Lib Dems and "proportionally earned seats"

I was having a look at data from previous elections recently with a particular focus on the number of seats and percentage of votes gained by the third party in the last few decades.

The first thing that is clear and which I already knew is that in the last three general elections, the Lib Dems have consistently increased their number of seats. The figures are:

1997: 46 (+26)

2001: 52 (+6)

2005: 62 (+10)

The huge leap in 1997 is often put down to our improved targeting campaign techniques championed by Chris Rennard.

There is something else interesting in the figures which I had not realised. I wanted to see what happened when you compare the percentage of seats we had won with the percentage of "proportionally earned seats" we would have got in a proportional system. I was particularly interested to see if and how this had improved over the years and indeed it has.

In fact when you look at the figures since 1970, this is what you get:

This chart shows vote share plotted against seat share:

As you can see there is not much correlation between the two. The vote share bounces around all over the place but the share of seats is clearly trending upwards with a few small interim falls.

This graph shows the "proportionally earned seats" share on its own though:

In every general election since the second one in 1974 (so the last 8, coincidentally every single one since I was born in July 1974) the proportion of seats won with respect to the share of the vote has increased.

The fact is that the third party has consistently got better and better at fighting first past the post elections.

Whenever I have been asked how I think the Lib Dems will do at the next election I have always said that I think our share of the vote will be higher than last time (22%) but that our number of seats may well fall. A number of other Lib Dems have taken me to task on this suggesting that I am not taking into account how difficult it is for our opponents to shift us once we win seats and that I am being too pessimistic. The data above backs up their view.

The history of elections in my entire lifetime would suggest that if the Lib Dem share of the vote goes up then our number of seats will too. If this doesn't happen then it will be bucking a 36 year trend.

I am becoming increasingly convinced that we will not lose seats at the coming election.

This has been cross-posted from Lib Dem Voice.