Ukraine war part 5: How Russia fights (an analysis of casualty figures)

In part 1 of my Ukraine series, I had provided my estimate of casualties on both sides.
I had updated by estimates of July, to include an analysis by Mediazona (with the BBC) on Russian casualties. This is the most intellectually rigorous (and therefore credible) work on the subject I’ve found. It was heartening to me that against my `back of envelope’ estimate in July of 40,000 Russian dead and irrecoverable casualties (dead + seriously wounded and prisoner) of 120,000 Mediazona’s estimate at the end of August was 47,000 dead and 125000 irrecoverable casualties.

Part 1 of the Ukraine series:
https://rpdeans.blogspot.com/2023/07/ukraine-war-understanding-numbers-part.html

In response to feedback that my blog is more supportive of the Russian position than the mainstream media, it must be reiterated that there is no equivalent analysis of casualties in Ukraine, therefore I have to confine myself to studying Russian casualty figures. The purpose of this post is to analyze what these casualty figures tell us about how Russia is fighting the war and how it matches Russian and western media claims. 

There is a prevailing view in the Western mainstream media that the Russian army (based mostly on stories from the early 90s) is a corrupt, inefficient and badly led organization. A far more credible view was in a paper by the Modern War institute at West point, which is both professional and credible and 
is what the US military reads. You can read extracts in this very informative blog:
https://simplicius76.substack.com/p/dissecting-west-point-think-tanks  
It suggests that Russia has anticipated how a modern war between peers would be fought, to a far better
extent that believed and is acting in an optimal way. It lends credence to the following analysis of casualties. 

Of the estimated 47000 dead Russian soldiers in Mediazona's analysis, 31665 have been identified and in most cases we have information on which arm of the military they belonged to and their age and rank. The ratio of estimated dead to identified dead, is 47000/31665 = 1.48 I will therefore use 1.5 as a multiplier to estimate the total dead for any category in which we have dead who are identified.

The figure of 47,000 dead has been represented in the Western media as being 3-4 times more than the total casualties the Soviet army suffered in Afghanistan, in a much longer period. This is a disingenuous comparison, because Afghanistan (or Chechnya) was a counter insurgency operation, not a mechanized high intensity war against a peer opponent of similar size and capability. 

Compared to the average Russian losses of 84 (killed) per day over 560 days, similar conflicts in the past had the following casualty rates for a roughly similar sized force as the Russian army.

In the 1973 Yom Kipper war, the casualty averse Israeli army had 130 dead each day.
In the 1971 India-Pak war, the victorious Indian army had 200 killed per day. Adjusting for army size,
it was similar to the IDF’s 130/ day in 1973.
In the 8 year Iran-Iraq war, each side had an estimated 75-80 killed /day, but neither lost the will to fight until after 8 years, despite having much smaller populations and economies than Russia.

In smaller wars (1982 Falklands and 1999 Kargil) the willing side – UK and India, lost approx. 3 men killed per brigade (of 4 battalions) per day. This would also translate to about 130 dead/ day for a force of similar size to the current Ukraine and Russian armies currently in combat.

Hence overall Russian casualties are in line, if not lower than similar conflicts in the past involving professional armies that won.  

Of more significance (and something I comment on in part 3 or my Ukraine series):
https://rpdeans.blogspot.com/2023/07/ukraine-war-part-3-missed-opportunities.html
There has been a steady decline in the rate of Russian losses.

Since 1^st April 23, there has been no week in which estimated Russian dead were over 60/ day (280 per week confirmed with a multiplier of 1.5 for unaccounted for dead).
Every month since March 23 has shown a lower figure of dead. Against confirmed dead of 500 week in the opening month of the war, there were 280/week in March 23, there were 140-200 dead/week in Jun-July and under 100/week killed in Aug.
At the current rate, a figure of 5200 killed/ year, in an army whose combat formations now number has 500,000 is something Russia can sustain indefinitely, if this force can be supplied. In Part 2 of my Ukraine series, I suggested that they can.

Conversely, whatever evidence there is, suggests Ukrainian casualties have increased in the same period that Russian casualties decreased. Statements from the Ukrainian govt suggest that the lowest level of casualties they sustained in this conflict was around 100 per day, which can be considered a baseline.

Apart from the first month of fighting, Russian casualties peaked during the battle of Bakhmut, as did Ukraine’s. However, Russian casualties in Bakhmut mostly comprised convicts recruited into the Wagner private military Corp (PMC). Of the 31665 identified dead, 5968 were known convicts and 7582 did not have a branch of service identified. Assuming 4000 of these 7582 were also convicts, there were 10,000 dead among the Wagner convicts (which ties with the Wagner estimate of 30,000 irrecoverable casualties – assuming 2 seriously wounded for each person killed). There were an estimated 50,000 irrecoverable Ukrainian casualties in Bakhmut – I have derived this by the number of Ukrainian brigades involved which were removed from the order of battle for reorganizing and by `deflating’ the Russian and Wagner claims of Ukrainian casualties. Thus Bakhmut was an exchange of the best Ukrainian formations for a smaller number of Russian convicts, after which the exchange ratio turned sharply in Russia’s favor. Between Dec 22 and March 23 almost half of all Russians killed were known to be inmates. Assuming half of the  `identified but unknown arm of service’ were also inmates, they accounted for 80% of all Russian casualties between Dec and March. Thus the best Ukrainian formations (excluding those being trained and held back for the June counter offensive) were bled at Bakhmut, while the Russian army was largely intact.     

An estimated 42% of all dead belonged to the Wagner private contractor group. These included both former convicts and professional mercenaries. This is actually similar to US /Allied casualties in Iraq and Afghanistan. 8000 members of OMC's were killed compared to approx. 7000 US military personnel.  

Given the lower number of Russian deaths during the `all or nothing' Ukraine counter offensive
and the higher number of Ukrainain casualties ( there are no official figures, I am going by the number of brigades involved, the mission profile (attacking into prepared defenses) and Western analysts and media reports), the loss ratio has moved significantly in favor of Russia. Since all Ukraine's NATO trained, or regular army brigades have had significant casualties and replacements will be conscripted manpower who had hitherto avoided military service, its possible that the loss ratio would decisively 
swing in Russia's favor from now on.   

The 300,000 men newly mobilized last Sept, lost 3534 men known to be from the newly mobilized. 
If one includes those identified without their service branch being known and applying a 1.5 multiplier, we have 6000 dead among the newly mobilized in 1 year, or just 2%. Thus, the newly mobilized were not, as some in the Western media suggest, conscripts thrown into a meatgrinder, but those with military training (hence listed as reservists), who were chosen from among the reserves, based on their prior experience and in theory, retrained, before being deployed.

The newly mobilized force of 300,000 started taking casualties from the last week of Sept, belying the claim that all volunteers were retained before being deployed. It may also indicate that there were serious shortfalls in the Russian order of battle (ORBAT) which required quickly deploying the newly mobilized. This need would have been exacerbated after the Russian army’s retreat from the Kharkov sector, to beyond the natural defensive line of the Oskol river, giving up the important logistics hubs of Kupiansk and Izyum. There was also pressure on the Russian forces in Kherson, west of the Dnieper, who has been fighting for the last year against numerically superior forces, with a vulnerable supply chain (across 2 bridges). Since June, the newly mobilized have accounted for about 25% of all deaths.

As of Sept 23, there were 270,000 volunteers who joined the Russian army. The rate of recruitment increased throughout 2023, with a current rate of 30,000 joining per month. By definition they joined willingly and knowing the possible dangers they might face and it would be logical to assume a higher proportion of them went into combat roles, compared to the newly mobilized.

There were 3110 dead from those known to be volunteers (or an estimated 5500 of all deaths).
Between the last week of April 22 (when volunteers were first asked for) and end Jan 23, there were an
estimated 60 deaths per week among volunteers, on a base that averaged around 50,000 volunteers.
However, since Feb 23, deaths were reduced to 30/ week, on a larger average of 150,000 men (assuming a steady increase from 100,000 in Jan to 270000 today). This is a 6 fold reduction in the casualty rate per 1000, compared to a 3 fold reduction in overall casualties per 1000, which indicates that volunteers are getting better training before deployment and being better integrated into their units. By contrast, till Sept 2022, serious shortages in the manpower of combat formations, because many conscripts who were part of a combat unit were absent as they could not serve abroad and due to casualties, would have resulted in volunteers with inadequate training being thrown into combat. Volunteers also currently account for around 25% of deaths.

Thus both volunteers and the 300,000 mobilized men, now have casualty figures proportional to their
presence, which would indicate that they are fully integrated into combat units. Apart from the former convicts in the Wagner group, there is no category of men who are incurring higher casualties as part of any deliberate policy.

Men of the national guard (who are mostly deployed as light infantry or in a anti insurgency role) did not take any significant casualties since Oct 22. They were most likely replaced by the newly mobilized and volunteers.   

There were 573 known casualties from the armored corps, or an estimated 1000 totally.  Assuming 1 tank destroyed for every tank crew dead (with 2 more wounded) we would have 1000 tanks lost as
against an estimate by Oryx of 2321 tanks lost (incl 1508 destroyed). There is a view among some analysts that Oryx has overstated Russian losses (same tank shown in multiple pics, or a Ukrainian tank shown as Russian) but even if the ratio of dead men to dead tanks is changed and only the figure of 1508 destroyed tanks is considered, tank losses have been negligible since April (new tank production is higher than losses). Since April deaths of tank crew have been below 2 a week. New tank production is an estimate 240 per year with a larger number of older (T-72 and T-80) tanks modernized.   

If one looks at the trend in tank losses by type (documented by Oryx), at the start of the war, T-80 losses comprised 23% of Russian tank losses and T-72 (mostly older models) were 71%. During May-Aug 23, the share of T-80s rose to 53% and T-T2s fell to 36% 
Since Russia does not produce the T-80, one can conclude that they are replacing older T-72s with 
refurbished T-80s. The superior speed (particularly in reverse) and heavier armor of the T-80 compared to the T-72 would have been considered important based on the Ukraine experience and worth the higher cost and limited range of the T-80.  
The production of new T-90's is higher than losses (taking Oryx figures which may be inflated). Since the proportion of T-90 losses have remained constant, it can be assumed that newly raised units are being held in reserve and equipped with new T-90s (there is evidence of this from Russian TV clips) and the BMP-3. Among losses of infantry fighting vehicles, the share of losses of the old BMP-1 increased from 28% to 44% which is because new BMP's reaching the front are BMP1s upgraded to the BMP-1AM model of 2018 (which is much cheaper and manufactured in a different production line from the BMP-3).  

The arm with the highest losses has been the airborne forces. 1937 of those identified were confirmed as being from the airborne forces (it would mean an estimated 3150 dead, if the 1.5 multiplier is applied, though I would assume that a member of the elite airborne forces would have that fact disclosed in an obituary. 25% of these casualties were in the first month of combat. For a 30,000 man force in constant combat in the most active sectors of the front, irrecoverable losses (incl. seriously wounded) of 20% in 20 months, is actually not high, but consistent with any light infantry unit in tough fighting. A single division, the US 101 airborne division (a third of the size of the Russian airborne forces in combat) suffered 2155 dead in 11 months in World War 2. The US 82^nd airborne lost 1509 men killed.

Reservist men from the airborne forces or new conscripts can replace losses, without diminishing the capability of the airborne forces – as has been demonstrated when the 76^th Guards Air assault division, in constant combat since day 1, giving a good account of themselves when moved to the Rabotino sector during the ongoing Ukrainian counter offensive.    

Russian Naval infantry (Marines) had a confirmed 766 deaths for its 4 brigades. This is a similar proportion of losses to the airborne forces.

Interestingly, only 3% of those killed were under 20 and 11% under 23. Almost all conscripts are likely to be under 23. The 4 districts where fighting has been taking place are now designated part of Russia where conscripts can serve. A conscript has 12 months service, of which 6 are in his unit. The expansion of the Russian army by 130,000 announced last year, corresponds to 6 months intake of conscripts (or 1 year with 50% joining the army) This gives a total of 260,000 (130,000 doing their 12 month service and 130,000 staying for another year), conscripts in an armed forces of 1.1 million or 24% conscript, who have taken only 11% of casualties. This is the opposite of the Western notion that Russian conscripts are being used as cannon fodder. Most casualties are in the 30s (i.e. professional soldiers). Among volunteers, the age of most of those killed is the 33-47 age group i.e. not young men who join out a sense of adventure, or to get a job, but older men more likely to have families. 
The age split of the current Russian forces in Ukraine is similar to the Indian army which raises the question of weather a younger soldier profile (through the Agnipath scheme is really the best way forward). 

13% of Russians live in Moscow and St Petersburg, but these cities contributed to just 3.5% of
deaths among volunteers or those mobilized. Assuming that deaths are proportional to enlistments,
the poorer areas and those further from Moscow had a larger share of volunteers per capita. Moscow and St Petersburg districts (semi urban with a lower income than the cities) had a higher share of volunteers. However, rich districts like Tyumen or Sakhalin had a higher than average proportion of volunteers. A strong motivator could be the salaries on offer for contract soldiers and possibly a reduced influence of Western culture (unlike in the big cities).    

Among the casualties were 234 listed as foreign.

Most officer casualties are in the 24-26 age group (younger than the men they command). In the early months of the war, there were several reports of Russian generals being killed by targeted strikes. In reality, Russian has lost 8 officers above the rank of Colonel. Russian doctrine, unlike in the West, requires senior officers to spend more time in forward positions and not in HQ. That said, in the Vietnam war, the US lost 12 generals (1 dead general out of every 5000 killed, compared to 1 in 6000 for the Russian army in Ukraine).

25% of the Russian army comprise officers. This is higher than the West, due to a lower number of experienced sergeants and warrant officers in Russia. In combat formations of the army, officers are likely to comprise about 12-15% of the total (since most officers will be in staff positions, headquarters or in technical roles. Although officers comprise 7.7% of the identified dead, it is between 15 – 26% of deaths in the regular army’s combat arms. In comparison, the Indian army, where officers lead from the front, has officer deaths around 5-6%. (6% in Kargil, 5% in the 1971 war). This is consistent with the Russian army tradition of young officers leading from the front.  

An interesting related statistic is about the number of men who have left the country to avoid being conscripted. In the early months of the war, there were numerous reports of `hundreds of thousands’, or `upto a million’ Russians fleeing the country. Cross referencing various media reports gives a figure of 500,000 men who left Russia on a one way ticket. Its possible that some returned, since there isn’t a large mobilization (confined to those who had previous military service) and they are mostly ineligible to be refugees and there was no bar on them leaving Russia.

Ukraine, which is undertaking fresh mobilization in the aftermath of the failed counter offensive (almost none of those who might now be conscripted would have military experience or be willing volunteers) has approached the EU seeking the return of those men who left the country in defiance of a law preventing men of military age from leaving. It turned out that there are a similar 500,000 Ukrainian men in the EU (Ukraine’s population base is now 20% of Russia’s).

Previous posts in this series:

https://rpdeans.blogspot.com/2023/07/ukraine-war-understanding-numbers-part.html

https://rpdeans.blogspot.com/2023/07/ukraine-war-understanding-numbers-part_18.html

https://rpdeans.blogspot.com/2023/07/ukraine-war-part-3-missed-opportunities.html

https://rpdeans.blogspot.com/2023/08/ukraine-war-part-4-ukraines-counter.html