On splits

Ken Boessenkool and Brian Topp offer their own take on vote splits in the recent federal election. And they're absolutely right to note that the Libs appear to have been the main beneficiary in last week's election: in fact, the full list of ridings where a party won 50% or more of the vote includes only two Lib seats (both in Newfoundland).

That said, though, I'm not sure that the observation speaks to much more than the inherent relationship between a party's baseline support and its top riding results.

By way of comparison, the same list from 2008 includes only seven NDP seats to a handful more Lib ridings, while the list of ridings with the lowest winning share included numerous NDP seats - suggesting that there isn't a huge structural difference between the two parties in how many vote splits they rely on when their party support is in a similar range. And it stands to reason that as a party's national support level drops, it will have a more difficult time compiling massive margins of victory in any given seat - not to mention little incentive to fight for them if it's possible to win a plurality with less.