Survivorship Bias (by Consistently Incredulous)

By -

I tripped across a Standard & Poors announcement last week that CBOE Holdings (CBOE) will replace Temple-Inland (formerly TIN) in the S&P MidCap 400 index as International Paper (IP) (S&P 500) completed its acquisition of Temple-Inland on February 13th.  This reminded me of a good Seeking Alpha post I read last year about Survivorship Bias in Index Performance.

I highly recommend the full post; but in a nutshell, survivorship bias in the indices:

“Specifically, in the process of rebalancing (selecting and or deselecting stocks) the indices it is the tendency for failed companies to be excluded from indices because they 1. No longer exist, 2. Their market capitalization has fallen or 3. Their industry is in decline (which likely caused the first two reasons); this is considered Type 1, survivor bias. Inherent in this type of bias is the error you make in just counting the survivors.”

A couple of other great lines from the post:

“Money managers, fund managers, investors and even Traders struggle with this issue of survivability bias because it can cause a real discrepancy between a thoroughly back tested trading model and the real life market. In mutual funds many well regarded fund managers believe that survivorship bias can also overstate a mutual fund's performance returns by more than 1.6%.”

“This is inherently problematic on two levels. On one level it creates a false level of optimism as the companies included in the index on or by some date excludes a significant number of major deadbeats, and in the same vein the companies that were replacements to the deadbeats most likely exhibited extraordinary growth in a short time period.  The second level is a bit more sinister as, when the ‘deconstructionists’ forgo inclusion of the dead beats in their five-year look backs, they gloss over the amount of risk you take on.”

As someone who primarily trades ETFs in intermediate-to-long time frames, I was more than a little concerned that survivorship bias affected my performance. So I took a look at the S&P500 close data back to September 1953 (the year we moved from 270-300+ trading days/year to the current 251-253) to see if anything jumped out relative to the 10, 50 and 200 SMAs and EMAs.  As one would expect, the % above is higher for the longer MAs:


In the period reviewed, price was above the 200d MAs over 2/3 of the time.

Although Standard & Poors is quite clear as to their current methodologies, I couldn’t find a running history of methodology changes – certainly not back to 1953 – to confirm which indices rebalance quarterly or by the same criterion. So I did a quick check of annual data on the 200s, and found no tendency (i.e. the data suggests we are neither tending to become more or less optimistic over time) with the annual averages almost spot-on at 67.3% and 70.2% on the 200SMA and 200 EMA, respectively:


As one would expect, the market reflected the economy.  Of the 59 years in question, there were 16-17 years where price was below the 200MAs > 50% of the time (28%, orange), and only 5 instances (8.5%, red) where price was below > 80% of that year.

Does survivorship bias affect my performance as an intermediate-long term trader?  Looking over my wins & losses over the last 15 years, I would say it did have an impact: generally speaking, it has augmented my long returns in non-recession years and augmented short returns in significant recession years (the red ones).  Where I tended to have weaker returns- both long & short- can essentially be summed-up when “fighting the trend” (including those periods when there was no trend).

I do realize that many Slopers have the ability to trade intraday, however for my current trading style; I will be more apt to lean on a buy/sell model 2/3 of the time and selectively short in periods of significant weakness.

One minute you're up half a million in soybeans and the next, boom, your kids don't go to college and they've repossessed your Bentley.”