Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
Support MFO
Donate through PayPal
Mark Hulbert: When You Realize How Much Luck Goes Into Investing, You Might Change Your Methods
I too suspect that most investors do not fully understand the tradeoff that exists between skill and luck when making investment decisions. Luck is a far more significant contributor then is commonly appreciated.
We are fooled by randomness (that's the title of an excellent book authored by Nassim Nicholas Taleb). The likely reason why we are fooled is that we don't recognize how large numbers of participants contribute to a respectable number of winners.
For example, if 1000 market forecasters exist, after a single forecast 500 are probably correct given an equal interpretation of the likely market outcome. For the successful forecasters, repeat this test again, and the successful number is reduced to perhaps 250. If the challenge is repeated 8 times, a simple probability calculation suggests that maybe 4 forecasters would be correct on all the 8 tests. These fortunate four might be skillful, but they just might be lucky..
These lucky few announce their prescient calls and are now respected as market forecasting wizards. The large number of initial forecasting candidates almost guarantees this outcome and the subsequent misleading interpretation. Indeed, we are often victims; we are fooled by randomness.
I think we’re setting up for something like a repeat of ‘87. Market action past few days looks shaky. Anytime a lot of folks start thinking it’s easy pulling 15-20% a year you’re looking for trouble - especially at a time of 1-2% on CDs. Might make it into January. Doubt it.
Market crashes and corrections do have a lot to do with luck. Like the old game of musical chairs.
For the record: Current DJ 23,400, S&P 2578, NAS 6728, 10-Year 2.38%
Most market crashes have a trigger and in a lot of cases that's unexpected. The economy both in the US and globally is doing reasonably well and appears to be picking up. Even the skeptics accept that a recession isn't very likely at the moment. I would tend to view what's been happening so far in November as consolidation rather than worry but that's just my humble opinion.
@MJG, I think it reinforces your point but if someone gets 5 out of the 8 forecasts you mentioned correct they're brilliant. A lot more of the guys who are able to get 5 right are lucky but in the same vein they look, and probably claim they're really smart.
@LLJB That’s reassuring. And I can see your point of view. Still, it’s beginning to defy logic with interest rates so low. (I’m looking for the rip-cord right now.) Hope you’re correct. Yellen does too. She’d like to leave town first.
Re: “Most market crashes have a trigger and in a lot of cases that's unexpected”. Yes. Absolutely. Here’s a few potential triggers:
High Yield Bonds Flattening or Inverted Yield Curve Rising Delinquencies N. Korea South China Sea Bob Mueller Fredo Any combination of the above
"Regardless of whether luck accounts of 92% or 98% of investment performance, the implications are the same: Almost all attempts to beat the market will fail."
A misleading statement because he's conflating beating the market with beating half of one's peers. Some other problems with the analysis:
It assumes that years are fungible. By that theory, even if only giant caps are doing well in a given year, a genuinely skilled manager in large (not giant) caps should still outperform his peers. Sure. Market conditions change over time. I'm happy with a manager who does very well some years and not excessively badly other years.
Because performance tends to bunch near the middle, tiny differences get amplified in percentile rankings. A small change in relative performance can easily shift rankings across deciles for one year. There is less instability over longer periods of time (thus highlighting another problem with focusing on year 1 vs. year 2).
There are enough studies showing that management skill does exist to raise doubts about the approach here. The usual question is not whether skill exists, but how to identify it prior to investing. MJG alludes to this with his coin toss.
"research [by Stanford faculty] suggests that the typical mutual fund manager is persistently skilled, and that top performers are especially good. It’s just that the market is so hypercompetitive that most investors can't benefit from the skill ..."
The article's introduction grabs you like a good novel:
It had been a little over a week since anyone had seen Karina Chikitova. The forest she had walked into nine days prior was known for being overrun with bears and wolves. Luckily, she was with her dog and it was summer in the Siberian Taiga, a time when the night time temperature only dropped to 42 degrees (6 Celsius). However, there was still one major problem — Karina was just 4 years old.
Comments
I too suspect that most investors do not fully understand the tradeoff that exists between skill and luck when making investment decisions. Luck is a far more significant contributor then is commonly appreciated.
We are fooled by randomness (that's the title of an excellent book authored by Nassim Nicholas Taleb). The likely reason why we are fooled is that we don't recognize how large numbers of participants contribute to a respectable number of winners.
For example, if 1000 market forecasters exist, after a single forecast 500 are probably correct given an equal interpretation of the likely market outcome. For the successful forecasters, repeat this test again, and the successful number is reduced to perhaps 250. If the challenge is repeated 8 times, a simple probability calculation suggests that maybe 4 forecasters would be correct on all the 8 tests. These fortunate four might be skillful, but they just might be lucky..
These lucky few announce their prescient calls and are now respected as market forecasting wizards. The large number of initial forecasting candidates almost guarantees this outcome and the subsequent misleading interpretation. Indeed, we are often victims; we are fooled by randomness.
Best Wishes
Market crashes and corrections do have a lot to do with luck. Like the old game of musical chairs.
For the record: Current DJ 23,400, S&P 2578, NAS 6728, 10-Year 2.38%
@MJG, I think it reinforces your point but if someone gets 5 out of the 8 forecasts you mentioned correct they're brilliant. A lot more of the guys who are able to get 5 right are lucky but in the same vein they look, and probably claim they're really smart.
Re: “Most market crashes have a trigger and in a lot of cases that's unexpected”. Yes. Absolutely. Here’s a few potential triggers:
High Yield Bonds
Flattening or Inverted Yield Curve
Rising Delinquencies
N. Korea
South China Sea
Bob Mueller
Fredo
Any combination of the above
A misleading statement because he's conflating beating the market with beating half of one's peers. Some other problems with the analysis:
It assumes that years are fungible. By that theory, even if only giant caps are doing well in a given year, a genuinely skilled manager in large (not giant) caps should still outperform his peers. Sure. Market conditions change over time. I'm happy with a manager who does very well some years and not excessively badly other years.
Because performance tends to bunch near the middle, tiny differences get amplified in percentile rankings. A small change in relative performance can easily shift rankings across deciles for one year. There is less instability over longer periods of time (thus highlighting another problem with focusing on year 1 vs. year 2).
There are enough studies showing that management skill does exist to raise doubts about the approach here. The usual question is not whether skill exists, but how to identify it prior to investing. MJG alludes to this with his coin toss.
Here's a random page pulled in searching for an example of a study on management skill:
https://www.gsb.stanford.edu/insights/jonathan-berk-are-mutual-fund-managers-skilled-or-just-lucky
"research [by Stanford faculty] suggests that the typical mutual fund manager is persistently skilled, and that top performers are especially good. It’s just that the market is so hypercompetitive that most investors can't benefit from the skill ..."
Here's the working paper referenced in the article:
http://www.nber.org/papers/w18184.pdf
The article's introduction grabs you like a good novel: a-little-knowledge-is-dangerous
Could be Russian Fake news, but I hope not:
dailymail.co.uk/news/article-2721673/The-real-Mowgli-Russian-girl-survives-11-days-nights-lost-Siberian-wolf-bear-infested-wilderness.html