Polling Projected Inadequately | Autonomy Scotland

Polling Projected Inadequately

In a recent post entitled Polling Projected Properly the Reverend Stu uses some figures which he suggests show us that voting for a smaller pro independence party in the regional vote is counter productive. In the comments section and online many have taken this article’s conclusion as further justification for an SNP/SNP vote. However, the figures used don’t actually show what that article says. The figures chosen actually support the argument that people who want to maximise pro independence seats should vote for a smaller party.

To understand why you will need to first know how the Scottish voting system works. And if you don’t you can read our guide to that here.

In the Wings article the Reverend makes a projection of seats by comparing the votes in 2011 to the polling data shown in the recent Electoral Reform Society report.  In that way he works out a figure to multiply the 2011 regional results by in order to project the 2016 results. In his own words:

For example, the SNP got 45.4% in 2011 and are now polling an average 46%, so to save on some calculations and slightly bias our experiment in favour of the smaller parties we rounded their multiplier down to 1. (It should strictly have been 1.013.)

Similarly, Labour’s current 19% polling average is 0.72 times the 26.3% they got five years ago, so that became their multiplier. For the Tories the number came out to 1.2, for the Lib Dems 0.76 and the Greens 1.8 (because they’re polling at an average of 8%, almost twice their 2011 share of 4.4%.)

The most important point to note is in his projection the Greens are being given just under double the regional vote they achieved in 2011. This represents about 8 percent of the regional vote which would be a good achievement.

So far so good.

For some reason the Reverend then decided to compare the amount of pro indy seats on the list in 2011 to the amount in the projection for 2016 and rightly concludes that there are 5 fewer pro independence seats on the list. However, the glaring error is to attribute this reduction to the shift to the Greens. If you are going to compare actual 2011 results to a projection for 2016 what you would need to do is clearly show the overall pro indy seats in both constituency and region as one affects the other. The Reverend glosses over the constituency part.

There are currently 71 pro indy MSP’s but there are 83 in the Wings projection. So, overall there is a big jump in pro indy MSPs in the Wings scenario. And the reason for this increase is that the SNP are polling to do much better in the constituency vote. It is their success in the constituency vote that is the reason for the reduced pro indy list seats. Not the Green surge. Quite simply, the SNP are doing too well in some constituencies to win any votes in the regions.

That said, comparing 2011 list votes with projected 2016 list votes isn’t helpful in order to make a decision on your regional vote.

In order to do a proper analysis, what we should be doing is comparing total indy seats projected when the Greens have 8 percent with total indy seats projected when the Greens have what they achieved in 2011. When you put the Greens back on 4 percent of the vote and give those votes to the SNP then there are 80 pro indy seats projected. So by giving the Greens fewer votes using the figures Wings used you actually have 3 fewer pro indy MSP’S.

In the Wings figures a doubling of the 2011 Green regional vote to the level they are currently polling gets 3 extra pro indy MSP’s. 

So if looked at properly, the figures Wings used are an argument for tactical voting in some regions not against it. I have put my workings on another page. The main findings are.

If you reduce the Green vote by half(2011 levels) and give those votes to the SNP we can can see the following happen in each region:

  • In Central nothing changes as the SNP win 1 seat in both scenarios and the Greens win none.
  • In Glasgow the Greens go from 2 seats to 1 and the SNP continue to win none. So, indy majority decreases with no SNP gain.
  • In Highlands and Islands number of pro indy seats is 4. The SNP take all 4 compared to SNP 3 Greens 1 in the Wings projection.
  • In Lothian the Greens go from 2 to one and the SNP win none. So, indy majority decreases with no SNP gain.
  • In Mid Scotland and Fife, the 1 pro indy seat remains the same but the seat switches from Green to SNP.
  • In N/E Scotland the 1 pro indy seat remains the same but seat switches from Green to SNP.
  • In South Scotland the number of pro indy seats remains at 2 but the Greens lose their projected seat to the SNP.
  • In West of Scotland the 1 projected pro indy Green seat is lost.

So, using the Wings figure, a non SNP second vote makes sense in Lothian, Glasgow and the West of Scotland as in each of these areas reducing the projected Green vote by half, does not lose the SNP any seats, and actually gifts the Unionists one seat at expense of the Greens. In the other areas reducing the Green vote by half damages the Greens at expense of the SNP but doesn’t reduce overall pro indy seats.

That’s not to say SNP 1 AND 2 is not a sensible vote. It is if you want to maximise SNP seats for whatever reasons, for the SNP definitely get more seats when they take half of the Green’s 8 percent projection. I’m voting SNP/Green because I support the Greens and want them to do well. However, for those who are non aligned to a political party but want to maximise pro indy seats, in regions where the SNP are projected to get all the constituency seats then the percentage bet is to vote Green with your second vote. It’s not an infallible tactic as there is no certainty but it leads to a much reduced chance of wasting your vote. I would promote the same tactic for Rise if they were polling as well as the Greens.

Just to labour the point, using the figures that Wings used for the Glasgow region the results were two pro indy seats:

Lab 4
Grn 2
Con 1
SNP 0 

If you divide the amount of seats the Greens achieve by half and give those votes to SNP, putting them back to 2011 levels, the results are:

Lab 5
Grn 1
Con 1
SNP 0 

So in that scenario Labour have been gifted a seat. And the SNP get no seats at all despite getting 95000 votes. That is a lot of votes for probably no extra seats. Same in Lothian where SNP can get 130,307 votes with no seat. And in West of Scotland where they could get 124,878 votes with no seats.

You could argue that the Wings/Electoral Reform Society data may not be reflected in the actual ballot which is a fair point. But it is the best info you will have in order to decide if your SNP second vote will count.

Just to clarify, I don’t think any pure SNP supporter should vote anything other than SNP/SNP, although in some regions you will be wasting that vote but it is worth it to err on the side of caution. However, for those on the pro indy left, in some regions voting Green is the sensible thing to do, and the figures used by Wings back this up.

Please comment below and enter an email address to receive notifications of new articles. Also, if you enjoy our content you can support the site for less than one pound per month by clicking here.


Spread the love

You may also like...


This site uses Akismet to reduce spam. Learn how your comment data is processed.

Inline Feedbacks
View all comments
Would love your thoughts, please comment.x