United States Election Year Better for the Stock Market is a Hoax

By Lawrence

Heard too much bullshit on this topic lately so let’s clarify this issue once and for all.

Following is a chart with 2 distributions.

Top one (middle pane) is the distribution of yearly net changes in percentage since 1929.

Bottom one (bottom pane) is the distribution of same measure for the election years only.

MBO Dow_20120212_144013

What have we learned from this?

  • The normal bias for any year is 67% of the time since 1929 Dow has a net positive year. i.e. close of the previous year is less than the close of the current year
  • For the election years it is 66.7% which is weaker in comparing to the normal bias. It also means any other year that is not an election year actually has better chance of closing the year positive because the election years dragged the rate lower
  • There are only 21 election years since 1929 so the statistics is not even that convincing because there is not enough samples to start with
  • At best we can infer that the performance of election years are similar to those from normal years. Definitely not better.

Share

Comments
  • HiggsBoson February 15, 2012 at 2:03 pm

    Sure not what I would have expected.

    Interesting to see examples on ways to use NT’s distribution plot indicator. I will have to check that out next. Lately I have been stuck on your pattern scanner. I can’t stop testing different patterns on Russell 2000 stocks. I have pretty much gone pattern scanner crazy, but I will get over it. 🙂

  • Lawrence Chan February 16, 2012 at 11:27 pm

    I use pattern scanner for a lot of scheduled end of day work.
    Automation saves time. =)

  • You must be logged in to comment. Log in