Forfty percent of all people know that....
...[one] can come up with statistics to prove anything. (quoth my hero, Homer Simpson).
I'm an avid consumer of statistics on the situation in Iraq. Partly this is because I find it difficult to get a handle on what's happening over there with the lack of good information. People like Michael Yon and Bill Roggio help, but don't tell the whole picture. One good source of statistics is the Brookings Institute Iraq Index.
Of course, the problem with statistics is how you interpret them. It seems possible for different people to interpret the same set of statistics to come to different conclusions (possibly supporting their preconceived notions).
Personally, my take on the situation is positive. Not all the statistics are positive, but some are, and I mix in my reading about events over there to make the full picture.
Currently, the best trends are those which show that Iraqi civilian and soldier/police deaths are dropping off rapidly, as are car bombs and such. The bad news is that Coalition deaths and injuries don't seem to be dropping off much, and reconstruction of infrastructure like electricity is not advancing as well as we would like. However, I believe there is a good reason for the trend in coalition casualties, which supports my belief that progress is being made, slowly but steadily.
Basically, what I see happening is that the fighting from the point of view of the coalition and IAF is shifting from being a defensive battle to being a more and more offensive one. A lot of deaths and injuries are currently occurring in the Anbar province - the sector in which the terrorists and insurgents are currently being targeted relentlessly. Troops fighting offensively are naturally at a disadvantage and can expect to be hurt more often.
In short, when areas become safer and therefore coalition casualties drop, that's an indication that it's time to move some troops into more dangerous areas, as they are no longer needed where they are. This creates a rise in casualties. Essentially, if the battles are being fought intelligently and troops are being distributed properly, coalition casualties should be essentially flat. And for all statistical intents and purposes, they are.
I don't believe the "oil spot strategy" is 100% correct, but I do believe that it is in part what is happening here. As areas are cleared of enemies, the number of peaceful areas (or relatively peaceful, depending) in Iraq goes up. The remaining enemies are squeezed into a smaller space. Meanwhile, those areas which are not being offensively targeted are instead being patrolled enough to at least prevent the situation from getting worse, especially with the influx of fighters from other regions. Ultimately, I believe their last stand will be Baghdad. Baghdad has been the most consistently violent place in Iraq, despite the large security force presence.
There are several obvious reasons for this - it's where the journalists are, it has the highest concentration of targets, it has the most mixed and varied population, it's easier to hide in a big city and one has to be careful mounting large scale military operations in such a densely populated area.
So, as other provinces become more normal, Baghdad will probably stay bad. Operations will eliminate some of the rotten eggs there, but others forced out of outlying areas will probably move in to take their place. The problem for them will be that their lines of supply will be cut. No new foreign fighters, no new weapons... it will be a sort of siege-from-within which will force them into submission. Of course, weapons are not hard to come by in Iraq - it was estimated they had 412 012 tons before the war and most of that fell into the wrong hands. But caches are being constantly discovered and destroyed and without outside supply they will eventually run out.
I think we really are seeing a move in this direction. Some people say "well, if we're winning, we should be able to pull our troops out now and let the Iraqis take over, right?" Well, it's not quite that easy. What I see happening over the next 6-12 months is that as more and more Iraqi police and soldiers take over defensive tasks, coalition troops will go more and more on the offensive. We're already seeing it now, it will happen even more soon. When this happens to the extent that all the offensives which are desired are already in progress, coalition troops will start coming home. I suspect, though, that probably about 25% of the number there now (say, 30-40 thousand) will be staying for a while. There are still a lot of internal and external threats that will need to be dealt with. We shall see - to a large extent it will be up to the Iraqi government to decide. I don't think the 25% level will be reached until at least 2-3 years from now.
I suspect if we cut and run now, two things would happen. Firstly, it would so embolden the terrorists that they would go on a major offensive. Secondly, the Iraqi security forces would be forced on the back foot, because we would no longer be placing pressure on the terrorists and insurgents by going after them. They will be forced to become purely defensive, and instead of the pockets of unrest shrinking, they will grow. That will probably continue until much of the country (probably minus the Kurds) are engaged in a civil war. We don't want that. We don't have to stick around much longer before we can effectively break the back of the terrorist network and allow the Iraqi forces to gracefully transition into a defensive posture, one where the situation is improving rather than degrading.
Of course, we hope the December elections will also help calm things down. It should provide Iraq with a fully representative government. There is evidence that terrorist and insurgent groups are splintering and disagreeing too. It's likely that with the representative government, if more reconstruction progress can be made (and despite some problems, progress is being made) a lot of people will realize they're better off and things are only going to get better over time, and some will stop their bickering and get back on with their lives.
So, when exactly will troops start pulling out? I'd say it will be when there are obvious and permanent drops in coalition casualties. This will be indicative of the pace of operations dropping. To pull troops back now would be to reduce the pace of the offensives, which would only draw them out, and require a longer presence. Why would we want that? Better to stay there until some of the coalition troops are just not needed any more, so they won't be making the job of their buddies who stay behind harder when they leave.
At least, that's my hope. I think we have passed the point of inflexion recently and we will see the rate of good news climb from now on. It is in everyone's interest (other than terrorists) to do this right.
Then we only have to worry about Syria, Iran, North Korea.... *sigh*