The biggest change has come in Ontario, where the 13-point Tory lead observed in December has become a statistical tie (Conservatives 38%, Liberals 36%).
National Voting Intention: Con. 34%, Lib. 28%, NDP 17%, BQ 11%, Grn. 8%
Again, we see how quickly Ontario can change. There is really nothing substantial to explain such a large shift, but it just speaks to the superficial volatility at play, voters can and will change on a whim. In reality, you have 20-30% of the electorate on the market come the campaign, a subset that will have the biggest impact on the ultimate results.
In addition, lots of talk about Conservatives targeting ridings. Although that hope may very well be true, it is equally true that as many are available for Liberal consideration. In this poll for example, the Liberals would likely gain seats at the expense of the government, relative to the 2008 results. In other words, assaults can be replaced by erosions in dramatic fashion, that's the nature of Ontario at the moment. In fact, I suspect more ebbs and flows come the campaign, with the result being more about timing than conviction.