong either...
 
Sarenco said:
But picking a single stock at random is not a simple coin toss. That's really the point.

The probability that you will pick a loser is far higher than the probability that you will pick a winner.

Again, the distribution of stock returns is not symmetrical - there are far more losers than winners. It's not a 50/50 bet.
Click to expand...
Ok so we have a market of 99 stocks which will lose everything and a single stock which will jump to 200. My expectation when holding the whole market is to double my money. a priori I have the exact same expectation if I just pick a stock at random.
But I concede that you have a point on the likelihood of losing. And your point is valid even if all stocks have the same log symmetrical* prospects..
A bit off topic for sure but GG did make an erroneous assertion about the prospects for Colm’s “strategy”.

* the chances of the stockmarket going to zero are obviously much less than of it going to +100%. So it would not be reasonable to posit that performance prospects are symmetrical. But log(zero)is a very big negative number and so it is quite a bit more reasonable to posit that performance prospects are log symmetrical
 
Last edited:
You must log in or register to reply here.
Back
Top