There are two things. I'll just make a general comment about the data series and what we're actually measuring here.
These are compilations by the International Energy Agency of data from national statistical sources. All those data series will have measurement errors around them, just as polling numbers have errors around them. So it is not the price of a particular transaction; it is confidence in the polling around it. Moreover, there may be differences in the procedures that statistical agencies use. What day of the week do we measure on? Do we measure every day of the month, or are we taking only the end of the month? You would expect, even in jurisdictions where prices are in fact identical, that the series would not be precisely identical. You have to ask whether they are statistically different.