Will a majority of voters in Britain’s upcoming referendum elect to remain in the European Union? (update)

Are there do-overs in this league? After a bit of reflection on my last forecast, I think further revision is in order. The central thesis of my last forecast was that, if we squint really hard, we can probably see the beginning of the Renwick effect[a]. I now think that I was mistaken. We might or might not see the Renwick effect in the future, but we’re not seeing it yet. Or at least I’m not seeing it yet, which is a somewhat different proposition.

Why the change of mind? Using the metadata from “The Financial Times”[b] we see that

  • The average results for the Brexit polls in April (so far) are as follows: stay=42.75, go = 40.83, don’t know=14.58.
  • The average results for the Brexit polls in March were as follows: stay=42.87, go=40.50, don’t know=18.75.
  • The average results for the Brexit polls in February were as follows: stay=43.38, go=39.38, don’t know=15.77[c].

I grant you that I’m taking all of this data at face value and not attempting to weight for the reliability, size, or temporal proximity of the polls. I leave that to smarter folks. But I just don’t see what I thought I saw: the beginning of a shift toward the status quo (i.e. stay in the EU) that the Renwick effect predicts. In fact, we see a very slight movement _away_ from the status quo, though I think that the size of the shift is too small to mean much. What’s most remarkable to me is the stability of the opinion polling data over the last 3 months. (It would be super-freaking-awesome if someone tried to determine whether this level of stability prior to the referendum predicts a greater or lesser degree of variability in the final weeks before the referendum, but the low sample size might render this information less than fully helpful.)

You know, this one might be a nail-bitter, folks.

[a] https://electionsetc.com/2016/03/14/first-forecast-for-the-brexit-referendum/

[b] https://ig.ft.com/sites/brexit-polling/

[c] So “stay”+”go”+”don’t know” doesn’t always equal 100% in the average of polls because it doesn’t do so in some of the individual polls that the FT reports. That’s odd; this seems like an exhaustive trichotomy, but I don’t think it makes too much of a difference to the outcome.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s