Thread by @KofmanMichael

An update on the war following a recent trip. Ukrainian forces are holding, but the situation has worsened since July due to mounting offensive pressure. Here I cover some of the negative and positive trends, along with the salient dynamics at the front. Long thread. 1/

Drones continue to be responsible for most daily casualties, with the front line defined by overlapping drone and artillery fire engagement zones 20-25km from the forward line of troops. This is commonly referred to as the ‘kill zone.’ 2/

One of the key dynamics this year has been a tug of war contest between Russian forces and the AFU over superiority in this kill zone, and its relative footprint. That in turn determines initiative on the ground, and the cost incurred in offensive, or defensive ops. 3/

Russian forces have been steadily shifting the zone this year. Ukraine held superiority and its squarely over Russian lines. Now the engagement zone is much more evenly situated over Russian and Ukrainian forces, reducing Ukraine’s previous advantages in drone employment. 4/

Drone units work to suppress and displace the opponent’s drone crews further from the front. Artillery is commonly used to strike closer drone launch points at 3-5km, while longer range drones strike artillery, logistics, and enemy drone teams further in the rear. 5/

On the Russian side Rubicon formations remains a leading problem for drone operators, not only the drone companies themselves, but because they train other Russian drone units to replicate their approaches focused on AFU logistics, drone crews, and intercepting ISR. 6/

One of the observed changes is the balance of casualties in the AFU has shifted from infantry to supporting roles, drone operators, logistics, etc. There is very little infantry forward, and in many AFU brigades infantry now bears less of the casualties. 7/

The AFU continues to adapt. For example through widespread use of UGVs for logistics, and in some assault roles. Most units now have a UGV platoon, company, or battalion. These require greater skill and training to employ, but hold considerable promise, reducing casualties. 8/

The airspace has become even more contested for longer-range ISR, with both sides establishing dense tactical radar coverage to detect drones, and on way attack munitions. Ukraine led in establishing this trend in 2024, Russia has sought to replicate and scale deployment. 9/

Artillery remains important to suppressing enemy forces and shaping how they attack, especially in bad weather, which is more prevalent this time of year. Fog, wind, and rain significantly degrade drone operations, enabling Russian infiltration through Ukrainian positions. 10/

Overemphasis on drones overlooks that the current dynamic is due to a combination of mining, use of drones, and traditional artillery fires. Hence maintaining adequate supply of arty ammo for Ukraine, especially longer-range munitions remains essential. 11/

There is a divergence in approaches being employed by the AFU to conduct counterattacks. Some units employ costly tactics akin to Russian assault units, others employ traditional combined arms supported by drones, and others emphasize drones in combination with infantry. 12/

Successful Ukrainian assaults benefit from systematic reconnaissance, suppression of enemy drone crews and artillery, along with effective coordination with the assaulting infantry. These elements can be seen in the AFU counterattacks by Kupiansk. 13/

On the ground the war continues to be defined not by trenches and cohesive defensive lines, but by 2-3 man positions with large gaps in between. These pickets are neither firing positions or observation points. The front is porous, presence does not confer control. 14/

Russian forces conduct infiltration by splitting platoons into 1-2 man groups trying to penetrate in between Ukrainian positions. They seek to advance as far as possible past Ukrainian positions, entrench, then accumulate. Many are killed but some make it through. 15/

The net result is a sizable grey zone, where nobody is sure who controls what, making the battlespace difficult to map. Russian and Ukrainian positions are comingled in cities, and tree lines. Hence the situation could be much better or worse than it ‘looks on the map.’ 16/

Russian KIA casualty rate has gone up considerably this year, but WIA decreased as a ratio because survival rates are much lower. Wounder are much harder to recover within the kill zone. Russian gains this year came at a steep price of unrecoverable personnel losses. 17/

Russian unrecoverable casualties (25k+) have been inching up towards their monthly recruitment rate (30-35k) such that in 2025 they had to focus on loss replacement as opposed to force expansion. The quality keeps going down. Desertion has also gone up within Russian ranks. 18/

Russian forces' approach allows them to maintain pressure, grinding their way through the front, but it does not generate momentum or enable operationally significant breakthroughs. This is why the ‘summer offensive’ is more of an all-year affair. 19/

That said, the winter is not conducive to infiltration tactics. There is no cover, Russian air support & UMPK strike rates drop off, and it is more difficult to advance. Hence the front may become more stable this winter, as was seen last year. 20/

Bad weather enables, and constrains, both sides. Ukraine can conduct rotations, replenish, and reinforce. However, it tends to favor Russian forces more right now because they are on the offensive, and Ukraine’s defense is largely based around the drone engagement zone. 21/

More recently Russian forces have renewed mechanized assaults. These are largely ineffective. Even so, ‘assault’ tanks take 30-40 FPV drones to stop, if not more (70 in one recent case). Mines remain the best way to immobilize armor and halt such attacks. 22/

Looking at the front, the main problem areas are Pokrovsk & Kostiantynivka, the area running from Huliapole to Pokrovske at the Zap-Dnipro-Donetsk border, and area surrounding Lyman where Russian forces slowly advance towards Sviatohirsk and the Izium-Sloviansk road. 23/

On Pokrovsk: The situation worsened over time as Russian forces kept infiltrating through AFU brigades holding the southern part of the city. Ukrainian positions grew increasingly thin. Worsening weather enabled Russian troops to get more men into the city in recent weeks. 24/

If Pokrovsk falls so does Myrnohrad, and the pocket closes. The immediate result is Russian forces will operate drones from Pokrovsk, using the city to displace AFU units another ~15km or more. Ukraine has defensive lines to withdraw to, and can reset the defense. 25/

Ukraine has brought in reinforcements to stabilize the flanks, but this may not be enough given the extent of Russian infiltration at this stage. Counter-attacking at this stage is costly, and has implications for other sectors like Zap, which will be without reserves. 26/

Pokrovsk is not encircled, but the issue is that drones now effect fire control in depths such that it is much more difficult to withdraw equipment, or extract people from a pocket. Hence a late exit akin to Avdiivka is now a more problematic proposition. 27/

The city has operational value. Its loss opens the path for Russian forces to keep advancing towards Donetsk borders, and widen the axis of advance west of Kramatorsk. It does not open those cities to be quickly taken though and so the significance should not be overstated. 28/

Russia’s advance this year has been lackluster, especially along the axes they prioritized. The costs have been high as well relative to the amount of terrain gained. But Pokrovsk can be an unfortunate postscript to that story, especially if it results in unnecessary losses. 29/

Much of the Russian advance has taken place further southwest at the Zap-Dnipro-Donetsk border, and has accelerated in recent weeks. Ukrainian defenses around Huliapole look increasingly incohesive. This area appears to be a lower priority for Russia, and Ukraine. 30/

A lack of reserves means AFU must take a firefighting approach. Counterattacking in Pokrovsk means few forces are available to stabilize the situation anywhere else, like Zap. Hence Russian forces might make increased gains along that axis. 31/

However, Russian forces continue to underperform tactically, and fall short of their objectives. This in part due to rife cynicism and falsification of information within their system, with units claiming positions and advances that never took place. 32/

Russian forces reduced equipment expenditures in 2024, because such attacks were ineffective, trading them for a steep increase in manpower expenditure which is not sustainable at this rate. In 2026 Russia will face the problem of its growing casualty vice recruitment rate. 33/

The strike campaigns are also an important factor. Russia has been targeting gas production, storage pumping, and seeking to further fragment electricity distribution. Ukraine faces a cold winter with electricity rationing across major cities. 34/

Ukraine has made strides in deploying new types of cheap drone interceptors, with many showing promise for scaled deployment to air defense units. The number of Russian OWA drones intercepted cheaply via these means continues to grow. 35/

What Ukraine also needs is a large, relatively cheap, set of ‘middle-strike’ options in the 30-300km range. Ukraine continues to have a capability gap beyond 30km, with many units sitting on targets they can’t engage in the Russian rear, especially 100-150km behind the FLOT. 36/

Manpower remains one of Ukraine’s biggest problems. Mobilization is hampered by corruption. AWOLs sap combat strength more than casualties, a worsening issue. The approach to force management, and force generation, leaves much to be desired. 37/

Bottom line: the coming months will prove difficult, and the situation will worsen before things improve or stabilize. But with sustained support, and pressure on Russia, Ukraine can stabilize much of the front over the winter, and is positioned to continue the fight into 2026.

There's a fair amount I couldn't include in this thread as it was already rather long. If you've made it this far - thanks for reading. The maps are used are from DeepStateMap. deepstatemap.live/en#12/48.31048…

For those interested in a past thread on this subject in July, you can find it here:

https://x.com/KofmanMichael/status/1950998968416215459?s=20