Still, it's worth noting that coalition casualties continue to decline sharply.
There were 13 coalition fatalities in Afghanistan in August, following on 14 in July. That compares with 52 in August 2012 and 46 in July 2012; it's the lowest combined total for July and August since 2005 (and the lowest August since 2004). For the year, the total is 122 coalition fatalities (96 US fatalities), down from 320 through the end of August last year. It may also be worth noting that fatalities for July and August this year were only half as many as in May and June, which hasn't been the pattern over the years and may indicate that the trend is still very much continuing.
I don't really have much to add; I mostly just think it's an important and massively underreported set of facts. I do continue to believe it provides evidence that Barack Obama is unusually quagmire-averse. That doesn't mean one should support a Syria strike, of course; one can certainly believe that the 2009 Obama surge in Afghanistan was a mistake, regardless of his willingness to abandon it after a while (although I don't know if that's the case).
On the other hand, there could easily be another 100 or more coalition fatalities between now and the end of 2014, and still more after that, depending on how many US troops wind up remaining. It's also fair to ask exactly what the point of that is. Yes, some of it is baked in (withdrawal is complicated and can't be done overnight), but some is certainly a cost of continuing operations. And of course there's always the question of what policy gains (if any) were abandoned when Obama decided to end it. Lists of fatalities only indicate (one of the) costs of a policy, not whether it is wise. One more thing: these stats are fatalities only. Full casualties statistics would capture even more of the costs.
At any rate, I do think the information about this is important and undercovered.