Random Passage

To the helpful person (heya, D.W.!) who suggested that maybe the partisan pavement pattern demonstrated of late is just the result of "the law of averages"...

The short answer is "noop".

The longer answers are below.

First, if the "law of averages" were truly at work, then, during the past four years, you would expect between 11/38ths and 12/38ths of any cohort of ten consecutive districts, or of any group of ten picked at random, to be opposition ones.

(Why 38? That's what you get when you subtract the one remote, roadless district of Torngat Mountains, and the nine central St. John's seats, which usually do not get dedicated, district-identifiable PRIP funding, or any at all. Eleven and twelve are how many of those 38 have, from time to time in the past four years, been held by opposition MHAs.)

In other words, you should expect, on an average, about three of the bottom ten districts, or three of the top ten, or three of any ten, would be opposition, and seven government.

Instead, in 2004, eight of the bottom ten were Liberals; nine of the top ten were governing Tories. Then-NDP held Labrador West was the only opposition district ever to rank among the top ten most-favoured districts under the PRIP program during the Williams era so far.

In 2005, seven of the ten bottom districts were held by Liberals; the entire top ten were government-held districts.

In 2006, five of the bottom ten were opposition-held (six, if you factor in the three opposition districts that got a whopping $840,000 between them, including the well-punished Bay of Islands district.) Again, the top ten were all good PC blue.

In 2007, all but three of the bottom ten were opposition-held; and, you guessed it, the Tories swept the top ten again.

In not one case does the "law of averages" make even the suggestion of an appearance.

Now, on the other hand, what if things truly were averaged out? What if there were no partisan pattern at all, that the amounts were distributed randomly among the districts without concern for partisan considerations?

You'd get something like the chart below.

The number of districts involved, the districts themselves, even the dollar amounts are exactly the same as in the real-world chart for the current year, 2007.

The only thing that has changed, is that the dollar values have been randomly re-assigned among the various districts. The dollar amounts themselves are the same as in the real world, they have just been re-distributed, in no particular, totally random, order, to other districts. If there were no partisan considerations at all factoring into PRIP allocations, the charts for each of the past umpteen years would look something like this:

Notice how the opposition districts are more evenly distributed, not all bunched together at the low end of the scale? And how only two of both the bottom ten districts are opposition, while only eight of the top ten are government?

There's your "law of averages".

Not here.