April 20th 2013: How high can the NDP go?

If my first campaign post was about the chances of a BC Liberals victory on May 14th (spoiler: right now, the odds are very close, if not exactly, zero), my second is about how manys seats the BC NDP can expect or hope for.

Looking at the most recent polls (from Ekos, Angus-Reid or Ipsos), the NDP is clearly ahead in term of voting intentions. If the election was held today, the only possible outcome would be a BC NDP government. However, would it necessarily be a majority one? Could the NDP sweep the province the same way the Liberals did in 2001? The short answers are, respectively, yes and yes.

For the majority, I'm using the numbers from Angus-Reid (Ekos is a little bit old and Ipsos doesn't provide enough details to the public. In any case, all three recent polls agree. On top of that, remember that the 1000 simulations take care of the uncertainty regarding polls. In other words, if the NDP is at 45%, there will be scenarios with the party is as low as 38% or as high as 52% in my model). So, using these numbers, the lowest possible number of seats for Adrian Dix's party is currently 47. Remember (or discover if you're new to this site), it means that even in the worst case scenario where the NDP would perform significantly lower than what the polls predict (and thus the Liberals or another party would perform much better than expected) AND be inefficient in the repartition of its votes, this party could not get less than 47 seats. This is good enough for a (small) majority. So, as of right now, not only is the BC NDP sure to win, this party is even sure to get a majority! In case you're curious, the best result for the Liberals would be 35 seats. For the Green and the Conservatives, 1 and 9 respectively.

Then we have the second question: what if the NDP actually performs even better than what the polls predict? After all, there is no reason (a priori) to believe that the polls could be biased or skewed in one direction only. Could the BC NDP sweep the province? BC isn't foreign to this situation. In 2001 the Liberals won 77 of the (then) 79 seats of the legislative assembly. They achieved this by getting "only" 57.6% of the vote. The electoral system and some bad luck (or inefficiency) for the NDP did the rest. So, could we see a complete reversal of the 2001 outcome?

Actually, it does seem to be possible. First of all, it would require an important underestimation of the popular vote share of the NDP. In my 1000 simulations, the NDP best case is to get 53% of the vote. This is going even beyond the normal margins of error (but not completely impossible, especially if the latest Ipsos poll was actually right). In terms of seats, the highest number Adrian Dix could get is 80! 80 out of 85 is pretty similar to 77 out of 79.

Please notice the implications of these numbers. As of right now, based on the polls and the past elections results, the BC NDP is more likely to win 80 seats than to lose the election! This is a pretty comfy situation Adrian Dix is in right now! In term of suspense, this is definitely not the best election.

My next article will look at the impact the Conservatives could have. In particular, how high can they get and who do they hurt.