A Comment in Response to a Comment...
Wherein I calculate Russian losses in the Ukraine conflict...
A commentator to my last article mentioned that he thought Russian Defense Minister Shoigu had admitted to Russia having 40,000 killed in action (KIA) last fall.
A Google search failed to turn this up. It did turn up western media reports based on a report by Russian opposition media outlets MediaZone and Meduza claiming such a figure.
I decided to put down my methodology for calculating Russian losses. I haven’t bothered publishing this before because frankly the disparity between Russian and Ukraine losses is so huge as to make the issue irrelevant. It is also irrelevant because as I am on record (as Andrei Martyanov says every 2 sentences) that it is a physical impossibility for Russia to lose this conflict (short of nuclear weapons being deployed).
Here is the response I made to the commentator:
Shoigu never said 40,000. That was a calculation done by Russian opposition media, i.e., it's likely bullshit.
After the first three or four weeks of the war in March, 2022, on March 25, Russia claimed 1,351 KIA. That was when they were getting ambushed and other problems in their initial rapid advance. That was I believe mostly regular army.
UPDATE TO MY COMMENT: That article also established that Russian wounded in action (WIA) totaled 3,825, which almost exactly tracks the usual 3:1 ratio used to estimate the ratio of KIA to WIA.
Since then they've been conducting offensive operations using the Donbass militias, the Volunteer Battalions, the Rosqvardia (which includes the Chechens) which are NOT regular army, and Wagner. So most of the ground fighting has been done by units that are not regular army, supported by Russian special forces, and with Russian regular army providing armor, artillery and air support as well as logistical support.
So the rate of Russian regular army losses has been minimal since they stopped their initial advance, mostly the result of hits on rear areas, artillery units and armor with probably some regular and special forces units casualties.
On September 21, 2022, Shoigu said 5,937 Russian soldiers had been killed since the start of the conflict. He also said almost 90 percent of the Russian wounded had been returned to duty.
My own estimate is based on the initial March report. I assume 1,300 Russian regular army killed per month. This closely tracks with Shoigu's statements.
In March, 2023, the BBC attempted an estimate by analyzing open source Russian death reports and came up with a figure of 16,071. Divided by 13 months, this comes to 1,236 per month - which almost exactly tracks my estimate.
Further the BBC article says:
"Throughout 2022, Russian sources typically reported about 250–300 deaths each week, doubling in January and continuing to grow again in February.
This also tracks.
As the intensity of the conflict has intensified since last fall when Ukraine conducted its limited offensives and since Bakhmut (again, Wagner is not counted as they are not regular army, but reportedly they lost some 20,000 troops) and now with the Ukraine summer offensive, we can expect the Russian toll to increase somewhat.
As for the KIA:WIA ratio, we have no way of telling for sure. However in June, after a three-day battle, Russia revealed that it lost 71 KIA and 210 WIA. This almost exactly tracks the usual 3:1 ratio most analysts use to estimate this ratio. On the Ukraine side, I've been using 1.67 which is a ratio established by the Ukraine International Institute of Sociology poll.
UPDATE TO COMMENT: See my above UPDATE which also indicates that a 3:1 ratio is appropriate.
Based on that, we can believe that Russia has lost since the start of the war the following:
KIA = 1,300/month x 17 months = 22,100
WIA = 3 x KIA = 66,300
Total KIA and WIA = 88,400
Compare this to my estimate of a minimum of 432,000 to a maximum of 857,000 Ukrainian total losses. At the low end, this is five times more Ukrainian casualties than Russian. At the high end this is almost ten times more Ukrainian casualties than Russian. Since this is mostly an artillery and stand-off weapon war, and the Russians have ten times more stand-off weapons than Ukraine, the higher figure is more likely.
To continue from my posted comment, I don’t have any figures on either Russian desertion or prisoner rates, so we would have to rely on the historic rates. But it’s not worth the effort as the totals will be so low as to be irrelevant to the overall losses on the Russian side.
However, an article in an Australian paper about a Ukraine prison camp says about 50 Russian prisoners are admitted per week and about as many are released in prisoner exchanges with Russia. We don’t know how many such prisons Ukraine runs, whether this article is at all accurate (it reads like western propaganda - talking about “Russian conscripts”, there are no Russian conscripts in Ukraine), or anything else of value. If we take the 50 figure, multiply by 4, we get 200/month or a total of 3,400 for the entire 17 months of the war. Not a significant number. Reportedly way back in May, 2022, the Donbass militias were claiming 8,000 Ukrainian POWs to date. My estimate based on a 5-10% historical range was 35-70,000 based on a Ukrainian mobilization estimate of 700,000 and is quite possibly too low.
If we assume a historical rate of 10% for the Russians, likely too high, then the Russians have lost perhaps 20,000 prisoners based on the initial mobilization figure of 200,000 (the fall mobilization troops have mostly not been committed yet.) Since most of the ground fighting has been done by units other than Russian regular army, this figure is probably too high and consists of Donbass militias, Volunteer Battalions, Rosqvardia (which are not regular army) and Wagner rather than Russian regular army.
Also, since this war is one of attrition by stand-off weapons - artillery, armor, missiles and air power - which Russia has a ten-to-one advantage in, the likelihood of Russian prisoners approaching Ukrainian prisoners is minimal.
The desertion rate for Russian forces is likely on a par with historical rates for say, the British army, which I mentioned in my previous article. Assuming a similar figure of 10 per thousand, the number of Russian desertions, based on an initial mobilization of 200,000 (including non-regular units) to a maximum of 500,000 (including the fall, 2022, mobilization) would range as follows:
Initial mobilization: 2,000
Total mobilization: 5,000
Again, a mostly insignificant number, as it was for Ukraine.
If we add everything up, as we did for Ukraine, we get:
Total KIA and WIA + POWs + Desertions = 88,400 + 20,000 + 5,000 = 113,400 Total Losses
If Shoigu was correct that 90% of the WIA return to duty, we subtract from that total 59,670 to get total losses as: 53,730, mostly KIA, POWs and desertions.
Compare that to my Ukrainian estimated losses from my previous article:
Low Estimate: 390,000 + 35,000 + 7,000 or 432,000
High Estimate: 780,000 + 70,000 + 7,000 or 857,000
This should put to rest the notion that the Russians aren’t taking casualties. However, it is clear that their casualties are at least 4 to 8 times less than Ukraine’s, perhaps more.
Are these “acceptable losses”? There are no “acceptable losses”, there are only losses. It is a political decision as to what losses are “acceptable” to the political authority. For Ukraine and the West, the annihilation of half a million to a million people is “acceptable” (as it was in the 2003 Iraq war - and those were mostly civilians.) For Russia, the loss of 50-100,000 is “acceptable.” Compared to World War II, it’s almost minuscule.
It is what it is.
Since your article of 11th July, I've tabulated the daily deaths and materiel dustruction reported by Russian MOD. That article reported 26,000 deaths. The sum of the daily totals up to yesterday (25th) is just over 14,000, so that up to 40,000 KIA in less than two month's action suggests 120,000 WIA over the same period. Not really sustainable over the longterm me thinks. James is doing sterling PR work on your behalf at MoA ;o)
Excellent analysis, which corroborates my own research in this area.