They were 1:300 just yesterday.
Yep. Then they were 1:64. Now they are 1:43.
Those odds are based on the margin of error on the orbital path, the center-line plus or minus a certain amount in two dimensions. The cross-section of the probable path is getting smaller as our measurements are becoming more precise, and the earth is still inside that path cross-section and a larger target as a percentage of that cross-section.
JPL and NASA have been given priority time on our deep-space sensor nets (including US Space Command assets if I am not mistaken) to develop a very precise orbital model of this rock. With this much effort being put into the measurement, eventually the orbital path cross-section will be smaller than the diameter of the Earth and we can say with a high degree of certainty whether or not it will hit the planet and roughly where if it does. We'll have a second chance to make fine-tuning corrections in 2013. Interesting stuff.
I would point out that based on the orbit of this rock, that it will be a regular recurring threat.