Reliability of Sovjet military deaths

Discussions on all aspects of the USSR, from the Russian Civil War till the end of the Great Patriotic War and the war against Japan. Hosted by Art.
Post Reply
Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#106

Post by Art » 12 Dec 2021, 11:18

thorwald77 wrote:
10 Dec 2021, 17:55
Total reinforcements sent to the front in section 1 of the document dated 14 February 1943 are listed at 5,489,418.
They say that the army received 5,486,475 newly conscripted personnel in 1942, which provided for:
- increase of the active army strength by 1,443,841 between 1.1.42 and 1.1.43
- increase of the number of patients in hosptals by 319,500
- compensation of medical discharges and furloughs (863,740)
- compensation of irrevocable casualties (2,544,505)
In section 6. of Krivosheev p.213 they list 5,328,395 conscripted in 1942.
in eight months beginning from 1.5.42. On the other hand Krivosheev counted all the men regardless of where they went to - army, navy, NKVD, NKPS railroad troops, labor columns, construction columns, jobs in civil agencies etc. That the numbers are close is incidental.
Document 130022SS from 14 February 1943 indicates that the Soviet General staff
It's not the document of the General Staff. RKKa General Staff lost controll over mobilization, organizations, replacements and new formation in the summer of 1941 an and only partly reocvered it thereafter.

AriX
Member
Posts: 184
Joined: 29 Jun 2015, 09:07
Location: Ukraine

Re: Reliability of Sovjet military deaths

#107

Post by AriX » 12 Dec 2021, 15:09

Soviet Armed Forces strenght on 6/22/41 - 5.7 mln.
Soviet Armed Forces strenght on 1/1/43 - 10,95 mln + 0.85 mln. in hospitals. Total : 11.8 mln.
Number of drafted and mobilised in
1941 - 11, 79 mln.
1942 - 5,4 mln
Total : 16.2 mln.

Do the math : (5.7 mln + 16.2mln) - 11.8 = 10.1 mln disparency of whom 0.98 mln are discarged due to medical reasons and 5.6 mln Tchadencos losses KIA, MIA and DOW.
c.2.6 mln are "lost"


User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#108

Post by thorwald77 » 12 Dec 2021, 17:36

Art thanks for the corrections on the 1/1/1943 balance of Soviet forces.

There is just one more thing, I have a question.

Krivosheev lists the detailed losses of the 64 fronts and fleets. Were these figures based on actual reports generated during the war that we can verify or were the statistics the estimated actual losses prepared after the war?

https://youtu.be/vYgUPD2khtw

Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#109

Post by Art » 13 Dec 2021, 10:58

thorwald77 wrote:
12 Dec 2021, 17:36
Krivosheev lists the detailed losses of the 64 fronts and fleets. Were these figures based on actual reports generated during the war
Yes, for the most part. With exception of the year 1941, where numbers include estimated numbers of missing in actions in addition to reported numbers.
Document 130022SS from 14 February 1943 indicates that the Soviet General staff had functioning system in early 1943 to account for losses in the war. The "resolute measures" taken in 194-42 were necessary to put this in place. I can imagine hundreds of Soviet women in Moscow working on adding machines under strict deadlines processing reports from the front. NKVD auditors would check and verify the reports. Timoshenko would present the final report to Comrade Stalin in the Kremlin. Timoshenko would deliver messages and manage the office in Moscow while Zhukov would be commanding the forces at the front.
Well, actually the authorized strength of the 8th Department of the General Staff's organizations directorate (registration of the personnel strength) was 37 military and 14 civilians (establishment tables issued in 1943)
ORGU.png
It was the same 8th department headed by colonel Podolsky that produced this summary table of casualties:
https://www.soldat.ru/data/images/zamo_ ... _29_74.jpg

User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#110

Post by thorwald77 » 13 Dec 2021, 19:28

Thanks Art for this information. Krivosheev apparently has given us a summary of the casualties reported during the war by the various fronts and fleets, the estimated number of reservists missing and convicts and deserters. To balance to the total conscripted 11,944 million he plugged 1,162.6 million.

In millions
  • Krivosheev 8,668.4
    Reservists Pows 1,000.0
    Convicts 557.0
    Deserters 376.0
    Defectors to west 180.0
    Subtotal 10,781.4
    Plug 1,162.6
    Irrecoverable losses 11,944.0
The acceptability of a plug depends upon the amount: a plug must be immaterial in order to be justified.

https://en.wikipedia.org/wiki/Plug_(accounting)

User avatar
TheMarcksPlan
Banned
Posts: 3255
Joined: 15 Jan 2019, 23:32
Location: USA

Re: Reliability of Sovjet military deaths

#111

Post by TheMarcksPlan » 22 Dec 2021, 11:43

Art wrote:
18 Nov 2021, 10:57
Let's return to the casualties electronic database (https://obd-memorial.ru)
At the present moment the database yields above 11 million records from casualties list submitted by military units in 1941-45. By year:
1941 - 1 673 763
1942 - 3 232 821
1943 - 2 859 979
1944 - 2 358 349
1945 - 1 229 716
As already explained, the database contains many duplicating and overlapping records. Hence the physical number of casualties is smaller than the number of records. In order to estimate the scale of duplicating I've made the sample search, using a very typical name "Sergey Sergeyev". The database yields 509 records corresponding to this name, including:
killed - 228
fallen - 37
died of wounds - 65
died of disease - 15
missing in action - 138
misc. (sentenced, wounded, died) - 26
After removing duplicating records that leaves 351 casualties. That's the maximal number, because I can't be sure that I found all duplicates. Corresponding correction factor is roughly 70%. By causes casualties are distributed as:
killed - 193
died of wounds - 46
died of disease - 11
missing - 91
drowned - 1
died in captivity
died of poisoning
executed - 1
suicide - 1
died of hunger - 1
unspecified reasons - 4

So uniformly applying 70% correction factor to all records we obtain the following rough estimates for the number of casualties
1941 - 1 170 000 vs. 3 140 000 irrevocable casualties according to Krivosheyev
1942 - 2 260 000 vs. 3 260 000
1943 - 2 000 000 vs. 2 300 000
1944 - 1 650 000 vs. 1 760 000
1945 - 860 000 vs. 800 000 (January-June)

Note that database includes casualties of regular military (Army, Navy, NKVD) and in some cases even irregular (MPVO or partisans). while official publication gives only casualties of the Red Army and Navy. Anyway, the conclusion is that in 1944-45 we have a reasonably good match between the database and official numbers. At the same time in 1941-42 the database underestimates the number of casualties. That is what can be expected from a common sense, since registration was obviously incomplete and, in particular, many missing and prisoners were not included in casualties lists. The discrepancy would be even more larger when considering that official numbers are also understated. The year 1943 is an intermediate situation, I can't say if the difference is real or resulted from a sampling error.
Thanks for this walkthrough. Some folks are making a lot of mischief online by reference to this database but your explanation of its (superficially paradoxical) over- and under-statement relative to Krivosheev makes perfect sense. It overstates losses before correction for duplicates; it understates them after correction because registration was incomplete (particularly for early years).

--------------------------------
TMP bookmark: Art's explanation of the Soviet casualty database
https://twitter.com/themarcksplan
https://www.reddit.com/r/AxisHistoryForum/
https://medium.com/counterfactualww2
"The whole question of whether we win or lose the war depends on the Russians." - FDR, June 1942

User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#112

Post by thorwald77 » 22 Dec 2021, 17:28

Krivosheev does not make sense. Here is the proof.
  • Krivosheev 8,668.4
    Forced labor 2,164.3
    Penal units 422.7
    Executions 135.0
    Deserters 376.3
    Defectors 180.0
    Civilians in military 91.0
    Milita 146.0
    Partisans 250.0
    Civilians killed 4,500.0
    Seige Leningrad 658.0
    Bombing Stalingrad 40.0
    2nd emigration 451.1
    Emigration Germans & Romanians 170.0
    Total 18,252.8
  • Krivosheev 8,668.4
    Forced labor 2,164.3
    Civilians exterminated 7,420.1
    Total 18,252.8


Notes:
1-Forced labor is actually balance of casualties not reported in Soviet official figures during the war. Krivosheev's data is incomplete, the proof is that his data does not agree to the balance of 11,944 million irrecoverable losses in the reconciliation of conscripts. Krivosheev deducted the balance of 1.1174 million in NKVD service, which included penal units, executions and deserters. Krivosheev's data is worthless, soft as shit.

User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#113

Post by thorwald77 » 22 Dec 2021, 20:23

Here is the corrected reconciliation of forces which includes the Far East Campaign 12.0, NKVD forces 1,169.8 and a deduction for civilians serving with the military (91.0).

All figures i (1,000)
Reconciliation of conscripts
  • Forces 6/41 4,826.9
    Civilians in military 74.9
    Conscripted during war 31,812.2
    Total called up 36,714.0

    Far East Campaign (12.0)
    Adj. Border troops(see below sch) 4.8
    Civilians conscripted for military service (91.0)
    Rejected medical reasons (1,154.8)
    Transferred to industry (142.8)
    Total available forces 35,318.2

    Evacuated to hospital (3,798.2)
    Transferred to industry (3,614.6)
    Released Pows (2,016.0)
    Transferred to Polish/Romanian forces (250.4)
    Convicts (436.6)
    Discharged (206.0)
    Deserters to interior (212.4)
    Forces 7/45 (12,839.8)
    Irrecoverable losses 11,944.2

  • KIA 5,187.2
    DOW 1,100.3
    Non-combat losses 541.9
    MIA 500.0
    POW 3,445.0
    NKVD border/Security 55.8
    Convicts,deserters 934.0
    Post war defectors 180.0
    Irrecoverable losses 11,944.2

    Transferred to NKVD/Civilian depts Bal 7/45
    NKVD border/Security 55.0
    Convicts,deserters 939.7
    Post war defectors 179.9
    Total @ 7/1945 1,174.6


    Adjs Krivosheev
    NKVD border/Security 55.8
    Convicts,deserters 934.0
    Post war defectors 180.0
    Total per Krivosheev 1,169.8
    Difference 4.8

Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#114

Post by Art » 23 Dec 2021, 08:17

Art wrote:
13 Dec 2021, 10:58
Well, actually the authorized strength of the 8th Department of the General Staff's organizations directorate (registration of the personnel strength) was 37 military and 14 civilians (establishment tables issued in 1943)
Administration for registration of enlisted men casualties established in 1943 was much larger. Its authorized strength was 64 military personnel and 414 civilians. This administration handled the lists of casualties submitted by units, questions of pensions, search of personnel by relatives etc and included among others statistical department and a card index department. While the General Staff received reports on the numbers of casualties this administrations received casualties lists and collected names of enlisted casualties and developed a database on this basis.
At the same time registration of officer casualties was assigned to the Main Cadres Administration, which included a registrations and statistics administration (authorized strength 26 military and 155 civilians), whihc in its turn included a department for registration of officer losses (6 military and 50 civilians).

User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#115

Post by thorwald77 » 24 Dec 2021, 15:14

Administration for registration of enlisted men casualties established in 1943 was much larger. Its authorized strength was 64 military personnel and 414 civilians
What was the authorized strength in 1941-42?

I can imagine a large office in Moscow with Soviet women working on adding machines, upstairs was a communications section receiving coded telegraph messages from the front.

Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#116

Post by Art » 27 Dec 2021, 22:41

Prior to 1943 all that belonged to Galvupraform as Central Bureau for registration of casualties. I don't have a detailed establishment, I suspect that numbers were broadly similar.

Paul_Atreides
Member
Posts: 606
Joined: 09 Sep 2008, 09:05
Location: Russia, St. Petersburg

Re: Reliability of Sovjet military deaths

#117

Post by Paul_Atreides » 28 Dec 2021, 07:35

Art wrote:
27 Dec 2021, 22:41
Galvupraform
Glavupraform.
There is no waste, there are reserves (Slogan of German Army in World Wars)

Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#118

Post by Art » 09 Jan 2022, 15:43

According to the regulation on registration of individual casualties in wartime approved on 15 March 1941 Red Army's division and brigades were to sumbit list of casualties to the Staffing and Equipment Administration of the RKKA General Staff three times a month by 1, 10 and 20 day of each month. Corps HQ submitted lists of casualties of the corps HQ and corps units in a similar manner. On 28 July 1941 the Staffing Administration was removed from the General Staff and trasnferred to the newly established Glavupraform. According to the NKO order of 29 July 1941 the Glavupreform consisted of 5 administrations and 2 separate departments. The first (organisations) adminsitration included the 9th department whose function was to register the army's strength, number of personnel losses and replacements. The second administration (staffing and equipment of troops) was to register individual casualties and carry out correspondence with relatives. For this purpose it included the 4th department ("Individual registration of personnel casualties"). Already on 9 August 1941 a new regulation on Glavupraform was approved which changed its structure somewhat, so now it consisted of 5 administrations and 5 separate departments. The first administration (organisations and tables of establishment) included the 8th department which registered strength and number of losses of the army. A separate 10th department was established with a function of registration of individual casualties. This department was to consist of 28 military personnel, 43 civilians and 30 clerks (privates).
On 19 November 1941 by the NKO order the department for registration of individual casualties was included in the Staffing Administration of the Glavupraform as its 5th department.
By the NKO order No.25 of 31 January 1941 the Central Bureau for Registration of Casualties was established within the Glavupraform on the basis of the 5th department of the Staffing Administration. According to the establishment issued on 10 February 1942 this Bureau was to consist of:
1st Department (registration and statistics) - 59 men, including 9 military
2nd Department (Information) - 70 men, including 7 military
3rd Department (Correspondence) - 113 men including 18 military
4th Department (Inspection) - 8 military personnel
5th Department (General) - 42 men, including 6 military
On 25 March 1942 the 6th Department (Orders) was etablished with a function to prepare orders on updating the officer registry and compile a card index of [officer] casualties.
This structure remained essentially unchanged until the Bureau was disbanded in 1943 and Red Army's administration for registration of enlisted casualties was established as described above.

This information is mostly from Shabayev and Mihalyov "Tragediya protivostoyaniya" (2002).

User avatar
thorwald77
Member
Posts: 528
Joined: 25 Sep 2007, 18:42
Location: usa

Re: Reliability of Sovjet military deaths

#119

Post by thorwald77 » 09 Jan 2022, 16:59

Thankyou Art for taking the time to post that information.

Art
Forum Staff
Posts: 7028
Joined: 04 Jun 2004, 20:49
Location: Moscow, Russia

Re: Reliability of Sovjet military deaths

#120

Post by Art » 23 Aug 2022, 12:39

Art wrote:
18 Nov 2021, 10:57
Let's return to the casualties electronic database (https://obd-memorial.ru)
At the present moment the database yields above 11 million records from casualties list submitted by military units in 1941-45. By year:
1941 - 1 673 763
1942 - 3 232 821
1943 - 2 859 979
1944 - 2 358 349
1945 - 1 229 716
Current state of records in the database (as of this month)
OBD.png
"All records" - all records pertaining to a specific months, some of them non-casualties, such as wounds, desertions, court sentences etc.
"Sum" - sum by months
"Totals" - total number of records of specific type.
"Fallen" - in most cases belong to two type of sources: 1) records of Navy and NKVD casualties 2) unit-level records of burials, where specific cause of death was not recorded.

As explained above, the database contains considerable duplication, so the physical number of casualties is smaller than raw numbers, presented in the table, by a large (and unknown) percentage.

Post Reply

Return to “The Soviet Union at War 1917-1945”