46

Ukraine independent media seems to quote figures from its General Staff, that sound large, and it's hard to get a sense if they are likely to be accurate, or inflated.

enter image description here

Are these figures seen as plausible/reliable? What do reputable third party onlookers say? And how did such numbers arise, if plausible? Are Ukraine's losses comparable?

14
  • 10
    "Independent" and "reputable" is already quite limiting to select your source of information. And "onlookers" - if you want to verify Russian losses, you need to have onlookers at Russian military side. But you cannot - war secret, mind you! So there are two possible "onlookers" out there, Russian Def.Ministry and Ukrainian one, neither is "independent" or "reputable". So the question is moot.
    – Vesper
    Mar 17, 2022 at 12:02
  • 7
    @aca My understanding is that for about half of the equipment, there's photographic evidence on social media (cf the answer of CDJB). That might give a vague baseline for how much to trust these numbers.
    – Arno
    Mar 17, 2022 at 12:15
  • 8
    @aca It's much easier to accurately determine when you destroy a tank or airplane (since you have a discrete missile launch followed by a big explosion) than when you kill or injure a soldier. Clearly the numbers are not 100% accurate, but the figures on destroyed equipment are definitely going to be significantly more accurate than the figures on losses of troops
    – divibisan
    Mar 17, 2022 at 15:12
  • 9
    Everyone over-states enemy losses. The true numbers won't be known until long after the fact, when the historians have had years to go over the primary sources.
    – Mark
    Mar 17, 2022 at 23:54
  • 3
    @divibisan Not only do vehicle kills go boom, but soldier kills might not even provide any evidence at all. A mortar round explodes behind a truck, the truck is able to drive off. If there's a cover over the truck you don't even know if somebody was there, let alone what happened to them. Thus even careful reporting by troops can't give you an accurate count. Mar 18, 2022 at 5:36

5 Answers 5

39

It's obviously impossible - given the fog of war - to accurately determine figures on losses sustained by either side, especially while the conflict is ongoing. There have been some efforts, however, to record verifiable equipment losses - most predominantly by the Oryx blog, based on work by the investigative journalist Stijn Mitzer as well as others. At the time of writing (March 17th), it has published evidence of the destruction/capture/abandonment of 1,381 Russian vehicles (429 of which are trucks/jeeps/etc), and 366 Ukrainian vehicles (85 being trucks/jeeps).

Comparing these to the figures published by the Ukrainian General Staff solely on the 429 figure, this is about half of the 864 reported. One has to bear in mind of course that Oryx' figures consist of equipment losses verified by image only - so are likely an underestimate according to the blog:

This list only includes destroyed vehicles and equipment of which photo or videographic evidence is available. Therefore, the amount of equipment destroyed is significantly higher than recorded here. Small arms, munitions, civilian vehicles, trailers and derelict equipment (including aircraft) are not included in this list. All possible effort has gone into discerning the status of equipment between captured or abandoned. Many of the entries listed as 'abandoned' will likely end up captured or destroyed. Similarly, some of the captured equipment might be destroyed if it can't be recovered.

The Pentagon, while not particularly independent perhaps, has - according to the NYT - published an estimate of the number of Russian losses, the 'conservative' side of which is 7,000 Russian troops killed, with 14,000 - 21,000 injured.

9
  • 6
    @Trilarion yes, that's true - although the 'Z' logos on Russian equipment helps narrow things down to this conflict. But it's true of course that equipment recorded as captured could then subsequently show up as a duplicate if destroyed.
    – CDJB
    Mar 17, 2022 at 12:51
  • 5
    @Trilarion Based on what Stijn Mitzer posts on Twitter, he seems to be rather meticulous about identifying these kinds of situations.
    – Arno
    Mar 17, 2022 at 16:18
  • 12
    It is worth noting that the Oyrx blog, while its numbers are solid (verified by photographic evidence and confirmed by third-party experts), are inherently biased towards being too low, because not every Russian loss can be verified to their standards period, let alone in real time, and it is making no effort to estimate what portion of total losses are being reported to it. Reporting is particularly difficult in the conditions where battles are most pitched and hence losses are most great. So, the blog estimate is best seen as a minimum. Pentagon data uses real time sat intel & estimates.
    – ohwilleke
    Mar 17, 2022 at 17:34
  • 4
    @Trilarion It is an underestimate. The verification by photograph or videos is done with timestamps and geolocations when possible rather than just counting wrecks in pictures. Images from other conflicts are frequently identified and ignored. Each vehicle confirmation has its corresponding image available. Feel free to find the duplicates.
    – David S
    Mar 17, 2022 at 18:31
  • 16
    @Trilarion the data is public - find the dupes. Your opinion has far less evidentiary worth than data and analysis.
    – Stian
    Mar 17, 2022 at 23:24
18

In addition to suggestions from other answers, here are some additional points.

Evaluating The Biases From Different Sources

The Oryx blog estimate, is a conservative lower bound because its confirmation requirement necessarily means that there are other unconfirmed losses that also exist and there is no estimation of what percentage of losses are unconfirmed. But, since its standards are consistent for all sides, the relative magnitude of each side's losses should be fairly accurate.

It is also fair to view Ukraine's reports of Russian losses as an upper bound on possible Russian losses, and to view Russian reports of Ukrainian losses as an upper bound on Ukrainian losses.

For example, the Russian News agency TASS claimed on February 28, 2022 that:

Russian troops destroyed 1,146 Ukrainian military infrastructure facilities during their special operation for demilitarizing Ukraine, Russian Defense Ministry Spokesman Major-General Igor Konashenkov said on Monday.

"Since the beginning of the special operation, the Russian Armed Forces have crippled 1,146 Ukrainian military infrastructure facilities, including 31 command and communications centers, 81 S-300, Buk-M1, and Osa surface-to-air missile systems, and also 75 radar stations," the spokesman said.

Operational-tactical aircraft of Russia’s Aerospace Force wiped out six armored columns of the Ukrainian armed forces, the general said.

The Russian combat planes "destroyed 311 tanks and other combat armored vehicles, 42 aircraft and helicopters (including those on the ground), 51 multiple launch rocket systems, 147 field artillery guns and mortars, and 263 special military motor vehicles," the Defense Ministry spokesman said.

While it is possible (and indeed likely) that Russian forces destroyed less Ukrainian military equipment than that in the first five days of the conflict, it is almost certain that it did not destroy more than that much Ukrainian military equipment by then. It does appear, however, from independent news reports, that almost all fixed Ukrainian radar stations were destroyed in the first day or two of the conflict.

Ukraine has not publicly reported on any of its own military losses, to the best of my knowledge, although it has reported its own civilian losses.

Russia has made reports of its own military losses, but these reports (of perhaps 500 Russian soldiers killed) are absurdly low compared to any other reasonable estimates from other sources.

The U.S. Pentagon numbers are informed by satellite images in real time and other sources of intelligence of a variety of means. Also, importantly, unlike the Oryx blog, the U.S. Pentagon makes at least a rudimentary effort to estimate the total losses from incomplete information.

This is probably the closest to accurate data available in absolute terms. So, it can be used to some extent to calibrate data like that from the Oryx blog, except that it is much less detailed in its assessments of equipment losses, instead focusing on lost lives and injuries.

Estimates of equipment losses provide some guidance regarding lives lost, injuries, and captured or deserting soldiers, but only a crude one. Ideally, one would look at changes in Oryx blog, Ukrainian government, and Russian government claims over time and compare them to changes in the Pentagon estimates over time, in an effort to calibrate this kind of estimate.

Other Estimation Tools

Expected Russian v. Ukrainian Losses

As of March 10, 2022 (a week old, and hence stale):

The [Oryx] website has counted significantly more Russian equipment losses than Ukrainian losses: 993 versus 277. Of particular importance, it has counted losses of 465 Russian heavy vehicles (tanks, armored fighting vehicles, infantry fighting vehicles, armored personnel carriers), versus 138 for the Ukrainians.

Forbes magazine offers another rule of thumb from military experts looking at past conflicts:

An invading army should expect three times the losses of the army on the defense, if the two armies are equally matched. These numbers suggest that the two militaries are fairly equally matched and that neither side necessarily has a technology advantage.

Deaths v. Combat Injuries

The estimate that two or three soldiers have suffer serious injuries, but are not killed for every soldier killed in modern warfare, consistent with the U.S. Pentagon's most recent estimates of Russian military casualties, is likely a common rule of thumb based upon experience in previous military conflicts.

Reality Checks From The Number Of Russian Generals Killed

Another relevant point is that there are credible reports that at least four Russian Army generals have been killed in the conflict. Multiple sources have also reported that something on the order of 180,000 to 200,000 Russian troops have been deployed in Ukraine. In the U.S. Army, there is about one general per 2,000 U.S. Army soldiers and the ratio is very similar in the U.S. Marine Corps. This ratio would be not all that greatly different in the Russian Army, so there are probably something on the order of 90-100 Russian Army generals involved in the Ukraine conflict.

There has been no military conflict since the early 19th century, at least, where Army generals died at a higher rate than rank and file Army soldiers in active combat (ignoring post-defeat retribution directed at surrendered or captured generals). So, it is safe to infer that it is highly likely that more than 4% of rank and file Russian troops are casualties.

Small numbers, like four general officer casualties, aren't a very precise estimate. Maybe the Ukrainians just got lucky and killed a disproportionate number of generals by sheer chance.

But this number does tend to corroborate the U.S. Pentagon estimate that 7,000 Russian troops have been killed (roughly 3.5%), and that another 14,000-21,000 Russian troops have been seriously injured (roughly 7-10%) for a combined casualty rate including soldiers both killed or injured (as well as soldiers who are captured or have deserted) something on the order of 10-14% of the Russian force.

Indeed, if anything, the fact that four Russian generals have been killed suggests that this 10-14% total casualty rate estimate for Russian troops is conservative and is probably on the low end of the actual total.

Other Inferences About Russian Military Resources

One can also make some more qualitative or crude estimates based upon the kind of tactics that are being employed by the respective sides in the conflict.

Guided Missiles and "Smart Bombs"

For example, there are strong indications that after the initial barrage of missiles from Russian forces on February 24, 2022, that Russian forces have made little or no use of guided missiles or "smart bombs", even though it has at least some of these in it national arsenal.

Guided weapons are, however, vastly superior.

For example, a U.S. guided multiple rocket launcher missile (typically considered a form of artillery) can hit a target with a 10-20 foot (3–6 meter) target radius at a distance of up to about 36 miles (60 km). In contrast, an unguided mobile howitzer artillery shell has a 900 foot (300 meter) target radius (i.e. about 2000 times larger an area where it is 50% likely to hit) at its maximum range of about 18 miles (30 km) and a 180 foot (55 meter) target radius (i.e. about 80 times larger an area where it is 50% likely to hit) at its optimum range. Both deliver explosive effects that are in the same general ballpark in magnitude.

So, it is reasonable to infer that if Russia had guided missiles left, it would be using them. Yet, the news reports from the conflict indicate that Russia is overwhelmingly shelling its opponents with conventional howitzer artillery rounds and unguided rockets, rather than using guided missiles.

A similar inference can be made regarding aircraft delivered bombs. A guided "smart" bomb is about 300 times more likely to hit its intended target than a 1970s-1980s era unguided aircraft dropped bomb. And, a "smart bomb" can be dropped from a higher altitude where it is less exposed to surface to air anti-aircraft missiles and anti-aircraft gun fire, which is a big concern because news reports have established that Western allies of Ukraine have supplied it with at least hundreds, if not thousands, of small, "man carried" Stringer anti-aircraft missiles and the equivalent.

Yet, news reports have also established that Russian aircraft making bombing runs in Ukraine have been using old school "dumb bombs".

It is fair to infer from this fact that Russian forces have pretty much completely exhausted their supply of "smart bombs", given, the greatly increased danger to Russian warplanes and pilots involved in using dumb bombs when "smart bombs" are available, and the fact that "smart bombs" allow bombing aircraft to make vastly fewer bombing runs in an environment where the number of bombing runs available to Russian forces is limited in an environment where it has not have air superiority.

Black Sea Fleet Involvement

We can also infer that the Russian Navy's Black Sea fleet, which consists of 45 ships (not all major surface combatants) and seven submarines, have had little or no involvement in the ongoing conflict after the initial February 24, 2022 barrage of missiles fired across Ukraine, many of which were guided missiles.

There are no reports of naval battles, which surely would have been reported if they occurred, either by the winning side's military, or by the U.S. Pentagon which would have observed this with their satellites.

Also, Russian warships have only two kinds of weapons that can be directed at ground targets: guided missiles, which don't appear to have been used in any significant number since the initial day of the conflict three weeks ago, and naval guns which are basically big unguided shell firing howitzers mounted on ships. But, naval guns have a maximum range of 12-24 miles (20-40 km) depending upon size, only a quite small portion of the shelling in the conflict has taken place close enough to the shore for that option to even be possible, and there are no reports of this being done.

The possibility of Russian Navy fire support for ground troops is also a disfavored possibility because the technical details of coordinating target locations between warships and ground forces are great and there are few, if any, suggestions prior to the conflict that the Russian Navy and Army had developed this capacity to coordinate).

Estimating Percentage Losses

It is also worth noting a couple of important issue comes up with trying to compare the losses to the side of the Russian and Ukrainian forces immediately prior to February 24, 2022, in order to determine losses on a percentage basis.

First, there is considerable disparity between sources over, for example, how many tanks and troops each side's army had prior to the conflict. The Forbes article linked above, for example, reports that:

Russia went into the war with 9,780 tanks while Ukraine had 2,170.

But, the World Almanac 2022 (I rely on the printed version here) citing The Military Balance, 2021 prepared by the International Institute for Strategic Studies, reported that as of November 2020, the Russian Army had 2,840 main battle tanks in its entire Army worldwide, while the Ukrainian Army had 858 main battle tanks.

In this case, the IISS number is almost surely more accurate, because I have seen numerous media reports over the last decade or two reporting that Russia has retired many thousand of its main battle tanks to cold storage for parts and the like, from a peak number close to the one cited by Forbes.

Since there is no good reason to believe that the Russian Army has deployed every single one of its operational and in service main battle tanks to the Ukraine theater in this conflict, moreover, the number of tanks that the Russian Army has deployed is surely significantly less than 2,840 (even adjusting for forces that are part of the Army of Belorussia deployed in this action).

The IISS reports that Russia, worldwide, has 900,000 active duty soldiers and sailors (including at least 150,000 sailors in the Russian Navy), 2,000,000 reserve soldiers and sailors, 185 air to air fighter aircraft, and 415 ground attack fixed wing fighter aircraft as of November 2020 (2021 for soldiers and sailors), while the IISS reports that Ukraine has 209,000 active soldiers and sailors, 900,000 in reserves (the reserve figure is higher than estimated in recent media reports), 71 air to air fighter aircraft, and 14 ground attack fixed wing fighter aircraft.

One reasonable tool to estimate the number of tanks deployed in Ukraine is to assume that if the number of Russian soldiers involved is about 25% of its active duty force outside the Navy, which suggests that something on the order of 620 tanks have been committed by Russia to the Ukraine war.

Given the Oryx and Ukrainian military estimates cited above, this suggests that at least 1/3rd and something less than 2/3rds of Russian tanks deployed to Ukraine have been destroyed by the military of Ukraine.

Second, while essentially 100% of Ukrainian military resources are in play in this conflict, a significant share of Russian and Belorussian military resources are deployed outside the theater of conflict, for example, in Syria and in military bases near the Baltic states.

Third, while the available supply of military equipment on the Russian side of the conflict is more or less fixed, because it hasn't had international allies supplying it with military equipment over the last three weeks, Ukraine has been receiving a substantial volume of lightweight military equipment resources from its Western allies, such as something on the order of ten thousand anti-tank missiles and at least hundreds of drones. This doesn't appear to have included significant numbers of tanks, manned warplanes and helicopters, or other manned military vehicles, however.

In the other direction, Ukraine's self-report that is destroyed the sole frigate, which was the most capable warship in its Black Sea Navy, to prevent it from being seized by the Russians, supported by video images, seems credible.

Still, even with these limitations, one can comfortably say that even the Russian estimates are not consistent with the possibility that the entire Ukrainian air force has been destroyed or rendered useless by Russian invaders, which means that Russia, after three weeks, still does not control Ukrainian airspace.

10
  • 2
    "the relative magnitude of each side's losses should be fairly accurate." I wouldn't count on that. The war takes place in Ukraine. It might be that proportionally more Ukrainians film destroyed Russian equipment than the other way around. Mar 17, 2022 at 20:09
  • 6
    @Trilarion it is not "might be". Military leadership on both sides promotes the difference. Russian soldiers don't have (in theory) any smart devices, everything they have is smuggled or looted and they are reluctant to go online for obvious reasons. Ukrainians (both civilians and soldiers) are encouraged to take pictures and publish them.
    – fraxinus
    Mar 17, 2022 at 21:13
  • 4
    @fraxinus Your reasoning is perfectly sensible. But, as Forbes notes, the relative proportions reported by Oryx are very consistent with the ratios in prior modern military conflicts analyzed after the dust settled which would be expected a priori in this conflict all other things being equal. So, while there are plenty of reasons that the numbers could be off in a biased manner, there is at least some corroboration that the numbers on Oryx are in fact, they are reasonably close to the mark proportionately.
    – ohwilleke
    Mar 17, 2022 at 22:56
  • 1
    satellite images in real time — that seems unlikely. Only the spy sats in low earth orbit have enough spatial resolution to confirm details of losses, and those don't provide information in real time (but only during daytime overpasses, which are 1–2 per satellite per day, and still need to be cloud free).
    – gerrit
    Mar 18, 2022 at 14:25
  • 1
    @TechInquisitor I've not seen any such reports.
    – ohwilleke
    Mar 18, 2022 at 20:34
11

In the current climate it's hard to find any source which can be considered both independent and knowledgeable. Those who know accurate data (the military commands of Russia and Ukraine) aren't independent. Those who are independent don't have direct access to the information, so they can't do more than make educated guesses.

This New York Times article claims to know from the CIA that the actual numbers of Russian losses is about half of what Ukraine claims. But as usual with intelligence sources, the CIA is not particularly keen on talking about their methodology. Perhaps they have hundreds of super-advanced secret spy satellites over Ukraine monitoring every single skirmish. Or perhaps they just hear what the Ukrainians claim and guess that they probably exaggerate by factor 2.

We will probably have to wait until after the war until the fog of war lifts and historians can come to a consensus about how many lives this conflict actually took.

2
  • 4
    CIA/Pentagon's estimates are pretty close to those of open-source Oryx blog. Some of these numbers seemed incredible to me on paper, given Russia's much smaller losses in Georgia (like 3 tanks); but then I saw entire columns destroyed in ambushes, especially 2nd line troops, so that certainly looks quite unlike Georgia. Mar 17, 2022 at 17:04
  • 1
    "the military commands of Russia" and "knowledgeable" - you are making an assumption that they are :) Russia's people in power (military and civilian) have a long and storied history of... shall we say creative accounting for their bosses' consumption.
    – user4012
    Mar 18, 2022 at 3:02
6

The problem is that there are no truly independent sources around and even the US officials the mainstream media quotes appear to be helping the Ukrainians out with the information campaign. Just going through the claimed aircraft losses, because they are by far the most manageable to figure out, quantitatively, yet also the "splashiest" and hardest to hide.

Eg1: https://www.timesofisrael.com/liveblog_entry/us-officials-say-2-russian-transport-planes-shot-down-over-ukraine/

https://theaviationist.com/2022/02/26/two-il-76s-shot-down-in-ukraine/

A second Russian Ilyushin Il-76 military transport plane was shot down near Bila Tserkva, 50 miles (85 kilometers) south of Kyiv, according to two American officials with direct knowledge of conditions on the ground in Ukraine.

The Ukrainians claimed to have shot down 2 large Il-76 transport planes and got backing from US officials for their story, yet the only pictures that were released were subsequently shown to have been those of a Ukranian Il-76 shot down over the Donbass in 2014(by people pointing out that there's no green grass in the area of the alleged shoot-down in February). In 2022, when everyone and their dog has a smartphone, other cameras are everywhere and companies like Maxar are providing high-res satellite photos of the conflict every day, losses like that are flat-out impossible to hide, yet no evidence of these shoot-downs has emerged until now.

E.g. 2 the story of the "Ghost of Kyiv" dogfighting Russian fighters and racking up an impressive kill-count over the city received massive dissemination and even mention from multiple US congresspeople, yet the only documented shoot-down over the city is a Ukrainian Su-27 - https://aviation-safety.net/wikibase/275891

So, TLDR, there will be no reliable data from any side until long after the dust has settled.

3
  • But you point out that Maxar has high quality satellite images. Are they good enough to recognize tanks or other equipment on it? If one carefully looks over the whole country every day, maybe in these satellite images, one could find (and attribute) destroyed larger gear from both sides (unless it's given cover after destruction, but a lower limit might be possible). Mar 17, 2022 at 22:40
  • 1
    @Trilarion not really, because you can't tell who's equipment it is from a satellite photo. But that's irrelevant to my answer about there being no independent sources. Other answers have been citing e.g. NYT citing US government sources as credible ones and not a faction in the information warfare side of the conflict and I've provided counterexamples. My point is that there are no independent onlookers to this conflict that don't have a stake in the outcome.
    – Eugene
    Mar 17, 2022 at 23:50
  • 1
    @Trilarion, the Maxar imagery I use for mapping is good enough to spot a vehicle and identify its general category (eg. truck versus small car), but not good enough to identify the operator or state of repair. I assume the pictures they're taking of the combat zone are of comparable quality. (The US intelligence agency photos are good enough, but they're not being released to the public.)
    – Mark
    Mar 18, 2022 at 1:45
5

Most of the publications in independent journals I seen assume that each of the fighting sides reports the own losses as lower and enemy losses as higher than is likely in reality, writing like Ukrainian figures (which tend to be high, with the latest at 13,500), Russian figures (which tend to be low, with the latest at 498).

Usually some estimations provided by the third party are considered more reliable. In case of Russian losses, most report about half of these values. The New York Times reports 7000, for instance (source).

4
  • 1
    Probably comes down to how reputable the NYTimes is in that regard. Mar 17, 2022 at 12:47
  • 1
    They earn money for they work, I need to pay them for the subscription. This somewhat limits on how much nonsense can they afford. But in this case it seems that they just took the number from the CIA release.
    – Stančikas
    Mar 17, 2022 at 13:32
  • It works both ways: NYT subscribers are mainly Americans who vote for Democrats. So NYT is unlikely to publish things that their readers would find objectionable. This obviously applies to all commercial media, and NYT still remains a rather reputed one.
    – Roger V.
    Mar 18, 2022 at 15:54
  • 1
    Here more depends if the reader wants objective truth and serious analysis. If one wants the truth, one pays for exactly the truth and does not pay for propaganda, so why to publish? The demography of the readers is just an interesting statistics.
    – Stančikas
    Mar 18, 2022 at 16:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .