Following a crash of our database server last night, I have migrated the database to Amazon’s new Relational Database Service. Hopefully, this will lead to better scalability and reliability. Some people (Seth O Logue and Old Nat…) lost comments in the migration, and I’m sorry about that. It is possible we will see some performance issues over the next couple of days as I tweak the configuration, but hopefully these will be short-lived. Right: off to bed now. Robert
Five Reasons Why UKIP Won’t Win In Buckingham (1) The Speaker’s Absolute Number of Votes Holds Up Very Well. Perhaps it’s because the speaker is a naturally prestigous role, but contested by the other parties or not, the speakers’ vote has tended to hold up very well. In fact if you look at the average votes the speaker has got, going back to 1974, and compares it to the average that parties polled in the surrounding elections then you see…
Two weeks ago, I introduced a new election forecasting model. This generated much heat, and a little light in the numerous comments. Many were scepticical, arguing that it used proportional swing, others asked how a model that produced such different results to Unified National Swing could be taken seriously. Many of the criticisms came from a lack of understanding of how the model – which we’ve nicknamed Voter Intention Prediction Analysis, or VIPA – works. This post explains why VIPA…
(This is a post from Robert Smithson, not Mike Smithson!) Electoral Calculus says a Conservative majority of 26.. Anthony Wells reckons it results in a 12 seat majority.. Hill & Knowlton has it as No Overall Control.. But could it really be 150 Tory majority? Yes. Because traditional election forecasting tools are flawed. Electoral Calculus, and the rest, use Unified National Swing as the bedrock of their analysis. But this has a fundamental defect: it assumes parties pile up votes…
(Spare a thought for our poor server) Robert
A message from Robert Smithson – who keeps PB running As some of you may have noticed, pb.com just had an hour of downtime. Checking through the logs, our machine has been running at 100% capacity for the last few days, and it’s time (again) to go through a necessary upgrade. Currently, Politicalbetting runs on a single server at 1and1.co.uk. On this machine there are two “programs” that run – one is the web server (apache2) and the other is…
Hillary has a plan; win the popular vote and persuade the super-delegates that she is not just the better candidate but has a moral claim to the nomination. Should the gap in pledged delegates be less than 100, then this might just work. The only problem she has is that the maths is hard; very hard. Excluding Florida, Hillary needs to pull back 410,000 votes between now and the final primary. (And she’ll be hard pressed to include Puerto Rico…
Surprisingly, the server migration appears to have gone through relatively painlessly. Hopefully the site should run a little more quickly now, especially under heavy load. Thanks for your patience, Robert