Bell Curve The Law Talking Guy Raised by Republicans U.S. West
Well, he's kind of had it in for me ever since I accidentally ran over his dog. Actually, replace "accidentally" with "repeatedly," and replace "dog" with "son."

Saturday, December 17, 2005

The Graph Bush Doesn't Want You to See

Nobody doubts that our troops in Iraq are doing the best they can. But despite their best efforts, despite growing expertise in the field, and despite billions of dollars in expenditures, insurgent/terrorist violence against our forces is only getting worse.


Bush asks us to "stay the course," but this strategy is failing to provide security to our troops and to the Iraqi people. Representative Murtha has it right: the problem is neither lack of will nor resources, but rather that our occupation force has become an obstacle to its own success. Our troops have accomplished a great deal in Iraq, but we have long past the limit of what we can achieve by military means. We are getting no closer to final victory by continuing it, and there is no reason to believe this will change.

It is time to bring our soldiers home.



[All figures are from U.S. Department of Defense. Most recent reported deaths included are for November 2005; most recent reported wounded included are for October, 2005. Graph (c) Dr. Strangelove]

10 comments:

Anonymous said...

I think it is important for Democrats to emphasize that withdrawl is a means to achieve our stated goals (a stable, responsible government in Iraq) not an admission of defeat. Howard Dean has done a disservice to his party by making statements about the impossibility of victory. While his point may have merit, it is political stupidity. Murtha's statements are both more likely to be accepted by pivotal groups of voters and (in my opinion) a more accurate assessment of the situation.

It is clear to me (and was clear to me before this invasion started in the first place) that democracy cannot be imposed by military means. It is a product of political and economic conditions which cannot be established by force.  

// posted by Raised By Republicans

Anonymous said...

Dr. S - did you find or create this graph? It's incredible. This needs to be on CNN. 

// posted by LTG

Anonymous said...

Wow, it didn't occur to me that Dr. S. did that himself. If you did, fantastic! How did you fit the lines? Do you have a significance indicator of some sort to give us?  

// posted by Raised By Republicans

Anonymous said...

Looks like an Excel chart. Dr. S., is that a least-squares plot?

Even if the average CNN viewer doesn't know what that means, it's hard to deny the violence is accelerating, not slowing down. 

// posted by Chris

Dr. Strangelove said...

Chris: yes, it's an Excel plot I put together and added a least squares fit (linear regression) for each data series.

Anonymous said...

What about the obvious outliers? Is the relationship significant? 

// posted by Raised By Republicans

Dr. Strangelove said...

RxR: questions of significance and outliers are important measures for any linear regression analysis. In this particular case, however, I ask you to understand that I am not actually hypothesizing a linear relationship between casualties and time. It surely is a very complicated, non-linear relationship, to which a straight line would naturally be a poor fit. I am merely using a least-squares linear fit as a tool to illustrate what you can see with your eyes: the violence is not decreasing, and if anything it's getting worse. One can achieve better fits with polynomial regressions of higher order (I have done so) but these merely add nuances that do not alter the fundamental trend you can see with the simplest fit. One can also plot a moving average, which shows the same thing.

That being said, here are the statistics for the linear fit:
For dead, R^2 = 0.1336
For wounded, R^2 = 0.2151

The first spike or outlier (April 2004) corresponds to the first major U.S. attack on Fallujah and associated coordinated strikes by Muqtada al-Sadr's forces. The second outlier (November 2004) corresponds to the second major U.S. attack on Fallujah (Operation Phantom Fury) involving 10,000 U.S. soldiers. This is the one in which white phosphorous was used.

Dr. Strangelove said...

If you omit the two outliers mentioned above, the R^2 for dead jumps to 0.223 and R^2 squared for wounded jumps to 0.421.

Anonymous said...

So length of the conflict alone explains a quarter of the deaths per month and over 2/5 of the number of wounded per month. Very interesting. 

// posted by Raised By Republicans

Anonymous said...

I wonder how this compares to charts from other wars.  

// posted by LTG