136 Hours

Less than a week

On Monday I will be hosting an Ask Me Anything for Spectra School and this one will be open to the public. Please email your questions to me and/or [email protected] and we will answer as many as possible in the one hour time allotted. Questions can be election related, trading and risk management related, or anything else! Here is the link to the AMA. It will take place at 4 p.m. NY time, November 4.

Out of curiosity, I took a look at how SPX and the US 10-year yield moved around prior elections.

2000 and 2008 stand out as those were recessions and massive equity selloffs both before and after the election, while the 2016 result captures the unexpected move towards stronger fiscal stimulus. This time, a Trump victory seems significantly priced, and a large fiscal impulse is thus also priced in. It’s very difficult to separate the momentum in Trump’s gambling odds, the momentum in the Trump trades (bitcoin, US yields, USDJPY, DJT stock, etc.) and the fact that the polls and best predictors are all still calling it a 50/50 race.

There seems to be this idea that election polls substantially underestimate the Trump voter and therefore cannot be trusted. But polling errors are more mean reverting than autocorrelated.

“We don’t always see the misses in the same direction,” said Chris Jackson, the senior vice president of public affairs for Ipsos. “I can tell you that the polling industry has done substantial changes to how we do our surveys to try to account for what we think was driving those errors in 2020. So while there undoubtedly will be errors in the future, they’re probably going to be driven by different things and go in different directions.” 

Here is the history of polling errors in the US.

Note that the media momentum in the 2022 mid-terms was strongly towards the Republicans and a Red Wave. That Red Wave never materialized.

Instead, the polls were pretty accurate, and the Democrats did well. Roe v. Wade and other issues helped the Dems. Also note that the sign of the polling error in the UK was red in 2010, blue in 2015, red in 2017 and then blue in 2019 and again in 2024. There is no guarantee that the polling error in the US election next week will be in favor of the Republicans. Polling errors are more random, mean reverting, and prone to overcorrection—not autocorrelated.

I just say this because there seems to be this air of inevitability around the result, and the result is anything but inevitable. If Trump does not win, there are going to be some massive market moves, as USDJPY, USDCNH, and yields will absolutely buckle now if Harris wins. Current levels now make a bet on blue more interesting than a bet on red. We have seen few (but not zero) bets like this put on by our client franchise—most of our customers here at Spectra have been focused on USDCNH, USDMXN, and USDCAD topside, along with EURUSD downside.

---

VW

EURUSD failed to bounce on the stronger inflation data today, which makes sense as the US data comes in strong again and Volkswagen’s woes remain in the spotlight. Job cuts and wage reductions there, along with falling oil prices, make the idea of any sort of inflationary impulse in Germany look extremely unlikely.

---

Final Thoughts

The BOJ tonight is expected to deliver nothing, and given zero expectations, it’s a bit of a free option to go in short USDJPY and cut quick if nothing hawkish emerges. The severe weakness in the JPY means there is a tiny, but not zero chance that the BOJ just gets busy hiking rates now instead of waiting until December or January.

Silver, bitcoin, gold, NASDAQ. Oh my.

*SUPER MICRO SAYS AUDITOR ERNST & YOUNG RESIGNED AMID REVIEW – oh dear.

Have a clever day.

https://x.com/annechovy2/status/1849948490597794221?s=51

HT  GM

good luck ⇅ be nimble