The Opinionated Bastard

Interesting and relevant analysis. I concur with the change in posture for confronting the terrorists. Increased Sunni participation will ultimately defang these predators.

I would like to see some pure statistical analysis on troop strength versus violence (bet there is a surprising correlation!) MiniTab is a great tool for this.

Keeping in line with numbers from Iraq to round off 2005, how about a comparison of the death rate of Coalition soldiers in Iraq versus Afghanistan for the last year?

The answer may surprise you. Especially if you are of the opinion that Iraq is a “scandal” compared to a “justified” Afghanistan…

excellent post. thanks for the facts. but your conclusion is wrong:

“The only way throughout history to destroy a radical political organization has been to co-opt it into the government.”

this is just wrong.

the confederacy was not coopted poltically.

neither were many other insurgencies or rebellions.

the black panthers, red brigades, and che’s foreign expeditionaries were all defeated.

king hussein defeated arafat and expelled him.

radical political groups that take up arms MUST NEARLY ALWAYS be defeated with arms.

These analyses are a determined effort to miss the forest for the trees. If you simply take six-month averages, then nothing much has changed in more than two years. The occupation of Iraq is Sisyphean and eventually the US will just get tired of it. If there is a partial pullout in 2006, American fatigue will be a main reason.

Another interesting metric is Iraqi oil production. Every OPEC country increased production from 2004 to 2005 in response to record prices, except for Iraq and Indonesia. Indonesia appears to be running out of oil. In Iraq, oil production fell because of security problems.

Seixon is right about one thing. It is not too late for the Bush Administration to snatch defeat from the jaws of victory in Afghanistan. Unlike Iraq, the Afghanistan operation is aligned with American interests, but Iraq takes the lion’s share of the resources.

Hmmmmmm, yesterday 11 US soldiers were killed and 130 Iraqis in suicide bombings and IED attacks. I find it very tacky that you utilize “statistical analysis” to make claims about the state of the Iraq war. I think it is reliance on “cold analysis” such as this that deluded this nation’s technocrats into thinking that IRaq would last “6 days, 6 weeks, maybe 6 months, but i doubt it.” (Rumsfeld, 2003) I bet that “statistical models” in the pentagon also supported that Iraq could be pacified with 150k soldiers. The point is, the importance isnt in what the data shows, but what it doesnt, and treating the Iraq war like a financial chart is disingenuous.You should undertake some interesting analysis and predict how many US soldiers pro-war people will get killed in Iraq, and then predict how long the US imposed government will last once the US is forced to leave, oh ya, for iraqi freedom and liberty the US doesnt plan to ever leave. Good god…

ernie

I got a prediction ernie.

We will leave Iraq within 45 years.

I mean we kept our Soldiers in Germany for 60 years after WW2 for their freedom and liberty.

Wait, we still have troops in Germany?

Why?

Ernie, after some analysis, I’ve come to the conclusion that you are an innumerate idiot.

I think it’s a mistake to make month-to-month comparisons. A better way to gauge progress is to compare 2004 to 2005. Iraq Body Count says they track civilian deaths, but if you look through their database it becomes clear that they track all Iraqi deaths (Iraqi police and army included). In 2004, more than 10,000 Iraqis were killed. In 2005, about 7500 were killed (or 1000 more than that if you count the 1000 killed in the August 2005 panic stampede).

Now go to icasualties.org and look at the number of American troops wounded. In 2004, it was 7989. In 2005, it was 5557. Look at coalition deaths by hostile fire. In 2004, it was about 6% higher than 2005.

Next, look at contractors killed. Foreign contractors killed: the number of foreign contractors in Iraq just keeps going up, but the number killed keeps going down, from 182 in 2004 to 94 in 2005. Again, hardly a sign of an insurgency gaining in strength.

Attacks on gas pipelines: in 2004, it was 325. In 2005, only 89. Again, this is not a sign of increasing strength among the insurgents. Instead, it is a sign of decreasing strength.

Again, don’t look month-to-month. Look year-to-year. The insurgency is clearly weakening, just not as fast as we’d like.

Ernie actually made a reasonable comment for once.

The 125 civilians killed attack is significant, but the daily numbers get even more noisy then the monthly numbers. It takes longer to plan an attack that kills 125 civilians. So what remains to be seen is if an attack like this can be followed up with more attacks on the same scale, or if its a one-off.

So yeah, there was a bad attack yesterday. Does it mean anything? I dunno, ask me in a month!

Hey Ernie,

Don’t soil your panties with grief. I guarantee you today, and tomorrow as in decades of yesteryear’s, that the members of the US military couldn’t care less about your anxieties, concerns, sleeplessness, upsets and other varied neurosis-es.

Why don’t you go save an endangered shew, paint a window sill at a urban youth center or meditate for world consciousness? Leave the imperferct, hard, killingly dangerous work to your betters.

ernie,

I gather from your garbled post that you are offended by analysis of warfare in numerical form.

This is something that you are going to have to learn to deal with. You may be used to a small environment that can be managed via intuition. But everything that is big, everything that is a system is analyzed numerically. Take unemployment, you may say that it denies the humanity of each individual unemployed person to make him or her a statistic but one person being unemployed doesn’t tell us much about the economy as a whole.

Someday you may see the big picture.

If my post is an indication of anything, it is that Afghanistan got worse during 2005, and Iraq got better. You wouldn’t know that listening to Howard Dean, though, intent on sending 20,000 of our soldiers from Iraq to Afghanistan, claiming that we are “more welcome” there. Sorry Doc, the numbers on the ground don’t bear that out.

“but the daily numbers get even more noisy then the monthly numbers”

I’d suggest trend-analysis, actually. I started putting together something about 6 months ago, but the data was sparse. My next attempt was going to involve the technique embodied in this software. I was also having some success fitting an ARIMA model to the casualty figures.

The only way throughout history to destroy a radical political organization has been to co-opt it into the government.

As reliapundit above pointed out, that’s not quite true. In particular the closest example we have is in Jordan where a terrorist organization was eliminated root and branch. We just don’t want to adopt their tactics. Yet.

“The only way throughout history to destroy a radical political organization has been to co-opt it into the government.”

Yeh, I have to take issue with this as well. If you reach even further back into history it becomes even less true. The Sparticans and the Zealots come immediately to mind.

Your manipulations of the graphs and data are wonderful illustrations of the techniques described in the book “How to Lie with Statistics”. This is a short and very charming supplement to Freshman statistics classes. Cutting off at a downturn knowing that an upturn is ahead is on page ten.

I’m quite aware that predicting the future from the past is iffy. It’s actually my job to do so: http://www.marketocracy.com .

So yeah, LIKE I SAID, what happens post election is going to be key. (unless we can arrange for Iraq to have elections EVERY month!)

Seixon: Certainly sound bites from Howard Dean are beside the point. The man isn’t in office and he isn’t even running for office. The central question is whether the United States is winning the war in Iraq. I am convinced that it isn’t. In fact, I think that the war in Iraq is too illogical for that to even be possible. The invasion was strategically retrograde, according to William Odom and others.

But even if it is beside the point, it’s not clear that Dean is wrong. Should you measure local hostility by the number of Americans killed, or by the fraction of Americans killed? The truth is surely somewhere in between. If there are 10 Americans in some country and they are killed in one ambush, that doesn’t look worse than if 100,000 Americans are in a country and 2,000 of them are killed in many attacks.

Things are also not clearly getting better in Iraq. Some statistics suggest improvement and some down. For example, oil production is down compared to 2004. Slightly fewer Americans were killed, but that could be because they got better at avoiding attacks, rather than preventing them.

But I agree that attacks have clearly gotten worse in Afghanistan. That part is beyond dispute.

Greg K:

This report - http://www.fas.org/sgp/crs/mideast/RS21626.pdf - is enlightening about oil production in 2004 and 2005. To sum up, the northern fields, in Kurdish territory, move oil via a pipeline to the Mediterranean. This pipeline was sabotaged in early 2003(?) (I’m skimming) and partly repaired in 2004, which enabled some northern-produced oil to be moved. But the pipeline was sabotaged again in late 2004(?)(-ish), and remained offline throughout 2005 because - here I’m reading between some lines but I think my inferences are pretty straightforward - the political changes taking place in Iraq included lots of negotiations about how oil revenues were going to be divvied up, particularly with regard to the Kurds, who had little incentive to protect the pipeline unless they had assurance that they’d see proportionate benefits from its use. 2005 was a mighty busy year in the political life of Iraq… Considering that Iraq continued to produce and export oil at levels not a lot lower than pre-war levels (more on this point in the next para), and that refining capacity in Iraq still significantly lags domestic requirements, forcing imports of refined product, thanks to Saddam’s complete disregard for his nation’s infrastructure, maybe the interim Iraqi government had other priorities than boosting production while crude was being reinjected into some fields because it couldn’t be moved.

[Deep breath] As to pre-war production, I understand (and will try to relocate the source) that Saddam produced the holy heck out of Iraqi fields prior to the 2003 invasion, just as he did prior to his invasion of Kuwait in GWI. I read that his overproduction may have permanently damaged some fields. “Reservoir management” was not at the top of his list - stockpiling may have been.

See this chart - http://www.wtrg.com/oil_graphs/PAPRPIQ.gif - of Iraqi oil production since the ’70s, and compare it to this one - http://www.utpb.edu/PBDPL/Statistics/Images/PBAnnOilProdTX.jpg - for Permian Basin (Texas) production between approximately 1940 and 2000. The Texan production chart smooths out the peaks and valleys, covering as it does a much longer time period, but it’s the profile of a mature field, with economics sort of subsumed in the data. The Iraqi chart, if we were to remove the political aspects from it (the mega-drops in production in 1981 - the Iran-Iraq war - and 1991 - GWI), would have the same shape. With no new wells coming on-line, we may well be just looking at a mature “field” (that’s still much more productive than American fields).

More: This article - http://www.usatoday.com/money/industries/energy/2005-10-10-iraq-oil-usat_x.htm - points to several problems with current and recent Iraqi oil production, almost all of which would have affected Saddam just as much if he’d remained in power: equipment outdated and ill- or not maintained, reinjection of crude fractions, no exploration. Only “insurgent attacks” and “looting” are problems unique to post-2003 production (and looting was a problem only in the immediate aftermath of the invasion).

Thanks for the analysis. I wonder how these news trends would compare to headline trends. I’m not sure how to quantize the severity of TV and newspaper coverage, but it would be interesting to chart public programming against the actual facts.

Greg said: “Should you measure local hostility by the number of Americans killed, or by the fraction of Americans killed? The truth is surely somewhere in between.”

What? The only reason you would want to just count numbers instead of proportionality would be to, oh I don’t know, pull wool over someone’s eyes.

The easiest way to fool people with statistics is by not showing them in perspective. Citing raw numbers instead of how they are proportional to the things you are comparing is doing exactly that.

Let’s think of the French riots for a minute. At the height of it, they were torching around 1,000 cars a day. To begin with, it was only around 100-200. At that point, people we making a big deal about it. What if they knew that it is common for 50-80 cars to be torched in France every day? Well, then that 100-200 sure doesn’t look all that dramatic anymore.

Leave out the perspective, leave out the truth.

Jamie: You’re giving a complicated explanation for something simple: Iraq’s oil production in 2005 was thwarted by insurgent attacks and political uncertainty. Oil is the core of Iraq’s economy, but they just don’t have it together because of the civil war. Wolfowitz’s theory that oil would pay for reconstruction turned out to be wrong. On the contrary, Iraq has a chicken and egg problem: They can’t restore their economy without oil, they can’t end the civil war without restoring the economy, and they can’t restore oil production because of the civil war.

No oil expert, certainly not Kumins, who you cited, supports your speculation that Iraq is simply running out of oil as Texas is. All analyses say that Iraq has many undeveloped oil fields, and that even its existing wells have 3 mbpd capacity. There are reports that Saddam Hussein’s regime damaged some of the oil fields, but at the most he busted a few kegs in an enormous cellar. For that matter, if your speculation were correct, it would make Wolfowitz even more wrong.

I find it a little hard to accept your analyses, due to the fact that these are middle-number magnitudes, hardly numerically large enough for assurances of statistical trend; plus the fact that the insurgents have always played the political game around election-time, which you are hoping to correct by an rough rule of thumb.

Of your first chart you say “there’s still a slight growth in IED deaths, but all the other numbers seem to be trending down.” Surely, without another four to six months of future data, the best one could say is all the other numbers appear to be trending rather steady. As with the chart on Iraqi police deaths.

Without more months on the “Killed and Wounded in Multiple Fatality Bombings,” I would say the trend is distinctly upwards at a 10-degree angle.

On the civilian deaths chart: are they including estimated airstrike casualties? It would appear that much of the air campaign is being hidden from us.

I think Mr Murtha may have it exactly right, unfortunately: in one year insurgent attacks have quadrupled, electricity and oil production are down, and the US doesn’t have much more force to send. Indeed, the Administration did not ask for more funds for rebuilding. It appears the Administration is effecting a phased withdrawal and reorganization of forces, much as Mr Murtha, and through him, the Pentagon, have demanded.

GM, among others. What i am saying is that statistical models are only as good as your assumptions, and if your assumptions are ideologically charged and you are subject to group think, then whatever statistical analysis associated with your views will be limited to ideology and group think.

Memory test, think about the lead up to the war (well better described as a voluntary invasion against a very weak country..). There were no WMDs, invading Iraq has increased terror, the US has no money to pay for this (China just announced they are tired of it, given the FEDs implication that interests rate hikes will stop. Things are changing fast, it is no longer patriotic to support illegal aggression and it is obvious this “war” is not making the US any more safe (whatever that means).

In short a really bad idea supported by people blindly and obediently to their conception of the US role in the world….

America, with the group think that runs amok among our hyper militarist culture…has truly stepped in a big pile of shit.

Suppor tyour troops, demand their return and a Bush impeachment.

ernie

Except the cynical part of Mr. Murtha was that the Adminstration had already begun that when he started demanding that. Its an old poltical game: Hold a press conference to demand what your opponents are already doing.

see: my scoop