There is little doubt that the Liberal Democrats shock win in the Richmond Park by-election, overturning Zac Goldsmith’s 23,000 majority, has left many in Westminster scratching their heads. On a relatively high turnout for a by-election, 53.6%, the Liberal Democrat Candidate Sarah Olney took 20,502 votes (49.7%), up 30.3% points from their 2015 General Election share, compared to independent Zac Goldsmith’s 18,638 (45.1%).

However, in many ways, the result is not that surprising. We present four main factors that meant the Richmond Park by-election turned into something of a “perfect storm” for the Liberal Democrats on the 1st December.


BMG bench-marking poll already showed Lib Dem gains

Firstly, BMG’s bench-marking constituency poll showed that the Liberal Democrats were already set to improve significantly on their 2015 vote share – it was clear that the trajectory was very much in the direction of the Liberal Democrats.

Conducted between 26-27th November, only 48 hours after Zac Goldsmith announced that he intended to resign his seat and stand as an independent, our poll put Mr Goldsmith ahead, but with the Liberal Democrats already up by 10 percentage points on their 2015 result. It is important to stress that this improvement in the Lib Dem numbers was detected by BMG prior to any major campaigning in the constituency.

The press also reported that internal campaigns data for the Lib Dems showed that the switch happened very late on in the campaign. More than a week out from polling day, leaked information from sources within the Lib Dem campaigns team suggested that their own canvass returns were telling them Zac was still ahead, by around 10 points.


Evidence of a large change in focus over the campaign period

Secondly, the campaign period allowed the Liberal Democrats time to overturn voters’ focus on to the issue of a 3rd runway at Heathrow, which fell silent in the national press during that time, onto some of the national issues of the day, namely Brexit. It is important to remember that polls, especially those conducted several weeks prior to the ballot, should only be viewed as a snapshot at that particular point and time, and not as a prediction for the actual result. Remember also that more than one fifth of those who responded to our poll were undecided and in our article detailing the results, we warned that Goldsmith’s “majority could easily drop to within reach of the Liberal Democrats if they put up an effective campaign that cuts through on national and not local/personal issues.”

Olney, by aligning herself to Mr Goldsmith by opposing Heathrow expansion, and with the Conservatives opting not to field a candidate, the issue of Heathrow was effectively neutralised. The by-election, therefore, turned into a contest where the issue of Brexit was paramount, with the Liberal Democrats putting Goldsmith’s support for the UK leaving the EU to the front and centre of their Campaign. Indeed, BMG’s October constituency poll found that, even in the early stages of the campaign, Brexit was seen as the most important issue to voters, chosen by 25% of respondents, compared to the 21% that selected Heathrow.


Richmond a ‘liberal’ Seat

Thirdly, the seat of Richmond Park could have hardly been more favourable to the Liberal Democrats. Whilst Mr Goldsmith won a 23,000 majority in 2015, it is important to remember that this was achieved following a big swing from the Liberal Democrats to the Conservatives across the board. In 2010, Goldsmith only has a small majority of around 4,000, and prior to this, the seat had been in Liberal Democrats hands since 1997. Furthermore, the seat was also perfectly suited to a Brexit and anti-Government protest, with an estimated 70% of voters in the constituency voting to keep the UK inside in the EU in June.


Zac didn’t have a campaign machine, the Lib Dems did

Finally, whilst Goldsmith was by all accounts a popular local MP, by opting to stand as an independent, he was at a big disadvantage when compared to the Lib Dem campaign machine. Not only will he have lost the “brand” recognition of being a Conservative, Goldsmith will have been without the Conservative constituency campaign data and their resources and personnel that the local party will have provided him.

By contrast, it was reported that the Liberal democrats had more than 700 volunteers knocking on doors in latter stages of the contest. Whilst senior Cabinet Ministers were unable to campaign for Goldsmith, the Liberal Democrats flooded the constituency with senior figures – it is reported that leader Tim Farron made 10 visits and Nick Clegg made six during the campaign.

Considering the intensity of the Lib Dem ground campaign, turnout is likely to be key in understanding the outcome. And that could be a key distinction in the interpretation of these results. It will be interesting to see whether voters actually switched in large numbers from Goldsmith to Olney, or whether many Goldsmith’s supporters simply didn’t turn out on the day. Indeed, with the turnout down over 30 percentage points from 2015, whilst the Liberal Democrat vote increased by around 30 points, they only won around 9,000 extra votes. So to deliver those extra votes took immense resource.

Richmond Park By Election Result (changes from 2015 General Election)

Liberal Democrat, Sarah Olney, 49.7% (+30.41)

Independent, Zac Goldsmith, 45.2% (-13.06)

Labour, Christian Wolmar, 3.7%(-8.67)

Monster Raving Loony, Howling Laud Hope, 0.5% (N/A)

Independent, Fiona Syms, 0.4%, (N/A)

Christian Peoples, Dominic Francis, Stockford, 0.4%, (N/A)

One Love, Maharaja Jammu and Kashmir, 0.2%, (N/A)

No label. David Powell (0.1%), (N/A)


BMG voting intention results (changes from 2015 General Election)

56% (-2) – Zac Goldsmith

29% (+10) – Sarah Olney/Lib Dems

10% (-2) – Labour

5% (-5) – Other

Base = 543 UK adults Aged 18+; Fieldwork conducted by telephone 26-27 Oct 2016.


Most Important Issue Results:

Brexit – 25%

Your local MP’s record/views – 22%

Heathrow 3rd Runway – 21%

Support particular party/candidate – 15%

Another issue – 9%

The Government’s record – 9%

Base = 543 UK adults Aged 18+; Fieldwork conducted 26-27 Oct 2016; note that results may not sum to 100% due to rounding.

*Note: Results may not sum to 100% due to rounding. Figures in brackets are changes on 2015 General Election. Readers should note; 19% said they were undecided; those who didn’t indicate which way they were leaning were excluded to calculate the above figures.

BMG Research surveyed a representative sample of 543 adults living in the Richmond Park constituency by telephone on 26-27 October. Data are weighted. Margin of error is approx 4.2%.

BMG Research are members of the British Polling Council and abide by their rules.

For a more detailed breakdown of results from this poll, or any other results from our polling series, please get in touch by email or phone.


0121 333 6006

robert-struthers Robert Struthers – Graduate Research Executive

Share this article: