It would appear that there is definitely some good news to report out of Iraq. After sustaining the third highest month of casualties in May of 2007, according McClatchy the casualties have declined for three solid months since.
This is most definitely a good thing, that is if it is true.
I decided to take a look at the casualty count as tallied by the organization McClatchy cited as their source for this piece, icasualties.org. If you take a look at their direct numbers you will notice one distinct difference, the numbers do not match.
If we just compare the past four months we can see that something is whacky. Remember, this is the same site that McClatchy cites as the source for their graphic yet on the last two data points, May and August McClatchy shows 123 & 57 deaths respectively whereas icasualties.org shows 126 & 81.
There is no question that any way you look at it August had less casualties than May but the disparity is real. For the most part these numbers jive +/- 3 or so casualties until we hit the month of August when the differential drops to -24 casualties. Even if you use the filtering methods on the site to differentiate between hostile fire and non-combat related deaths the numbers don’t agree [although they moves much closer to the +/- 3 noted before].
This seriously begs the question what changed in the method of counting combat casualties in August? Does it have anything to do with the upcoming Patreus report on “the surge?” And lastly, what type of deaths did the Pentagon decide not to include in their official count to account for this change?
UPDATE: Upon further review of the icasualties.org website I came across a trend comparison that is worth noting. According to the keepers of that site,
A Proposed Benchmark for Evaluating Surge Success Based on the Fatality Trend Line (Blue in the Chart Below):
If the Fatality Trend Line Drops Below 10 for an Extended Period (At Least Ten Weeks; A Pattern Not Seen Since the 2003 Occupation), and
If it is Accompanied by a Parallel Reduction in Iraqi Deaths (to the early 2005 level of less than 100 Per Week, Down from 300-400 Now – See US vs Iraqi Fatalities tab),
Then the Surge will Have Achieved a Substantive Measure of Initial Success.
UPDATE II: Being the lazy man I am (not to mention the fact that my entire day AFTER the Labor Day event was spent being pulled away from here to do chores around the house) I asked Engram over at Back Talk to give me some help with the question of why there is a spike in non-combat related deaths from his expert statistical point of view. Of course the answer didn’t need an expert at all, only someone with some more time to research.
Good, so no nefarious fuzzy math at play.
Now there is still the pesky question as to how a three month increase in US casualties over this time last year [June-August 2006 = 169 deaths - June-August 2007 = 261 deaths], has somehow been interpreted as proving the success of the surge.
Others blogging this story: Captain’s Quarters, Jules Crittenden, Wake up America, Back Talk, The American Street, Sister Toldjah, Burkean Reflections, Group News Blog, Connecting.the.Dots, Blue Girl, Red State and Show Me Progress