The AP Strikes Again

We had to check our calendar when this wire service headline popped up on our computer screen last night:

U.S. troop deaths up after drop in July

Barely one week into the new month, AP writer Sally Buzbee seems anxious to tamp down encouraging reports about last month's drop in U.S. casualties, suggesting that August will be much worse.

"The spate of recent U.S. deaths — 19 so far in August — seems certain to intensify the debate over U.S. progress to calm Iraq and gain ground against militants ahead of a key September report to Congress."

U.S. deaths had dropped slightly in July to 79 — the lowest monthly tally since 70 were killed in November. Before July, more than 100 American troops died each month in the April-to-June period as the U.S. military struck out at insurgents on dangerous streets and cities across Iraq.

But U.S. commanders say rogue Shiite militias have stepped into the gap left as Sunni insurgents have been pushed back, and are now responsible for most attacks on Americans in Baghdad and surrounding districts. Such a trend would elevate fears that Iraqi forces are not yet able to maintain security even when insurgents are beaten back. Large numbers of Iraqi police are believed also to hold allegiances to Shiite militia groups

A bit later in her report, we discover the basis for Ms. Buzbee's assessment, a comparison of "average" daily troop deaths in Iraq:

Although U.S. military deaths dropped in July, a wider accounting of the figures offered a sobering overview.

The daily average for U.S. troop deaths in July was at least 2.55 — higher than the daily averages of 2.25 last year, 2.32 in 2005 and 2.33 in 2004. So far in August, the daily average is 2.71.

Never mind that American casualties remain astonishingly low, in comparison to past conflicts. Never mind that the security situation in Iraq has improved dramatically in recent months--a development noted by AP military reporter Robert Burns last week. And never mind that Ms. Buzbee's barometer of operational success--the daily casualty average --would judge all military campaigns as ghastly failures. With the Petraeus Report looming on the horizon, the AP correspondent feels compelled to temper good news with her analysis that our death toll is on the rise (again).

But are the August numbers higher than July?. Using data from the Iwar database, we compared U.S. deaths for the period 1-7 July and 1-7 August 2007, and discovered the number of fatalities in the first week of August was actually lower than during the same period last month. Here's the day-by-day breakdown of the two periods:

1-7 July 2007

1 July: 7
2 July: 1
3 July: 0
4 July: 3
5 July: 7
6 July: 9
7 July: 1

Total for the week: 28

1-7 August 2007

1 August: 1
2 August: 5
3 August: 0
4 August: 5
5 August: 4
6 August: 5
7 August: 1

Total for the week: 21

The decrease is also reflected in Ms. Buzbee's favorite benchmark, the daily average. Calculated in terms of total deaths per day (3.0) and combat fatalities per day (2.85) , the averages for the first week of August are lower than during the same period last month.

But you won't find that in the AP report, which focuses (instead) on comparisons with years past, providing the statistical basis for their claimed "increase." Still, even that analysis is slightly flawed; the reported "jump" is actually lower than many analysts predicted, given the on-going troop surge and the increase in combat operations.

However, we must caution against reading too much into the AP assessment--or our own comparison, for that matter. Obviously, the death of even a single military member is too many, and we look forward to the day when we won't be writing about casualties in Iraq.

Beyond that, any "analysis" of this sort is militarily irrelevant, as any student of history knows. For the record, an average of 239 Marines died during each day of the month-long battle of Iwo Jima (1945), and we lost 358 troops a day during the three-month Normandy campaign in 1944. Of course, both battles ended in decisive Allied victories.

We may yet see the a similar outcome in Iraq. But we're guessing that the AP's coverage of that event would take a very different slant, particularly if Ms. Buzbee is doing the reporting.