Cycles Analysis: Predicting A Trump Victory

|
Includes: SPY
by: Topdown Charts

Summary

The presidential polls appear to show a repeatable short-term cycle.

Applying cycles analysis results in a projection that Trump's polling will peak around the date of the election.

If the analysis proved prescient it could be flagging a non-consensus outcome that could deliver a shock to markets, particularly the S&P500 and the Mexican peso.

While I don't count myself as a political expert, I do have a certain level of expertise in market analysis, and this often involves the analysis of cycles. In this piece I apply some basic cycle analysis methodologies to the cycles in the US presidential polls - specifically for Trump.

Note: this is not a political view, this is an exercise in cycle analysis that has meaningful implications for investment strategy.

While the US presidential election outcome may or may not have the same dramatic effect as the Brexit, it's worth noting that there remains a divergence between the polls and the betting markets (just like Brexit) [less than 20% betting odds for Trump vs. ~40% in the polls], and while sentiment seems to have gone against Trump lately, the following analysis shows that it could still end up a very close race.

The basic approach for measuring cycles is to measure the time elapsed from bottom to bottom, I have applied this method to a 10 day average of the daily average of the 4 most recent polls (a smoothed poll of polls) which is supposed to give a better representation of the trend in voter preferences.

In doing this we can calculate the average cycle length - given that we know when the last cycle bottom was we can thus project when the next cycle low will be using the historical average. What we can also do is measure the average duration from the bottom to top of cycles and thus project when the next top or peak may occur.

Using the averages the results are that the bottom would be the 12th of October and the next peak the 5th of November. There are a number of caveats to the analysis e.g. this is just a measure of duration of cycles, not magnitude... also it uses averages and there is quite a bit of variance within the average.

But the key point is that if the cycle continues as per usual then Trump's polling may peak very close to the election date. So don't rule out anything yet - it could still be a very close race.

Investment Implications

As for the market implications, a couple of charts are worth highlighting. First is the presidential polling spread compared to the S&P500.

Aside from the potential adverse impact of rising policy uncertainty on markets, an initial shock effect from a "non-consensus" or "unexpected" outcome would rattle markets. This is what I was referring to earlier when I mentioned about the divergence between the betting markets and the polls. While the link between the S&P500 and the polling spread day to day is fairly loose - you might argue that the market leads the changes in the polling spread, but either way, on the day what will matter is the outcome vs. consensus.

The other market where there is a more clear link is the USDMXN (Mexican Peso). Just like Brexit, the currency was about the most pure way to express a view on the politics. For now the peso is trading at levels that could be said as complacent with regards to a potential Trump victory. Thus the peso is a potential market to take a view in or use as a hedge.

Bottom line: Investors need to take note of the cycles analysis exercise in this article which projects Trump's polling to peak around the date of the election. This would present a potential non-consensus outcome which could shock markets such as the S&P500 and the Mexican peso.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.