Most Stocks Lose Money

[MachineGhost]

From 1983 to 2007, the Russell 3000 rose nearly 900%, yet:

* 39% of stocks had a negative total return.
* 18.5% of stocks lost at least 75% of their value.
* 64% of stocks underperformed the Russell 3000.
* A small minority of stocks significantly outperformed their peers.
* 25% of stocks accounted for all of the gains.
* 75% of stocks collectively had a total return of 0%.

[AgAuMoney]

From 1983 to 2007, the Russell 3000 rose nearly 900%, yet:

* 39% of stocks had a negative total return.
* 18.5% of stocks lost at least 75% of their value.
* 64% of stocks underperformed the Russell 3000.
* A small minority of stocks significantly outperformed their peers.
* 25% of stocks accounted for all of the gains.
* 75% of stocks collectively had a total return of 0%.

Wow, nice stats.  Where'd you find them?

64% of the index underperformed that index.  That's nearly 2/3rd's.  Couldn't have a better counter example to the fallacious "half better and half worse" comment that so often appears.  (That line of thought confuses the mean and the median.)

Which really means the odds of picking a winner by chance are worse than 50/50.  The challenge in any stock picking given these statistics is to pick sufficiently MORE of the 25% with positive return than of the 39% with negative.  Obviously you cannot do that by random chance (assuming you were going to be picking enough to actually get a random selection).  Or in other words, you have a better chance to rule out the bad than to pick the good.  (And the way I begin is by eliminating any that do not have a sufficient record of growing their dividend.)

I'm not sure how to interpret 75% net to 0%.  Given that 39% had negative return, some of those 75% must have had positive return to make the entire set net to 0%.  But should not those positive returns have been included in the 25% that "accounted for all of the gains" ?

Or maybe the 25% and 75% just coincidentally add up, and some companies appear in both sets.

[MachineGhost]

I tracked down the study: http://www.theivyportfolio.com/wp-conte ... bution.pdf

[clacy]

Ok, now we just need to figure out how to find the 25% of stocks that account for all the gains. Which stock is the next AAPL?

[MachineGhost]

Ok, now we just need to figure out how to find the 25% of stocks that account for all the gains. Which stock is the next AAPL?

This is where having access to expensive backtesting software with point-in-time fundamental data would come in handy.  Anyone got $15,000?  I will make sure you get your money's worth -- guaranteed!

The more pedestrian among us can just avoid all stocks not making a 52-week or all time high.  Since a market-cap weighted index is a momentum strategy by definition (at least those indexes that don't penalize winners by removing and promoting the stock to another higher, market cap weighted index), you only need to avoid all of the crap in the index to outperform.

[TripleB]

I love this thread and it's giving me insight into my own 31-stock portfolio I constructed 2 weeks ago to replace an Index fund in my PP (against the advise of HB).

Interesting stats in my own 2 week sample that mirror this post.

I purchased 31 stocks in various sectors. Some sectors have no representation because I simply don't want to invest in them, such as Telecomms and Financials.

Of the 31 stocks,
78% lost money
22% made money

61% underperformed the SP500
39% overperformed the SP500

OVERALL

SP500 = -2.58% returns over this time period
BBB's 31-stock Portfolio = -2.12%

I beat the market, even including the $7 trading fees (7*31 stocks= $217) that are figured into the cost basis.

2 weeks is meaningless and insignificant so I'm not touting my own horn too much.

Interestingly, I heavy weighted Defense Contractors and they are down about 5% over the last 2 weeks due to Obama's victory. 5 out of 31 of my stocks are Defense Contractors or 16%.

1 Stock gained over 10% (a Healthcare)
1 Stock gained 5% (another Healthcare)

Only 7 of the 31 stocks have any positive gain, and other than those 2 that I mentioned, the other gains are trivial (like 0.5% gain for example).

It's hard to say that Covance (CVD) with 10%+ return carried the portfolio because it's only about 2% of the total portfolio (I didn't equal weight all 30 stocks). Thus CVD only accounted for 0.2% gain of the total portfolio.

For the most part I selected a few stocks that did HORRIBLE compared to the index (in this case, Defense Contractors), I selected a few stocks that did GREAT compared to the index (healthcare) and I selected a bunch of stocks that did marginally better than the index and a bunch of stocks that did marginally worse than the index.

This experiment of the last 2 weeks goes to show that the old research showing that you need 30 stocks to get diversification is spot on. If I only selected 5 or 6 stocks from my pool of 31, my results could have a remarkably different spread, depending on which 5 or 6 I picked... ranging from losing to the index by several percent to beating the index by several percent.

Instead, by picking 31 diversified stocks, I was able to get really close to the SP500 on a daily basis, and over the 2 week period. I looked at it daily and I was beating the market by 0.1% most days and losing by 0.1% on a couple days... averaging out to an insignificant 0.46% gain over the SP500 over 2 weeks. Extrapolated out to a full year that's 11% annualized, however I'll likely have 2 week time periods where I lose to the SP500 by 0.5% as well too.

Also, I'm "cheating" in that 2 of my 31 holdings are actually funds. One is the VG Utility Fund VPU and the other is VGENX energy. I like VGENX rather than VDE because the mutual fund hold international stocks and VDE is all domestic. I decided to go with this route because I believe utilities and energy are too difficult to represent in 3 to 4 holdings each sector due to the large array of things they cover. Additionally, I find a lot of them fungible. When it comes to other sectors, I definitely believe Target is better than Wal Mart, but with respect to a Utility in Georgia versus a Utility in Maine, I can't really compare them and feel like indexing in that sector makes the most sense on a cost adjusted basis.

[TripleB]

Of more interesting note to my above post, I also track the PP the same way.

PP is up 0.01% over the same time period where the SP500 lost 2.58%

Gotta love the PP!

(Note - my PP is up more than 0.01% because my stock portion beat the SP500 by 0.4% so I'm up 0.11%)

[Thomas Hoog]

interesting

My results: dated: 1 -11 -2012 (dividend & costs not included)
Period: 2002 - 2012; 10 years
Global stocks
Benchmark: MCSI AC World standaard core: (- 5 %,  )
My stocks: 1 % 
Stocks with negatieve return: 38 %
worse then 75 %: 6 %
Top 3:
IBM: + 64 %
Fidelity Funds - South East Asia Fund A-Acc-EUR: 62 %
Robeco Emerging Markets Equities D (EUR): 57 %

Top 3:
Bank of America:  - 86 %
BAM (dutch constructor) : -75 %
BBVA (spanish bank): -55%

You win some, you lose some