Casualties in Iraq for the Month of September

I am tracking casualties from two sources now, Iraq Coalition Casualty Count (ICCC), which is my usual source, and Iraq Body Count (IBC), which I am attending to more closely now that I realize that their numbers agree with those presented by General Petraeus. These are both anti-war sites that track casualties based on media reports, so you can be sure that the numbers are not biased to make matters look favorable for George Bush. And matters do look favorable right now.

ICCC updates their figures frequently, so I have their September numbers to share with you today. IBC is slower, so we won't have their September figures for a while yet. ICCC is fast, but they are also less thorough than IBC. Despite their differences, the two sites identify the same trends.

According to the ICCC database, casualties in the month of September were astonishingly low. Their database has been updated through September 29 as of this moment, but there are no stories of mass-casualty attacks from Iraq as September 30 draws to a close in Baghdad, so this is pretty close to the final story:


Various aberrations have been removed from their numbers (as I have explained countless times, such as here). The light blue bars show the months during which the troop buildup unfolded, and the dark blue bars show the months during which the surge has been operational. If anyone had any doubts at all about the effectiveness of the troop surge, those doubts should vanish now.

People sometimes object to that trend line that I've drawn through the data. Another way to show a trend is to using a running average. Here are the same casualty figures, except that each bar now represents the average of that month's casualties as well as the casualties from the previous two months:


This method is slow to believe a trend. If casualties spike in the upward or downward direction in a particular month, the 3-month average value will not spike up or down to the same degree. That is, the trend has to continue for a while before it will be fully believed, so to speak. Do you notice a trend of late?

In September, violence was down in every category, especially deaths due to al Qaeda's suicide bombers (who are attempting to incite a civil war) and to Shiite death squads in Baghdad (who entered the battle in response to al Qaeda's relentless provocations). Here is the Shiite death-squad story:


Like the overall casualty story, you can see that these killings were out of control last fall, but they dropped substantially when the US troop buildup occurred in February. I now realize that things were even worse than that last fall (according to the IBC numbers) and the drop in February was not as large as it seems in this figure. In the Fall, ICCC did not regularly add in deaths from the Baghdad morgue, but IBC did that. Also, in February, ICCC did not include daily body counts from Baghdad for every day of the month (at least 10 days seem to be missing), which is one reason why the drop associated with the beginning of the troop buildup in February looks so large. I'll show you the story according to the more thorough IBC pretty soon (the trends are the same, but the specific values differ). In more recent months, IBC and ICCC have come into much closer agreement, in part because ICCC now includes daily body counts for every day of the month. Despite the more careful tracking, there were fewer than 300 deaths attributable to Shiite death squads documented in September. Thus, one of the main sources of civilian casualties in Iraq is increasingly under control.

As of 8:00 P.M. September 30 (Baghdad time), documented deaths from suicide bombers in the month of September come to a mere 75. Last month, it was over 600, with 500 coming from that one strategically useless attack against the widely despised Yazidi sect. Another 120 people were killed this month by car bombs, most of which are probably also attributable to al Qaeda in their relentless efforts to incite civil war. These deaths will presumably bounce around from month to month, depending on whether a bomber happens to get lucky in a crowded place full of defenseless civilians, but the trend would appear to be down. This fits with many stories suggesting that al Qaeda is on the run, but only time will tell if it is a permanent trend.

What about US military casualties? Here is that story:


As you can see, despite the troop surge, casualties are way down. All of this has happened despite the fact that Ramadan began on September 13, and despite the fact that al Qaeda threatened an offensive during that time.

I had expected the news today to report the astonishingly low number of civilian and military casualties for the month of September. After all, there is an on-going debate about whether or not General Petraeus was betraying his nation by deliberately lying about the improving casualty situation in Iraq. Casualties from the month of September are directly relevant to that debate. In fact, they would appear to settle the issue. Things might get worse down the line, and al Qaeda will certainly manage to sneak in more mass-casualty attacks before this is all over, but as of right now the trend in violence is very definitely in the downward direction (just as General Petraeus said was the case).

But what do people read when they pick up the paper today? This:

Petraeus admits to rise in Iraq violence

The top U.S. commander, back from his trip to Washington, says Sunni Arab militants have carried out a 'Ramadan surge.' But he notes that the level of attacks remains lower than a year ago.

By Alexandra Zavis, Los Angeles Times Staff Writer
9:57 AM PDT, September 29, 2007

BAGHDAD -- Gen. David H. Petraeus, the top U.S. commander in Iraq, acknowledged today that violence had increased since Sunni Arab militants declared an offensive during the Muslim holy month of Ramadan.
...
Militants with the extremist group Al Qaeda in Iraq have launched a string of deadly bombings in recent weeks, including one this week that killed at least 24 people at a reconciliation meeting between Sunni and Shiite tribal, religious, political and security leaders in the Diyala provincial capital, Baqubah.

At least 11 Iraqis were killed in bomb blasts, mortar and gunfire today. The U.S. military also announced the deaths of two soldiers in small-arms fire, one during combat operations in a southern section of Baghdad and the other in Diyala.

That's just pathetic. ICCC includes those 24 deaths in the 75 deaths for the month that are attributable to al Qaeda's suicide bombers. Would anyone reading this story know that casualties are way down in Iraq? Or would they instead assume that violence continues apace and that our lying general was forced to admit it? Well, as someone once said, you go to war with the media you have, not the media you wish you had...

UPDATE: OK, CNN gets some points for their willingness to cover the issue:

September U.S. death toll in Iraq lowest in 13 months

BAGHDAD, Iraq (CNN) -- The U.S. monthly death toll in Iraq has dropped to its lowest level in more than a year, although the figure is "still too high," the military said Sunday.

September's preliminary death toll is at least 62, according to military reports, the lowest number since August 2006 when 65 American troops were killed in the war.
...
As for Iraqi civilian deaths, authorities have reported that 296 bodies were found dumped in Baghdad this month -- the lowest monthly total this year. In August, there were 428 bodies found.

Those body counts are close to the ones you get from ICCC (which I have at 255 and 414).