Author Topic: Successful investor: How to get more in retirement  (Read 8994 times)

katthjul

  • 5 O'Clock Shadow
  • *
  • Posts: 62
  • Location: Sweden
Successful investor: How to get more in retirement
« on: November 25, 2017, 11:23:27 AM »
Some questionable advice for investing your retirement savings that was published in a large Swedish newspaper this summer. My translation is only a summary of the article, but I tried to keep the quotes close to the source material.

Successful investor: How to get more in retirement

The successful investor explains how he increased his retirement savings from 20k (2000) to 1.5 millions (2017). He believes in being an active investor and does not understand why most people stay in the default pension fund (a target date fund with a expense ration of 0.07% - 0.12%) which never shows up among the top performers.

1. Find out what is hot right now.
Quote
Every other day, I spend 10-15 minutes reading what is happening in the market and what people is talking about. Is it about petrol or gold? Or metals or tech? How hard can it be [to keep up with the trends]?
2. Only invest in the winners.
Quote
I look at the funds available that is doing well. Then you simply pick one of those. If any of them has a low expense ratio, I pick that one, of course. I try to stay in funds with an expens ration below 0.7%. Should a fund drop down the list of top performers replace it with another among the top 5.

Source: Investor gets more in retirement than from working

clarkfan1979

  • Magnum Stache
  • ******
  • Posts: 3352
  • Age: 44
  • Location: Pueblo West, CO
Re: Successful investor: How to get more in retirement
« Reply #1 on: November 25, 2017, 11:29:06 AM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.

obstinate

  • Handlebar Stache
  • *****
  • Posts: 1147
Re: Successful investor: How to get more in retirement
« Reply #2 on: November 25, 2017, 01:33:16 PM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.
No, that's not really how it works either. If it's difficult to pick winners, it must be difficult to pick losers as well. By negation:
  • Suppose there is no reliable method for building a portfolio that outperforms the market.
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
  • Build a portfolio that will under-perform the market.
  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.
  • Therefore, there is a reliable method for constructing a portfolio that will outperform the market.
  • Contradiction. QED.


Indexer

  • Handlebar Stache
  • *****
  • Posts: 1463
Re: Successful investor: How to get more in retirement
« Reply #3 on: November 26, 2017, 08:30:34 AM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.

Keep in mind this is an arena where the best of the best are competing against each other. Being really good makes you average. People who did really well normally benefited from luck. They invested in the thing that was popular. When it becomes unpopular or the markets shift in other ways their strategy won't work as well. Their performance will likely regress back to the average.

People who did really terrible might be doing so terrible because of bad luck. Their strategy was out of favor and it could be in favor in the future. Sure, those guys might outperform for awhile, regressing back up to the average. Finding them is incredibly difficult, because many people who did terrible did so because they are in an arena where they can't compete. They will thus continue to do terrible. Separating the two is really difficult, just about as impossible as picking next year's best performers.

Let's use football(or any sport) as an analogy. If two really good teams go up against each other they should play pretty similar, but luck can turn the tide. A lot of luck could even help a below average team beat an above average team. However, the worst team in the league has a terrible record, not due to luck, but just because they can't compete. I wouldn't put money on the Cleveland Browns to win their next game, and I don't even know who their next game is against... a record of 10 losses and 0 wins speaks for itself.
« Last Edit: November 26, 2017, 08:39:27 AM by Indexer »

renata ricotta

  • Pencil Stache
  • ****
  • Posts: 703
Re: Successful investor: How to get more in retirement
« Reply #4 on: November 26, 2017, 10:03:19 AM »

2. Only invest in the winners.
Quote
I look at the funds available that is doing well. Then you simply pick one of those. If any of them has a low expense ratio, I pick that one, of course. I try to stay in funds with an expens ration below 0.7%. Should a fund drop down the list of top performers replace it with another among the top 5.

Source: Investor gets more in retirement than from working

Huh? How would anyone ever make money with this approach? By definition, he is buying when prices are high and selling when they are low, in exchange for more expensive stocks.

obstinate

  • Handlebar Stache
  • *****
  • Posts: 1147
Re: Successful investor: How to get more in retirement
« Reply #5 on: November 26, 2017, 12:12:22 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Bear in mind that I am talking about the average performance of the chosen security class, not average performance of mutual funds vs. the S&P 500. Mutual funds often underperform the stock market because there are additional costs that have to be considered -- paying management, tax inefficiencies, etc. -- which I do not mean to capture in the proof I added. So, if you're trying to prove me wrong about being able to select under-performing mutual funds, I'd accept back-testing counterproof in the form of an algorithm that selects mutual funds that underperform the AUM-weighted average of mutual funds, before fees, by a significant fraction. If you're talking about stocks, the same, but with respect to the S&P 500 average.
« Last Edit: November 26, 2017, 12:13:56 PM by obstinate »

Telecaster

  • Magnum Stache
  • ******
  • Posts: 3551
  • Location: Seattle, WA
Re: Successful investor: How to get more in retirement
« Reply #6 on: November 26, 2017, 01:46:24 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Bear in mind that I am talking about the average performance of the chosen security class, not average performance of mutual funds vs. the S&P 500. Mutual funds often underperform the stock market because there are additional costs that have to be considered -- paying management, tax inefficiencies, etc. -- which I do not mean to capture in the proof I added. So, if you're trying to prove me wrong about being able to select under-performing mutual funds, I'd accept back-testing counterproof in the form of an algorithm that selects mutual funds that underperform the AUM-weighted average of mutual funds, before fees, by a significant fraction. If you're talking about stocks, the same, but with respect to the S&P 500 average.

Fair's fair.  You don't provide a backtest of your theory,  you just claimed it works and walked away. 

But there is a point that didn't follow, namely:

  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.

That's not logically true. The inverted portfolio could also underperform because there are more than two possible winning portfolios, right?  If you asked 20 people to construct a winning portfolio, you'd have 20 different portfolios.  And one or more of those might actually beat the market that year!  So if you did the inverse, you shot yourself in the foot.  Or your winner portfolio only modestly loses to the market, but the inverse the does better but still loses.  You are assuming a binary outcome when in fact there are many options. 





Telecaster

  • Magnum Stache
  • ******
  • Posts: 3551
  • Location: Seattle, WA
Re: Successful investor: How to get more in retirement
« Reply #7 on: November 26, 2017, 01:48:16 PM »

2. Only invest in the winners.
Quote
I look at the funds available that is doing well. Then you simply pick one of those. If any of them has a low expense ratio, I pick that one, of course. I try to stay in funds with an expens ration below 0.7%. Should a fund drop down the list of top performers replace it with another among the top 5.

Source: Investor gets more in retirement than from working

Huh? How would anyone ever make money with this approach? By definition, he is buying when prices are high and selling when they are low, in exchange for more expensive stocks.

He's basically following a momentum strategy--which can work well as long as the market is strong.  We've had an usually long bull market so it is possible he's done well.  The danger with momentum strategies is if the market goes down or even sideways you can blow your fingers off. 

Indexer

  • Handlebar Stache
  • *****
  • Posts: 1463
Re: Successful investor: How to get more in retirement
« Reply #8 on: November 26, 2017, 10:52:58 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Bear in mind that I am talking about the average performance of the chosen security class, not average performance of mutual funds vs. the S&P 500. Mutual funds often underperform the stock market because there are additional costs that have to be considered -- paying management, tax inefficiencies, etc. -- which I do not mean to capture in the proof I added. So, if you're trying to prove me wrong about being able to select under-performing mutual funds, I'd accept back-testing counterproof in the form of an algorithm that selects mutual funds that underperform the AUM-weighted average of mutual funds, before fees, by a significant fraction. If you're talking about stocks, the same, but with respect to the S&P 500 average.

In the real world mutual funds that outperform in the short term tend to regress to the mean(perform worse) over time. In other words, "Based on regression to the mean, if you only pick past winners you are most likely picking future losers." You claimed, "No, that's not really how it works either."  Clarkfan1979's statement is correct, which makes your statement incorrect. I'm not really concerned with how your proof came to the incorrect conclusion.

1. We are talking about mutual funds. The person you originally quoted, clarkfan1979, the OP, and the quoted article were all talking about mutual funds within a pension plan.
2. I don't need to do back testing. It's already been done. The list of sources that show outperforming active funds regress to the mean over time is extensive; to the point I just assume people know it as common knowledge. My favorite sources: Little Book of Common Sense Investing(John Bogle), and A Random Walk Down Wall Street. In addition, the Wall Street Journal recently did a piece about Morningstar's Star system being a mirage.

https://www.wsj.com/articles/the-morningstar-mirage-1508946687

Summary: Morningstar ranks funds based on past performance. 5 stars= best. 1 star= worst. 10 years later the average 5 star has dropped to a 3 star and the average 1 star rises to a 1.9 star. In other words, funds that outperform(5 star) tend to regress to the mean and become average(3 star) over time. The worst performing funds(1 star) tend to get closer to the mean, but they are still performing below average(1.9 stars).

i_have_so_much_to_learn

  • Stubble
  • **
  • Posts: 195
  • Location: SF Bay, CA, USA
  • started a chatroom: https://discord.gg/agSSAhXzYD
Re: Successful investor: How to get more in retirement
« Reply #9 on: November 26, 2017, 11:26:45 PM »
Some questionable advice for investing your retirement savings that was published in a large Swedish newspaper this summer. My translation is only a summary of the article, but I tried to keep the quotes close to the source material.

Successful investor: How to get more in retirement

The successful investor explains how he increased his retirement savings from 20k (2000) to 1.5 millions (2017). He believes in being an active investor and does not understand why most people stay in the default pension fund (a target date fund with a expense ration of 0.07% - 0.12%) which never shows up among the top performers.

1. Find out what is hot right now.
Quote
Every other day, I spend 10-15 minutes reading what is happening in the market and what people is talking about. Is it about petrol or gold? Or metals or tech? How hard can it be [to keep up with the trends]?
2. Only invest in the winners.
Quote
I look at the funds available that is doing well. Then you simply pick one of those. If any of them has a low expense ratio, I pick that one, of course. I try to stay in funds with an expens ration below 0.7%. Should a fund drop down the list of top performers replace it with another among the top 5.

Source: Investor gets more in retirement than from working

This is literally the opposite of Warren Buffet's famous quote, which says that the short term is just a popularity contest (that eventually expires).

Laura33

  • Magnum Stache
  • ******
  • Posts: 3479
  • Location: Mid-Atlantic
Re: Successful investor: How to get more in retirement
« Reply #10 on: November 27, 2017, 10:15:17 AM »

2. Only invest in the winners.
Quote
I look at the funds available that is doing well. Then you simply pick one of those. If any of them has a low expense ratio, I pick that one, of course. I try to stay in funds with an expens ration below 0.7%. Should a fund drop down the list of top performers replace it with another among the top 5.

Source: Investor gets more in retirement than from working

Huh? How would anyone ever make money with this approach? By definition, he is buying when prices are high and selling when they are low, in exchange for more expensive stocks.

He's basically following a momentum strategy--which can work well as long as the market is strong.  We've had an usually long bull market so it is possible he's done well.  The danger with momentum strategies is if the market goes down or even sideways you can blow your fingers off.

This.

I think the ongoing dispute here about whether he's picking future losers is only tangentially addressing the critical point:  he's not just choosing average stock funds, he's choosing specific market sectors.  I am assuming that his retirement plan will have one or two funds within a number of specific sectors, so maybe an aggressive growth fund, a dividend-focused fund, a value fund, a bond fund, something that invests in commodities, something that invests in REITs, etc. etc. etc.  So assume he starts with an equal weighting, and then the next year value funds suck and growth funds are taking off.  So he sells the value funds (low) and buys more of the growth funds (high).  Then the next year, he picks the biggest losing sector, sells it (low), and buys the biggest winner (high).  Rinse, repeat.  Classic momentum investing.

The think about momentum investing is that it can usually work for several years -- IIRC from a few years ago, when you are at the beginning of a market rally, buying the top-performing sector (aggressive growth funds) will make you significant money for 2-3 years before it falls out of favor.  Couple that with, in this guy's case, one of the longest-running bull markets in history, and I'd be surprised if he hadn't made a buttload of money.

The problem, of course, is that eventually even the most raging bull market ends.  It may end creakingly slowly, or it may end in the blink of an eye.*  And unless you are a massively sophisticated investor (and, honestly, even if you are), you will have no clue when that is going to happen.  So how do you know when you should toss those aggressive growth funds and put your money in other asset classes/types of funds?  Based on history, the answer given by people like this guy is "after stocks have cratered and my portfolio is worth half its former value."** 

So I agree completely that this guy is setting himself up for future underperformance.  To my mind, though, this is consistent with the law of averages/regression to the mean, not because of it.***  We know from the past 100+ years of experience that markets are cyclical, because the economies that provide the basis for those markets are cyclical.  The booming economies that lead to bull markets are inevitably replaced with recessions that lead to bear markets, until the economy starts to perk up again and the market swings.  As a result, we can predict, with a high degree of certainty, that a strategy that maximizes your returns during bull markets will very likely fuck you over completely during bear markets.

*Hint:  the greater the rise, usually the more sudden and dramatic the fall.

**Added bonus here that this guy doesn't even realize that he is following a specific and well-known investment strategy -- he just thinks he's got this whole "investing" thing nailed.  And since he doesn't even know that he is following a strategy, he has no clue about that strategy's limitations (i.e., that it works only in a bull market).  So unlike those sophisticated momentum players, this guy won't even be looking for signs of a pending bear market to signal that he needs to ditch that approach for a different one; he'll just continue blithely on, content in his belief that he's smarter than the average bear. 

***IMO, the law of averages/regression to the mean would apply if you were focusing on performance within a specific subcategory of the market, where performance is determined based on stock-picking prowess instead of which sector of the market is hotter or colder.  The guy who "wins" that competition for the first couple of years is not likely to be the winner in later years, because very very few people can pick winners and accurately predict the future consistently over a period of several years. 

bacchi

  • Walrus Stache
  • *******
  • Posts: 7056
Re: Successful investor: How to get more in retirement
« Reply #11 on: November 27, 2017, 12:23:30 PM »
Some questionable advice for investing your retirement savings that was published in a large Swedish newspaper this summer. My translation is only a summary of the article, but I tried to keep the quotes close to the source material.

Successful investor: How to get more in retirement

1. Start a blog about your successful investing or ER skillz.
2. Put advertising on said blog.
3. Pray to the internet gods that your clever wit and outstanding writing draws enough clicks.
4. Profit.
 

Psychstache

  • Handlebar Stache
  • *****
  • Posts: 1594
Re: Successful investor: How to get more in retirement
« Reply #12 on: November 27, 2017, 12:36:25 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Username checks out

trollwithamustache

  • Handlebar Stache
  • *****
  • Posts: 1146
Re: Successful investor: How to get more in retirement
« Reply #13 on: November 27, 2017, 02:26:10 PM »
 The Momentum investors who get of the bus at the right time do pretty well for themselves...


scottish

  • Magnum Stache
  • ******
  • Posts: 2716
  • Location: Ottawa
Re: Successful investor: How to get more in retirement
« Reply #14 on: November 27, 2017, 05:26:06 PM »
Yup.   Works great until it stops working.    <shrug>

i_have_so_much_to_learn

  • Stubble
  • **
  • Posts: 195
  • Location: SF Bay, CA, USA
  • started a chatroom: https://discord.gg/agSSAhXzYD
Re: Successful investor: How to get more in retirement
« Reply #15 on: November 27, 2017, 05:29:34 PM »
Yup.   Works great until it stops working.    <shrug>

True for everything.

MrMoogle

  • Handlebar Stache
  • *****
  • Posts: 1136
  • Age: 38
  • Location: Huntsville, AL
Re: Successful investor: How to get more in retirement
« Reply #16 on: November 28, 2017, 02:06:43 PM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.
No, that's not really how it works either. If it's difficult to pick winners, it must be difficult to pick losers as well. By negation:
  • Suppose there is no reliable method for building a portfolio that outperforms the market.
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
  • Build a portfolio that will under-perform the market.
  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.
  • Therefore, there is a reliable method for constructing a portfolio that will outperform the market.
  • Contradiction. QED.
You're second assumption is false:
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
Therefore the rest that follows is not valid.

obstinate

  • Handlebar Stache
  • *****
  • Posts: 1147
Re: Successful investor: How to get more in retirement
« Reply #17 on: November 28, 2017, 11:30:54 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Bear in mind that I am talking about the average performance of the chosen security class, not average performance of mutual funds vs. the S&P 500. Mutual funds often underperform the stock market because there are additional costs that have to be considered -- paying management, tax inefficiencies, etc. -- which I do not mean to capture in the proof I added. So, if you're trying to prove me wrong about being able to select under-performing mutual funds, I'd accept back-testing counterproof in the form of an algorithm that selects mutual funds that underperform the AUM-weighted average of mutual funds, before fees, by a significant fraction. If you're talking about stocks, the same, but with respect to the S&P 500 average.

In the real world mutual funds that outperform in the short term tend to regress to the mean(perform worse) over time. In other words, "Based on regression to the mean, if you only pick past winners you are most likely picking future losers." You claimed, "No, that's not really how it works either."  Clarkfan1979's statement is correct, which makes your statement incorrect. I'm not really concerned with how your proof came to the incorrect conclusion.

1. We are talking about mutual funds. The person you originally quoted, clarkfan1979, the OP, and the quoted article were all talking about mutual funds within a pension plan.
2. I don't need to do back testing. It's already been done. The list of sources that show outperforming active funds regress to the mean over time is extensive; to the point I just assume people know it as common knowledge. My favorite sources: Little Book of Common Sense Investing(John Bogle), and A Random Walk Down Wall Street. In addition, the Wall Street Journal recently did a piece about Morningstar's Star system being a mirage.

https://www.wsj.com/articles/the-morningstar-mirage-1508946687

Summary: Morningstar ranks funds based on past performance. 5 stars= best. 1 star= worst. 10 years later the average 5 star has dropped to a 3 star and the average 1 star rises to a 1.9 star. In other words, funds that outperform(5 star) tend to regress to the mean and become average(3 star) over time. The worst performing funds(1 star) tend to get closer to the mean, but they are still performing below average(1.9 stars).
"Underperforming" has a specific meaning here, which I don't think j you've captured. Underperforming doesn't merely mean doing worse than you did before. It means doing worse than average. I don't dispute that the top funds in a given year will have average performance in the next. I dispute the idea that they have below average performance.
« Last Edit: November 28, 2017, 11:34:03 PM by obstinate »

obstinate

  • Handlebar Stache
  • *****
  • Posts: 1147
Re: Successful investor: How to get more in retirement
« Reply #18 on: November 28, 2017, 11:35:05 PM »
@obstinate: Yes, that is how it works. It isn't a logical contradiction.
I provided a proof, and I don't believe you pointed out any errors in that proof. I'll only accept dispute in two forms:
  • Strong: select a specific point in my proof, and demonstrate that it does not follow from the previous point.
  • Less strong: back-test your theory against real stock market data, and demonstrate that it is the case.
(I mean, you're free to post whatever you want to dispute what I say, of course, but I'm not going to respond to it on the merits unless it's in one of the above categories.)

Bear in mind that I am talking about the average performance of the chosen security class, not average performance of mutual funds vs. the S&P 500. Mutual funds often underperform the stock market because there are additional costs that have to be considered -- paying management, tax inefficiencies, etc. -- which I do not mean to capture in the proof I added. So, if you're trying to prove me wrong about being able to select under-performing mutual funds, I'd accept back-testing counterproof in the form of an algorithm that selects mutual funds that underperform the AUM-weighted average of mutual funds, before fees, by a significant fraction. If you're talking about stocks, the same, but with respect to the S&P 500 average.

Fair's fair.  You don't provide a backtest of your theory,  you just claimed it works and walked away. 

But there is a point that didn't follow, namely:

  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.

That's not logically true. The inverted portfolio could also underperform because there are more than two possible winning portfolios, right?  If you asked 20 people to construct a winning portfolio, you'd have 20 different portfolios.  And one or more of those might actually beat the market that year!  So if you did the inverse, you shot yourself in the foot.  Or your winner portfolio only modestly loses to the market, but the inverse the does better but still loses.  You are assuming a binary outcome when in fact there are many options.
In the case of a portfolio and it's complement, there are only two possibilities. One must be above the average, and one must be below (or else they can be exactly equal).

obstinate

  • Handlebar Stache
  • *****
  • Posts: 1147
Re: Successful investor: How to get more in retirement
« Reply #19 on: November 28, 2017, 11:36:44 PM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.
No, that's not really how it works either. If it's difficult to pick winners, it must be difficult to pick losers as well. By negation:
  • Suppose there is no reliable method for building a portfolio that outperforms the market.
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
  • Build a portfolio that will under-perform the market.
  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.
  • Therefore, there is a reliable method for constructing a portfolio that will outperform the market.
  • Contradiction. QED.
You're second assumption is false:
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
Therefore the rest that follows is not valid.
That's kind of the point of proof by contradiction.  :)

MrMoogle

  • Handlebar Stache
  • *****
  • Posts: 1136
  • Age: 38
  • Location: Huntsville, AL
Re: Successful investor: How to get more in retirement
« Reply #20 on: November 29, 2017, 10:35:40 AM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.
No, that's not really how it works either. If it's difficult to pick winners, it must be difficult to pick losers as well. By negation:
  • Suppose there is no reliable method for building a portfolio that outperforms the market.
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
  • Build a portfolio that will under-perform the market.
  • Invert this portfolio (i.e. buy a weighted set of everything else).
  • Since the loser is by definition likely to under-perform the average, and your inverted portfolio must outperform the average.
  • Therefore, there is a reliable method for constructing a portfolio that will outperform the market.
  • Contradiction. QED.
You're second assumption is false:
  • Suppose there is a reliable method to build a portfolio that under-performs the market (e.g. picking things that have done well in the past).
Therefore the rest that follows is not valid.
That's kind of the point of proof by contradiction.  :)
You're saying that picking the previous winners is just as good as picking anyone else, statistically, and therefore you are not "most likely picking future losers."  I'll agree to that.  I missed your point the first time around.

You're (generic you) on average going to perform as the mean.  The real problem is diversification.  Since you're only choosing one sector, you have much more risk than if you were spread out across the whole market.  Right now, almost all markets are rising, so the author will do fine in any situation.  The problem is if the sector you just moved to right before a crash does worse than average, you end up selling low and buying high.  For example, the financial sector was doing really good, until financial institutions started closing in 2008/09, although it probably wasn't the top sector right beforehand.

You've basically said the problem with this method.  Since all companies are pretty much equal in the long run, you reduce risk choosing all instead of just the top 5.  You still want the top 5, but you also want the bottom 5, and all the 5's in between.

TheGrimSqueaker

  • Magnum Stache
  • ******
  • Posts: 2606
  • Location: A desert wasteland, where none but the weird survive
Re: Successful investor: How to get more in retirement
« Reply #21 on: November 29, 2017, 12:01:03 PM »
We're at risk of falling into a false dichotomy and deciding that each stock is going to be a "winner" or a "loser", when in reality most of them are going to be less than one sigma away from the mean and be simply "average". The challenge when trying to pick a portfolio of "winner" stocks that will consistently outperform the mean is that you're more likely to pick an average performer and you'll end up with a few real stinkers. However it's important to notice that, over the long term, the absolute worst performers get culled from the herd.

It's definitely possible to build a portfolio of companies that consistently underperform. Look for the ones that are in danger of failing.

Group companies with similar industries and of a similar size together. Have a group of airlines, a group of drugstores, a group of junk food manufacturers, and so on. Then, using each company's annual report or some Motley Fool type synopsis, go through each group comparing the companies to one another. This way you're comparing apples with apples. Look for signs of stress such as:

  • High debt to equity ratio (probably over-leveraged)
  • High debt to earnings ratio (possibly a candidate for bankruptcy)
  • Increasing debt over the past three years
  • Low upside-downside
  • Not one of the top three in terms of market share
  • Excessive executive compensation
  • Excessive insider trading
  • Bursts of high volume trading without noticeable change in stock price ("wash" trading)
  • Company headquarters is in Utah
  • A history of unusually high dividends

Pretend you're a corporate raider considering a company to acquire in a leveraged buyout, and select the company that's the least likely candidate. That's the dog of a company you want on your loser list.

Every once in a while someone will have a massively excellent idea and patent it, or else bring in a turnaround CEO who knows how to restructure the company to return it to profitability, but for the most part failure attracts more failure. The best ideas are more likely to be nurtured and made profitable in a company that basically functions and is healthy. For every company that pulls itself out of the quicksand and actually does something impressive, there are several others for which trading will just plain be halted or they'll go out of business. Not all companies survive bankruptcy proceedings, and the ones that do generally don't have vulture-friendly stocks.

Investment-wise a person doesn't really want "the bottom five" as much as "the top five" and "five from the middle". For an average-performing company to become outstanding takes less luck and effort than for a poorly-performing company to become average, because a company with serious problems has generally already been gutted of whatever wealth or assets could potentially be passed on to shareholders. Debt service alone is a serious hit to cash flow, and that's just as true of companies as it is of individuals.

Telecaster

  • Magnum Stache
  • ******
  • Posts: 3551
  • Location: Seattle, WA
Re: Successful investor: How to get more in retirement
« Reply #22 on: November 29, 2017, 12:59:06 PM »

In the case of a portfolio and it's complement, there are only two possibilities. One must be above the average, and one must be below (or else they can be exactly equal).

They must be the average of each other.  We're talking about the market average.  It is definitely possible to have two portfolios, both of which under perform the market. 


MrMoogle

  • Handlebar Stache
  • *****
  • Posts: 1136
  • Age: 38
  • Location: Huntsville, AL
Re: Successful investor: How to get more in retirement
« Reply #23 on: November 29, 2017, 01:12:33 PM »

In the case of a portfolio and it's complement, there are only two possibilities. One must be above the average, and one must be below (or else they can be exactly equal).

They must be the average of each other.  We're talking about the market average.  It is definitely possible to have two portfolios, both of which under perform the market. 

He's comparing X and 1-X, where 0<X<1.  Not X and Y where there is no relationship.  If X < 0.5, then 1-X  is > 0.5.

TGS, maybe I was unclear, maybe other people were making different assumptions than I was.  The author was talking about "funds" not companies.  I took that to mean sectors.  By the way it was worded, I also took it to mean short term performance, where a company with the characteristics you mentioned might have and be a "top performer", since short term looks like noise to me.  I very well could be wrong with my assumptions.

TheGrimSqueaker

  • Magnum Stache
  • ******
  • Posts: 2606
  • Location: A desert wasteland, where none but the weird survive
Re: Successful investor: How to get more in retirement
« Reply #24 on: November 29, 2017, 01:38:41 PM »

In the case of a portfolio and it's complement, there are only two possibilities. One must be above the average, and one must be below (or else they can be exactly equal).

They must be the average of each other.  We're talking about the market average.  It is definitely possible to have two portfolios, both of which under perform the market. 

He's comparing X and 1-X, where 0<X<1.  Not X and Y where there is no relationship.  If X < 0.5, then 1-X  is > 0.5.

TGS, maybe I was unclear, maybe other people were making different assumptions than I was.  The author was talking about "funds" not companies.  I took that to mean sectors.  By the way it was worded, I also took it to mean short term performance, where a company with the characteristics you mentioned might have and be a "top performer", since short term looks like noise to me.  I very well could be wrong with my assumptions.

A mutual, hedge, or investment fund is composed of large numbers of individual investments, is it not? If so, to have an underperforming fund would require the same kind of decision making as it would take to have an underperforming portfolio, i.e.: picking some severely mediocre investments. I think it would probably be easier to do this than it would be to create an average-performing fund, much less one that beats the average for a while.

Scortius

  • Bristles
  • ***
  • Posts: 475
Re: Successful investor: How to get more in retirement
« Reply #25 on: November 29, 2017, 01:49:28 PM »
You guys are missing one of the main arguments in favor of index funds, and that is that the distribution of performance has a high degree of skewness (third moment). Thus, it is much more difficult to pick winners than it is to pick losers. This reflexivity argument isn't valid for a skewed distribution.

scottish

  • Magnum Stache
  • ******
  • Posts: 2716
  • Location: Ottawa
Re: Successful investor: How to get more in retirement
« Reply #26 on: December 01, 2017, 03:43:43 PM »
You guys are missing one of the main arguments in favor of index funds, and that is that the distribution of performance has a high degree of skewness (third moment). Thus, it is much more difficult to pick winners than it is to pick losers. This reflexivity argument isn't valid for a skewed distribution.

Nice.  Do you have a good linky?

Scortius

  • Bristles
  • ***
  • Posts: 475
Re: Successful investor: How to get more in retirement
« Reply #27 on: December 04, 2017, 11:01:03 PM »
You guys are missing one of the main arguments in favor of index funds, and that is that the distribution of performance has a high degree of skewness (third moment). Thus, it is much more difficult to pick winners than it is to pick losers. This reflexivity argument isn't valid for a skewed distribution.

Nice.  Do you have a good linky?

Late bump as I just noticed this reply.

I remember reading a few articles posted here that really illustrated the point well, and a quick Google search finds a few, but perhaps not the best ones I remember.  Regardless, here are two from some of our favorite sources that help summarize the idea. I believe the first one was one of the ones I read earlier.

https://seekingalpha.com/article/4042396-many-investors-fail

http://www.etf.com/sections/index-investor-corner/swedroe-keep-skewness-perspective?nopaging=1

I seem to remember there being some seminal academic articles cited at some point, but I couldn't find anything that stood out on Google Scholar.


YoungInvestor

  • Bristles
  • ***
  • Posts: 409
Re: Successful investor: How to get more in retirement
« Reply #28 on: December 05, 2017, 04:57:54 AM »
This is a recipe for disasters. Based on regression to the mean, if you only pick past winners you are most likely picking future losers.

This understanding of regression to the mean is a huge pet peeve of mine, because it is flat out wrong in a probabilistic way.

If you spin a quarter 3 times and get 3 heads, your next one  still just has a 50% of getting tails.

However, over time, of you soon it another thousand times, your aggregate will look a lot like 50/50. That's what the regression to the mean is.

In the case of stocks, it means that future returns, likely being average, will, over time, bring the long term cagr to the market average.

Otherwise, as someone else noted, it would be easy to beat the market by just picking whatever's last year was bad.

EricL

  • Guest
Re: Successful investor: How to get more in retirement
« Reply #29 on: December 22, 2017, 10:18:57 PM »
The markets are doing really well.  So monkeys throwing darts at stock listings can make money.  Ditto day traders.  (Not that I’m comparing the two)  But when the market takes a downturn everyone suffers.  The best example I can think of was the day trader that made $100 million. Then had to join the Navy when the markets “corrected.”  Monkeys do OK though. They work for bananas.