By Big Serge, Substack, 4/10/23

Author’s Note: I had been intending to publish an article on Soviet operational art this week, but the emergence of the leaks diverted my attention and led to this article instead. We’ll return to military history shortly.

Another winter has ended, and spring has again arisen on the war in Ukraine. Amid the thaw and attendant mud, Russian forces – including the indominable Wagner Group – have pushed the Ukrainian grouping in Bakhmut to the brink, with the AFU now clinging its last defensive toehold in the city. Bakhmut has become the largest battle of the 21st century, and is now entering its climactic phase.

Nevertheless, battlefield developments have been upstaged to some extent by the apparent leak of classified US military intelligence documents which provide a sweeping view into the inner workings of the Pentagon’s war.

I am not entirely clear on Substack’s content policies as it relates to such documents. It is certainly too late for the US Government to contain the leak, as the images have by this point been shared, screenshotted, and downloaded countless times, but that does not preclude an attempt to limit its circulation via a whac-a-mole campaign of content deletion. In any case, desiring neither to violate US law nor run afoul of Substack’s content rules, prudence dictates that I ought not embed the images directly in this post, but they are not hard to find – the “Rus Fleet” Telegram channel has them up at the moment, for example. Use your own discretion.

While I will not be posting the leaked documents either here or on twitter, I would like to talk about them. If they are indeed authentic (and it appears that they are), they offer important insight into force generation and combat power in Ukraine – and perhaps even more importantly, into the intelligence framework that the Pentagon is working with. None of the items adduced paint a particularly rosy picture for either the AFU or its benefactors on the Atlantic seaboard.

A Brief History of the Leaks

Let’s briefly indulge in an overview of the leaked documents as such before we think about their contents. They take the form of photographs of physical pieces of paper from an American intelligence briefing. This implies that the particular nature of the breach is a leak (personnel with legitimate access to the documents illegally disseminating them to the public) rather than a hack (someone gaining illegitimate access through intrusion of one form or another). The pages have visible creases on them, and a hunting magazine can be seen on a table in the background. Many of the pages are marked for sharing with NATO allies, but some stipulate US eyes only.

The general impression is that an American folded the briefing documents up, put them in his/her/their/xer/xem/plur pocket (the American military is a Diverse and Inclusive institution, and the leaker could have any, all, or no gender), took the pages home and photographed them. It was almost certainly not a Russian asset – if the documents had been acquired by Russian intelligence, they would have kept it internal.

Now, the obvious question is whether the documents are real. There’s probably at least some rational basis to suspect a misinformation operation. All militaries engage in a range of intermingling intelligence (seeing what the enemy is doing), counterintelligence (hiding what you are doing), and misinformation (lying about what you are doing). Perhaps, one may muse, these documents were not leaked at all, but indelibly planted on the internet to mislead.

I was originally rather agnostic about the documents’ authenticity, but I have come to the view that they are genuine (let’s rate it a 90% likelihood of authenticity and a 10% likelihood of forgery or misinformation). My reasons are essentially as follows:

-The timeline of events suggests an authentic leak. While the documents only started to circulate widely in the last week or so, they were actually first posted to the internet (as best as I can tell) on March 1st – but nobody noticed, apparently. The documents didn’t attract mass attention until a pro-Russian telegram channel found them and reposted them after badly photoshopping the casualty estimates to show much lower Russian losses. Ironically, it was these falsified edits that sparked mass interest in the documents. To me, this suggests that the documents are not part of some sort of Pentagon misinformation campaign, because they essentially sat idle in the remote corners of a Minecraft Discord server for an entire month. If American intelligence wanted to circulate fake documents, one suspects they would have actually circulated them, rather than dropping them in an obscure corner of the information space and leaving them to languish.

-The documents have perfect internal consistency. The full leak includes dozens and dozens of pages which are totally consistent down to the level of delivery dates, inventory listings, and arcane unit identification. This goes even above and beyond the perfect use of acronyms and military symbiology. Creating these documents would be a colossal undertaking and would require both precise subject matter expertise and a mammoth amount of cross-referencing to prevent contradictions – unless, of course, the documents are genuine, in which case the material would be consistent because it is real.

-The documents are relatively low on actionable intelligence. They contain no planning details of Ukraine’s coming offensive operations and only hazy outlines of Ukrainian force dispositions. A ruse intended to deceive the Russians would be expected to contain highly actionable (but false) intelligence.

-Finally, both the government and the media are proceeding as if the documents and the associated security breach are real, and they are attempting to both limit the spread of the documents online and track down the source of the leak.

All of this to me suggests that these documents offer a genuine look into the Pentagon’s handling of the war. We can retain some measure of caution and doubt, but let us proceed on the presumption of their authenticity and think on what we can learn from them.

Ukrainian Force Generation

The most significant implication of the documents is simple: Ukraine’s combat power is significantly degraded, and in particular their mechanized units and artillery forces are in very rough shape.

The relevant material here in particular is a page entitled “US Allied & Partner UAF Combat Power Build”, which details the force generation, training, and equipment tranches that will create the mechanized package which Ukraine will use in its spring offensive. The plan calls for a force of twelve nominal brigades, nine of which will be equipped by NATO and three internally generated by the Ukrainians. The leak does not offer insight into the three Ukrainian brigades, but the intended complement of the nine NATO brigades is meticulously listed).

All told, the combat power build calls for these brigades to field a total of 253 tanks, 381 Infantry Fighting Vehicles, 480 Armored Personnel Carriers, and 147 artillery pieces. This implies that these will be brigades in name only, and will in fact be far understrength. Parceling these systems out across nine brigades will give an average strength of a mere 28 tanks per brigade, along with some 95 IFVs/APCs and 16 artillery tubes. Compare this to a US Army Armored Brigade Combat Team, which would have almost 90 tanks and almost 200 IFVs/APCs. An American Stryker Brigade (a lighter, rapidly deployable formation) would have about 300 Strykers – the Ukrainian 82nd Brigade is listed to receive only 90.

In combat power terms, therefore, these new brigades are going to be far understrength. Their tank strength, far from being full brigade level, amounts to less than an American armored battalion.

Another key aspect of the force build document is the training schedules. This document dates from the beginning of March, at which point five of the nine brigades were listed at “Training 0% Complete”. Only one of the brigades was more than halfway trained, rated at 60% complete. Despite this, six out of nine were scheduled to be ready by the end of March and the remainders by the end of April. This can only be achieved with significantly truncated training times, and these are detailed in the document. Leopard tank training, for example, is listed at only six weeks. Just for context, American tankers can pencil in 22 weeks of training for the Abrams.

The overall picture, therefore, is rather foreboding for Ukraine. The leaked documents do not give us insight into the three brigades that Ukraine is expected to generate with their indigenous assets, but the nine NATO trained and equipped brigades are slated to be significantly understrength and manned by personnel who are receiving a hugely accelerated training course. These brigades will almost certainly need to be deployed in groupings to be capable of the requisite combat tasks.

An ancillary but important note at this point is the fact that, as best we can tell from these documents, Ukraine’s prewar tank park is almost completely gone. Ukraine went to war with about 800 of its workhorse T-64, but the NATO combat power build notes only 43 now on hand. There are others, of course, that are currently being operated by Ukrainian frontline units, but the build plan indicates that Ukraine has virtually none in reserve to equip this vital attack package, on which all their hopes will depend.

Meanwhile, a separate element of the leak paints a similarly dismal picture of Ukraine’s ranged fires. Buried on a page marked “NOFORN” – which means No Foreign Nationals, even allies, are supposed to see it, is a logistics table showing 155mm shell deliveries and expenditures. This bit is rather shocking.

We have known for quite some time that Ukraine is facing a critical shell shortage, but the leaked documents reveal just how acute this issue is. Ukraine’s usage rate is very low right now – the report claims only 1,104 shells had been expended in the previous 24 hours – compare this to the 20,000 or so shells that the Russian army is firing on a daily basis. Even more alarming for Ukraine is the note that they have only 9,788 shells on hand.

Even with a low burn rate that leaves the AFU massively outgunned, they have enough on hand to sustain combat for a little over a week, and they rely on a trickle of deliveries from the USA to keep these stocks stable. The report noted a shipment of 1,840 shells departing in the next 24 hours. Batches of this size are obviously insufficient for Ukraine to build up its stocks, and can only serve to backstop and replenish daily expenditure. There is no possibility of America quickly ramping up the size of these deliveries, because a mere 14,000 shells are produced per month. US officials hope to get this number up to 20,000 this year, but this is still below Ukraine’s current burn rate.

The implication is pretty straightforward. Ukraine is on a shell ration that leaves it unable to offer more than token fire, and it will likely have to live with this shell ration for the duration of the war.

The overall picture of Ukrainian combat power is atrocious. Their overall combat effectiveness faces a hard ceiling due to systemic shell shortages, and the mechanized package slated for the spring offensive is going to be far less potent than advertised. Those nine NATO-created brigades will have the striking power equivalent of (if we are being generous) perhaps four genuine full strength brigades, augmented by three internally generated Ukrainian brigades of dubious quality. Ukraine’s hopes for a glorious assault on the Russian land bridge to Crimea will rest on, at most, 400 tanks and perhaps 30,000 men.

Should this force dash itself to pieces against the well prepared Russian forces in the south, an important question would present itself. If this was the best force that NATO could generate for Ukraine, what will the second team look like? Will there even be another force? This understrength and undertrained mechanized package may be Ukraine’s last serious roll of the iron dice.

The American Analytic Framework

While the leaked documents certainly do not paint an encouraging picture of Ukraine’s force generation, they also offer a similarly shocking glimpse into the state of American military intelligence.

One of the things that immediately jumps out when one looks at the operational reports (the pages showing detailed situation maps) is that the Pentagon apparently has far more information on Russian dispositions than on Ukrainians units. Russian units are strongly accounted for – their locations are precisely marked, unit designations are identified, there are assessments as to which Russian units are combat capable or not, and there are very specific estimates of Russian frontline strength (IE, 23,250 men on the Zaporizhzhia axis and 15,650 men on the Kherson axis).

In contrast, Ukrainian units are not given combat capability designations, their locations are more generally indicated, and there are huge ranges on the assessed manpower (10,000 to 20,000 men on the Donetsk axis – an enormous margin of error!) This, incidentally, is another reason why I think the documents are genuine. If the intent was to put forth disinformation to confuse or deceive the Russians, one would expect actionable (but fake) intelligence about Ukrainian deployments – yet there is no such thing here. Ukrainian strengths and dispositions are presented vaguely and inconclusively, so the only thing the Russian army might extrapolate from this report is that the Americans don’t really know what’s going on with Ukrainian forces.

Indeed, this is the inescapable conclusion. The Pentagon does not seem to have a strong sense of Ukrainian unit strength, location, or activities. They also list their assessed Ukrainian KIA at a mere 16k-17.5k. This is an absurdly low number – where could they have gotten it? In fact, it is a direct copy-paste of the casualty numbers reported publicly by the Ukrainian Ministry of Defense.

The fact that the Pentagon does not seem to have any independently generated intelligence about the Ukrainian army is shocking. They seem to have be relying on Ukrainian propaganda numbers and publicly available deployments data, like the open source Deployment Map. For the record, this is not a knock on the Deployment Map site – I use this resource frequently and find it very useful. The point, of course, is that the Pentagon, with its nearly unlimited resources, does not seem to have any unique insight or intelligence streams of its own in this regard. They gesture vaguely at the map and mutter, “there’s probably a brigade or two in this area, maybe 8,000 men. Or 4,000. We don’t really know.” In fact, all of their axis strength assessments for Ukraine have a 100% margin of error (that is, the upper limit of the range is double the lower limit).

One can only conclude that the tail is wagging the dog. The Ukrainians are able to extract material, training, and cash from the west, but there is little accountability or honest information flow in return. There were hints of this earlier in the war – that Ukraine is a sort of black box which sucks in resources but does not communicate honestly in return; American officials have complained (and Ukrainian leaders have confirmed) that Kiev simply does not tell DC all that much. Apparently this remains an issue well over a year into the conflict. One particularly alarming footnote in the leaked documents states:

“We have low confidence in Russian (RUS) And Ukrainian (UKR) attrition rates and inventories because of information gaps, OPSEC and IO efforts, and potential bias in UKR information sharing.”

Good grief.

One other issue is the Pentagon’s estimate of Russian vehicle losses. It seems that here too they are copy-pasting external estimates. In this case, they appear to be using the “documented” vehicle losses from the Oryx project. Oryx is… interesting. In theory, they are tabulating visually documented equipment losses, which sounds very scientific and hard to dispute. Furthermore, the sheer mass of pictures they have accumulated is something of a verification deterrent – nobody really wants to sort through thousands of pictures and keep score.

However, Oryx has been audited and found wanting. There are a variety of issues that cause them to overcount Russian losses, in some cases drastically. These include double counting (multiple pictures of the same vehicle), wrongly identifying Ukrainian vehicles as Russian losses, counting as lost vehicles that have no apparent damage, accepting images that have obviously been photoshopped, and so on. In one particularly egregious case, a picture of a Ukrainian Msta howitzer had its crew photoshopped out and was marked as destroyed Russian artillery piece. I mean, look at this:

According to Oryx this is a destroyed Russian howitzer, and not a very badly photoshopped Ukrainian gun. Please ignore the conspicuous shadow.

The issue is essentially that Oryx gathers data passively, by having people on social media send them pictures, which they then look at and mark as verified losses. Social media, however, has a pro-Ukrainian bias which leads to a flood of allegedly destroyed Russian vehicles coming in, and Oryx seems to have a weak filter that uncritically verifies almost all of these claims. As a result, Russian losses are drastically overcounted, and Ukrainian losses are undercounted.

Okay, so what? Let Oryx run their little counting project, no harm done – right? Apparently not. The Pentagon’s leaked documents claim 6,000 assessed vehicle losses as of March 1, which lines up with Oryx’s claims (now up to 6,486 destroyed vehicles as of April 10). This is a strong data point confirming suspicions that the American Defense Department is increasingly outsourcing intelligence to OSINT (Open Source Intelligence). It is fairly clear at this point that there is an incestuous amplification between OSINT and the American defense and political establishment. When Oryx counts absurd photoshop hack jobs as destroyed Russian hardware, this becomes a meaningful data point feeding the Pentagon’s battlefield assessments.

It would seem that, much like in the case of Ukrainian force generation and losses, the Pentagon simply does not have any sort of robust or meaningful insight of its own. There would seem to be no independent intelligence streams at work here – only a mindless regurgitation of Ukrainian MOD propaganda numbers and dubious open source projects like Oryx. The American military increasingly seems to be a hollowed out simulacrum of its past glories, decaying behind a façade of shiny machines and bloated budgets – a trillion dollar technobureaucratic jobs program coasting on the residual patriotic fumes of red state American boys.

It has long been apparent that the Kiev regime has no real plan, no firm path to victory, and only a tenuous and unfriendly relationship with reality. Far more terrifying is the thought that the Pentagon is much the same.

Air Defense at the Brink

One last major revelation from the leak is the greatly degraded state of Ukrainian air defense. Very simply, Ukraine is quickly running out of munitions, especially for its critical S-300 and BUK systems, and it can only endure two or three more wave strikes before breaking completely.

Air defense systems can be complicated to talk about for people who aren’t familiar with the nomenclature. This is because there are a large number of different systems required for a modern air defense, which must be “layered” with different systems that intercept targets at various altitudes, phases of flight, and trajectories. The conversation can quickly become even more muddled because the launch systems have both a Russian designation and a NATO designation, and their munitions have different designations still – just for example, the air defense system which the Russians call the S-300 is designated the SA-10 by NATO, and it fires a variety of different interceptor missiles which have their own names, like the 9M83. Multiply this by the many different types of air defense systems currently in use in Ukraine, and you can see how it can easily decay into a morass of acronyms and serial numbers.

In any case, the key thing to understand about air defense systems is the layering aspect – if one node in the layer fails, not only does one lose full spectrum coverage, but the burn rate on the remaining systems increases because they are now bearing an undue load. Ukraine is now almost completely out of interceptors for the S-300 and BUK systems, which comprise almost all of its medium to long range defense. At the current burn rate, they are projected to run out by the first week of May and have had to make hard choices about where and what to defend. There is no prospect of acquiring more interceptors for these systems because they are manufactured in Russia.

To backstop these capabilities, NATO has been rushing its own systems to Ukraine and providing crash course training. What is notable, however, is that NATO is opting to send Ukraine new systems. Germany, for example, sent Ukraine four brand new IRIS-T systems in October. This was a cutting edge weapon in its first run out of the factory. The downside, of course, is that because it is new, there are no deep stockpiles of munitions from past production runs to call on – therefore, surprise surprise, the leaked Pentagon documents claim that Ukraine is already out of IRIS-T interceptors.

The leak furthermore revealed that Ukraine will be outfitted with two newer NATO systems – the American-made Patriot PAC-3 and the Aster 30-SAMP/T (I apologize for this horridly long designation, but I didn’t name the blasted thing) which is a joint Italian-French creation.

Here’s the issue. The US Department of Defense only purchases 230 PAC-3 interceptors per year, and the new procurement schedule does not ramp this number up at all. The Aster system is just now coming online, and Italy and France have contracted to have 700 missiles delivered in the coming years.

What all of this means is fairly straightforward: the Pentagon’s plan to shore up Ukrainian air defense will force NATO to dip into its own stocks very soon, and we will see the artillery situation repeated with air defense interceptors. There simply is no surplus or large scale production to tap into to supply Ukraine; they can only be propped up by directly eating away NATO’s own stocks. All of this occurs at the same time the Russian Air Force is becoming more and more assertive, using new glide bomb conversion kits to deliver colossal FAB bombs from safe distances.

Conclusion: Asleep at the Wheel

At first glance, the worst thing about this remarkable leak is the fact that it happened. This is a bewildering and embarrassing breach; an American citizen seems to have simply walked out with highly classified documents, which were then permitted to sit on a Minecraft Discord server for a month without anyone being the wiser. One must wonder how, and perhaps even more importantly why someone would do this.

Yet the leak as an act of subterfuge or treason is less significant than what the documents show. They show a conspicuous lack of alertness or long-range planning on the part of the Pentagon. American leadership seemingly has to contend with Ukraine as a black hole which sucks in money and munitions and gives nothing back; there is no strong sense of Ukrainian frontline strength, losses, or planning, and the Pentagon seems to lack any sort of independent intelligence streams.

Meanwhile, the material situation in Ukraine is degrading rapidly. Their artillery arm is running on fumes, with a miniscule shell ration and no reserve stocks to speak of, fed by a trickle of deliveries from the USA. Air defense is similarly worn thin, and the plan to repair this crucial umbrella threatens to quickly become vampiric and drain NATO interceptor stocks. The entire strategic logic of Ukraine has reversed. Rather than becoming a cheap way to drain the Russian military, NATO finds itself drawing down its own stocks to prop up the hemorrhaging Ukrainian state, with no clear endgame in sight. The proxy has become a parasite.

There does not seem to be any long term plan to sustain Ukraine’s war. The Pentagon’s procurement plans do not indicate any real intent to ramp up production of key systems. For FY2024, they have ordered a modest 5,016 GMLRS – the missiles launched by the famous HIMARS system. Ukraine has already fired nearly 10,000 GMLRS, making this yet another system where Ukrainian expenditures vastly exceed supply.

To salvage the situation, Kiev must place its hopes on one desperate dice roll with a mechanized attack package comprised of half-strength brigades wielding a disparate inventory of different vehicles and systems. This Frankenstein’s monster of armies – sewn together with a bevy of different tanks, IFVs, APCs, and artillery systems drawn from all corners of the NATO alliance, will likely be asked to smash through the heavily fortified and robustly manned Russian lines in the south, where it will be pulverized and become only so much more mulch for the Pontic Steppe.