“Electoral history is littered with unexpected landslides”. That is the favourite saying of my friend Sir David Butler. During the period when I was described as “Australia’s leading psephologist” (from about 1968 to about 2008) he was described as “Britain’s leading psephologist”. However, he enjoyed that description for much longer than I did. Indeed he assisted in coining that new-fangled word “psephology” which means the study of elections, election trends and electoral systems.

Anyway, the reason for his coining of such a dictum was his experience. Because no one ever REALLY knows the result of an election in advance pundits tend often to say: “It will be close”. For my part I can think of quite a number of cases where the expected winner DID win but the victory was significantly bigger than anyone expected. The British general election of 1945 is the first example. There was a general feeling that Labour would probably win but no expectation of the landslide defeat for Winston Churchill’s Conservative party which actually occurred. In the United States in 1980, there was a general feeling that Ronald Reagan would probably win but no expectation of his landslide win – and certainly no expectation that he would lead his party to twelve years in the White House and he himself would become one of the celebrated “great” American presidents of the 20th century, the other four being Franklin Roosevelt, Theodore Roosevelt, Harry Truman and Woodrow Wilson.

I can think of these Australian landslide victories in the same kind of atmosphere. Federally there were the Menzies victory of 1949 and the Fraser victory of 1975. In New South Wales there was the Wranslide of 1978. In Queensland there were the landslide victories of Peter Beattie in 2001 and of Campbell Newman in 2011. In South Australia there was the landslide victory of the Liberal Party in 1993. In Western Australia there was the landslide Labor win at the general election held on Saturday March 11 of 2017!

The results of this latest WA election are now complete. In the case of the Legislative Assembly the result was fully known on Monday March 20 and it was a Labor win with 41 seats, a majority of 23 over the combined non-Labor numbers of a mere 18 seats, 13 Liberals and five Nationals. The result of the Legislative Council election was fully known on Saturday March 25 and it produced a slightly better result for Labor and The Greens than I predicted. Details are given below. The vital statistic for the Legislative Assembly election is that there was a two-party preferred vote swing to Labor of 12.5%. Labor’s share in March 2013 was 42.7% and it rose to 55.2% this time. For Liberal-National the statistics are 57.3 and 44.8, respectively.

On Thursday February 9, a month and two days out from polling day, I had this article published on the SWITZER website “Leadership shakeup for WA and SA?” in which I predicted that Labor’s Mark McGowan and Steven Marshall of the Liberal Party would be the new premiers of Western Australia and South Australia, respectively. We can now remove the question mark for WA and I have no doubt that – a year hence – we shall remove it from SA also. However, I wrote these detailed predictions for the Legislative Assembly election:

“Two Labor-held seats – Collie-Preston and West Swan - have become notionally Liberal on the new maps. My prediction is that Labor will get a big enough swing to hold both Collie-Preston and West Swan. Labor will also gain these seats from the Liberal Party, Balcatta, Belmont, Bicton, Forrestfield, Joondalup, Kalamunda, Morley, Mount Lawley, Perth, Southern River and Swan Hills. That means Labor will, on my prediction, have 33 seats. I think the Liberal Party will get a consolation prize, however. The Liberals will gain the seat of Roe from The Nationals. . .So the total number of Liberals will be 20 and there will be six Nationals.”

As things turned out there was a seat shifting from National to Liberal but it was not Roe, as agricultural a seat as one would ever find. It was the mining seat of Kalgoorlie. But The Nationals also lost their leader’s seat of Pilbara to Labor. It was a win for the big mining companies who would have paid more tax if the tax reform promoted by Brendon Grylls had been implemented. Apart from losing the mining seats of Kalgoorlie and Pilbara this was a good result for The Nationals. It was an exceptionally good result in their area of traditional strength, agricultural Western Australia.

It is worth noting that EVERY seat I predicted Labor would gain from the Liberal Party WAS actually gained by Labor. Therefore, it is also worth noting some details of the seven seats gained by Labor I predicted would stay Liberal. I list them in order of the size of the swing needed by Labor as shown on my pre-election pendulum. They were Jandakot (18.1%), Kingsley (14), Darling Range (13), Bunbury (12.2), Murray-Wellington (12), Burns Beach (11.4) and Wanneroo (11.2). In essence, therefore, whereas I predicted a solid Labor win, those extra seven gains from  Liberals plus Pilbara gained from The Nationals turned the predicted solid win into a result now universally recognised to have been a landslide. It delivered Labor its biggest win in the state’s history with a majority of 23 seats compared with the seven-seat majority I predicted.

Of the seven seats named above the swing was biggest in Bunbury where the sitting member had retired. It was 23.1%, a clear case of a very large swing due to the combination of “retirement slump” (an American psephological term, self-evident in its meaning) and the general swing. The second and third biggest swings were in Darling Range and Jandakot, each being an even 19%. The fourth biggest was in Wanneroo where it was 18.5%. Kingsley at “only” 14.7%, Burns Beach at “only” 14.4% and Murray-Wellington at “only” 13.4% were moderate swings by comparison. Of course all those swings were higher than the state-wide swing of 12.5%. In each of Jandakot, Kingsley, Burns Beach and Pilbara Labor won the seat by defeating a minister in the Barnett government.

I asked my friend in the WA Liberal Party whether there were ANY individual results which could be said to have been good for the Liberal Party. He nominated Kalgoorlie, Dawesville and Riverton, in that order. My pre-election pendulum showed that Dawesville needed a swing of 12.7% for Labor to win. However, it was a good Liberal result because the sitting member had retired and the swing to Labor escaped the effect of retirement slump illustrated so starkly by Bunbury. The same pendulum shows that Riverton needed a swing of 12.8% for Labor to win. It had been noticeably strengthened by boundary change but it was in the past the nearest seat to the median on my pendulums, including being the ACTUAL median at the elections of 2001 and 2013. It was a Labor seat from 2001 to 2008 when Mike Nahan took it for the Liberal Party. On Tuesday March 21 the decimated Liberals met to replace their leader so Nahan is now the Leader of the Opposition.

The last paragraph of my February 9 article read this way: “Under the semi-proportional system for the Legislative Council (where there are 36 members) the result in March 2013 was 17 Liberals, 11 Labor, five Nationals, two Greens and one for the Shooters and Fishers Party. My prediction for this election is 14 Labor, 13 Liberals and three each for Nationals, Greens and Pauline Hanson’s One Nation Party.”

In other words I predicted that the Labor-Greens combination would be 17 seats and the combination of the rest would be 19. The actual result is 18 apiece. However, Labor will need to provide a President of the Legislative Council who loses his/her vote on the floor as a consequence. Labor and The Greens, therefore, will quite often be out-voted by the rest, if they combine together. The new numbers are 14 Labor, nine Liberals, four each for Nationals and Greens, three for Pauline Hanson’s One Nation Party and one each for the Liberal Democrats and the Shooters, Farmers and Fishers Party.

So, what lesson for South Australia can we learn from Western Australia? My belief is that the feeling will grow it will be a Liberal victory of landslide proportions. Perhaps, then, it will be a close result. I am inclined to reverse Butler’s dictum and say this: “Electoral history is littered with expected landslides which turn out to be close results”.