Jump to content

How Hot is Ukraine Gonna Get?


Probus

Recommended Posts

7 minutes ago, danfrodo said:

The 154K troops -- is that dead or casualties?  I know is dumb question but I am confused because I see such widely varying numbers and most sources don't say what the number means.

According to the matching charts at Ukraine War Room on Reddit, that's Russian dead. The format of the charts there are different, but the numbers are the same across the board.

Edited by Splinty
Link to comment
Share on other sites

Just now, Battlefront.com said:

When looking at the broader scope of combat power, there is no doubt about it.  At the beginning of the conflict we created a rather lengthy list of factors that gave Ukraine's smaller physical presence on the battlefield a significant boost against Russia's larger presence.

Morale is the big one.  Doesn't matter how much stuff you have if your soldiers abandon it and head back home because the other guy's soldiers are fighting like demons.

Another early example was Ukrainians "phoning in" intel to local units and/or national "tip lines" that the Ukrainian MoD set up.  The example of a Ukrainian refugee seeing a video of his house occupied by Russians calling such a line and asking for his house to be smashed to pieces is never far from my mind.  Russia is able to do some of this through traitors and agents, but not to the extent Ukraine demonstrated.

Oh, and then there's the usual situation where the defender knows the terrain better than the attacker.  This was made worse for Russia because they went in with a lack of maps and many that they did have were hopelessly out of date.  I remember an interview with an old guy who was asked by some Russians were X village was and they were already in it!  He helpfully pointed down a road leading out of the town and the Russians thanked him and sped off away from their objective.

Now, couple this with the flip side of the equation which are the things which Russia did to reduce the effectiveness of its own forces.  Tires that shredded, soldiers with terrible training, open channel radios, inadequate infantry staffing, ships that couldn't defend themselves, no doctrine for meaningful air operations, politically dominated military decision making, etc.

What this means is on the one hand Ukraine was able to fight far better than an accountant would guess at, on the other hand Russia fought far worse than what the accountant would guess at.  This is, in a nutshell, where the pre-war analysts fell down hard.  Combat Mission guys, on the other hand, better understood these sorts of factors.  It's why nobody wants to take King Tigers into forests or fight with Conscripts against Veterans.  The details matter.

Steve

This whole line of thinking got me onto the idea of information mass.  The old rules of physical mass have largely fallen part in this war in may instances.  But perhaps mass still applies but has been offset into another domain.  I would bet a pretty large wager that the UA has got an information-mass advantage right now.  If we measured the raw data coming in (e.g. bits), being analyzed and turned into information, and then integrated into knowledge leading to learning advantage the UA looks like the Colossus, not Russia.  Further information is much harder to attrit.  You either attack the repositories (nearly impossible in this day and age) or you let time render it less relevant.  All you can really do is attrit the mechanism of information collection and analysis/processing, and the RA simply cannot do this.

So if information mass counts as much as bullets on the modern battlefield, it may go some way to explaining why the UA has crippled a far physically larger foe.  Mass may still very well matter but just not how we traditionally think about it. 

Link to comment
Share on other sites

@sburke

Major Aleksandr Bondarev, chief of flights safety, senior inspector-pilot, 559th bomber aviation regiment (Morozovsk airfield, Rostov oblast), 1st mixed aviation division, 4th AF/AD Army of Southern military district. Was killed on 3rd of March 2023, when his Su-34 was shot down over Yenakiyeve, Donetsk oblast. Second crewman survived but was heavy injured. UKR command counted this victory for 301st AD regiment (S-300), but eyewitnesses claimed the bomber became a victim of friendly fire from other Russian jet, who flew behind the order of Su-34 and launched several missiles, one of which hit the bomber. 

 

 Lt.colonel Mikhail Yermolin, commander of MLRS battalion of 126th coastal defense brigade, 22nd Army Corps of Black Sea Fleet coastal forces, Southern military district. Necromancer recently issued his documents and allegedly his body, but now is official confirmation from Russian social media with a data and place of death. He was killed on 2nd of March 2022 during Russin attack on Voznesensk. The body at lest up to the winter was kept in refgrigerator in Ukraine.

 

Edited by Haiduk
Link to comment
Share on other sites

9 minutes ago, The_Capt said:

This whole line of thinking got me onto the idea of information mass.  The old rules of physical mass have largely fallen part in this war in may instances.  But perhaps mass still applies but has been offset into another domain.  I would bet a pretty large wager that the UA has got an information-mass advantage right now.  If we measured the raw data coming in (e.g. bits), being analyzed and turned into information, and then integrated into knowledge leading to learning advantage the UA looks like the Colossus, not Russia.  Further information is much harder to attrit.  You either attack the repositories (nearly impossible in this day and age) or you let time render it less relevant.  All you can really do is attrit the mechanism of information collection and analysis/processing, and the RA simply cannot do this.

So if information mass counts as much as bullets on the modern battlefield, it may go some way to explaining why the UA has crippled a far physically larger foe.  Mass may still very well matter but just not how we traditionally think about it. 

So, on this point, and as per the discussion earlier.

3:1 is a useful fiction, which derives from Lanchester's laws. Specifically, it's the break-even point - it's the stage where you're going to achieve parity in attrition. You will wipe out the opposing population, and lose one third of your number. Lanchester's laws are not a good model for anything in reality, but they were the first to really do it - and they do offer a simple approximation that isn't a bad place to start from.

That's... sort of it. It isn't an idea that has any more weight than that. Specifically, "mass" has never referred to raw numbers, outside of a highly abstract piece of maths. Mass should, and has always, contained more than just numbers - it has to mean "combat effectiveness", with all that implies, outside of abstract situations where you're lining up clones with identical weapons.

So yes, there will be information advantages. There will be motivational advantages. There will be terrain advantages. All of the above and more.

Does this equate to a "force ratio"? Well, that idea is still pretty suspect. It's a potentially powerful tool for historical analysis, and a highly useful gamism, but it's generally better as a post-hoc justification when applied to the real world.

Link to comment
Share on other sites

25 minutes ago, Vanir Ausf B said:

Kofman was far more pessimistic at the time about Russia's prospects than you are remembering.

 

Well, it's still showing what I've been saying about Kofman.  All he says here is "may fail in the Donbas offensive in the coming weeks".  He was still couching things in a way that was favorable to Russia.  I understand that he was already starting to change his earlier tune, but he is one of the foremost experts on this stuff and yet he just didn't seem to willing to frame the possibilities in a way that was, in my opinion, consistent with the reality we were seeing.

Now, contrast Kofman's assessment with what we came up with here after a detailed analysis of this offensive might wind up looking like.  We took a look at terrain, forces, objectives, etc.  Hell, we even saw MS Flight Simulator used as a terrain analysis tool to complement some excellent map/force studies. 

This analysis was not favorable to Russia's possibilities of success.  Focusing on my take on the collective discussion, I said this on the same day as Kofman:

"Whatever the case might be, something is different today than the previous 4 days.  Different is often a sign of change.  I've plopped down my guess ("the offensive got cancelled"), though I don't rule out other possibilities including a temporary pause for some reason (shortage of ammo, ground conditions, etc.). "

https://community.battlefront.com/topic/140931-how-hot-is-ukraine-gonna-get/?do=findComment&comment=1923298

I'm not only saying that I don't think Russia is going to come out on top, I am speculating that they got the stuffing kicked out of them during the preliminary phase enough that they cancelled the whole offensive!  Kofman, by contrast, is saying he still needs a couple of weeks to figure out how it went.

On the 18th of April I tried very hard to give Russia the benefit of the doubt that mass would win them a chunk of territory.  This is the best outcome I could picture:

"From the early reporting it would seem that this offensive is, as expected, going to be very costly for the Russians even if they should ultimately succeed."

 

Here's a recap from April 29:

 

To finish this post off, what I'm saying about Kofman (and others) is that even after they adjusted to being flat out wrong about the war when it started, they were slow to fully admit where things stood and very reluctant to project forward.  Analysts are supposed to do these sorts of things, especially ones that get paid for such work.


Steve

Link to comment
Share on other sites

Earlier, I translated a description of current RU "Meat Assaults". According to the context, it is from the Vuhledar direction. The obvious question is - what is going on in different directions? This is a comment from another RU propagandist associated with RU regular soldiers fighting in the Svatove-Kreminna direction:

Quote

Dear brother Vladlen Tatarsky talks about the assaulting of enemy positions by our troops. The same situation is practically everywhere, except for the Wagner PMCs...

 

Edited by Grigb
Link to comment
Share on other sites

14 minutes ago, domfluff said:

Does this equate to a "force ratio"? Well, that idea is still pretty suspect. It's a potentially powerful tool for historical analysis, and a highly useful gamism, but it's generally better as a post-hoc justification when applied to the real world.

For sure it is super tricky.  Military planners have been trying to quantify these sorts of factors for a long time and haven't succeeded.  Yet we know it's all extremely important!

Imagine trying to design a new aircraft.  You know all the physics concepts of what makes a successful design, but you don't know any of the math or how to evaluate one design over another.  The only way to know if it flies or not is to build it, launch, and hope for the best.

IMHO it is better for militaries to take some sort of guess at the impact of "soft factors" than to ignore them in their calculations.  That is, in a nutshell, what Combat Mission is all about.

Steve

Link to comment
Share on other sites

Just now, domfluff said:

So, on this point, and as per the discussion earlier.

3:1 is a useful fiction, which derives from Lanchester's laws. Specifically, it's the break-even point - it's the stage where you're going to achieve parity in attrition. You will wipe out the opposing population, and lose one third of your number. Lanchester's laws are not a good model for anything in reality, but they were the first to really do it - and they do offer a simple approximation that isn't a bad place to start from.

That's... sort of it. It isn't an idea that has any more weight than that. Specifically, "mass" has never referred to raw numbers, outside of a highly abstract piece of maths. Mass should, and has always, contained more than just numbers - it has to mean "combat effectiveness", with all that implies, outside of abstract situations where you're lining up clones with identical weapons.

So yes, there will be information advantages. There will be motivational advantages. There will be terrain advantages. All of the above and more.

Does this equate to a "force ratio"? Well, that idea is still pretty suspect. It's a potentially powerful tool for historical analysis, and a highly useful gamism, but it's generally better as a post-hoc justification when applied to the real world.

So we still use force ratio/rules of thumb in planning but it is a start point, not the final answer.  For example, if we are planning to attack an opponent 3 times your size with rough parity of combat power, well you had better have some offsets.

I think what is breaking the calculus in this war is the level of impact information advantage is having.  It is followed up by how that information is being generated. 

Another key aspect is qualitative factors.  I spoke to an old op research guy from back in the day and he noted that the US calcs for the Gulf War did not take into account qualitative factors, while the UK did...and the UK numbers were far more accurate in the end (he was a Brit himself, so there is that).   Analysis of this war did take into account some qualitative aspects (the mighty BTG) but failed to really grasp information quality as a thing.

Back to hard force ratios.  Yes, I went through War By the Numbers  as a phase back a few years ago.  But what became apparent is that large sampling of battles and their force rations/losses is really just one apples-to-oranges-to-Tuesday exercise.  For example, the only way to do some math on Bakhmut is to take historical battles similar to Bakhmut - but there really aren't any.  One cannot look at the Somme, or Verdun or even Mosul as they all had very different conditions at play.  So you can take 600 engagements and do some math but everyone of those engagements is unique and can skew the numbers for all sorts of reasons.  Bakhmut can only really be compared to similar battles in this war (e.g. Severodonetsk) but as we have seen even in these conditions are really different.  So people built some pretty complex models, and still are but I have yet to see one that can do the job complete justice - outside CM, of course. 

We can see trends of attackers in similar situations and derive some deductions that inform but this all get down to "every battle is the same, every battle is unique" paradoxes.  It is possible that the RA has pulled off 1:1 casualty ratios but given their shortfalls of effective combat power compared to those of their opponent, and the fact they are attacking over prepared ground, it does not really add up.  At best one comes up with some "gut feelings" of how things should be happening and compare to results.  Then spend a decade trying to figure out what happened. 

Link to comment
Share on other sites

5 hours ago, Maciej Zwolinski said:

Also, since a few days the official reported losses by UKR ministry of defence show very significant losses of RU artillery. A solid counterbattery effort - could that signify preparations for a counterattack somewhere?

By itself is unlikely - good CB is extremely helpful for defense against RU attacks as well.

There are signs about UKR offensive preparation. Unfortunately, they are not credible enough to report them. We have to wait. 

Link to comment
Share on other sites

1 hour ago, The_Capt said:

First off China is coming up fast, there is no getting past that.  However it is not there yet, nor does it have global coverage.

So some numbers for context:

https://www.statista.com/statistics/264472/number-of-satellites-in-orbit-by-operating-country/  (obviously not all military ISR)

Better assessment of pair up here:

https://www.rand.org/paf/projects/us-china-scorecard.html

Really long write up here:

https://www.uscc.gov/sites/default/files/2020-05/China_Space_and_Counterspace_Activities.pdf

Bottom line China is not messing around but is still focused regionally.

Backed up by this:

https://www.fpri.org/article/2021/05/chinas-maritime-intelligence-surveillance-and-reconnaissance-capability-in-the-south-china-sea/

Another good short summary here:

25%20MCCABE.PDF

So what?  Well China had a finite number of systems and it is a big-@ss sky.  Solar orbits gives a satellite stability on the region it covers but the earth spins around underneath it.  Satellites (except for geostationary, which are actually pretty far out) are built in constellations designed to keep eyes on specific regions as continuously as possible.  They hand off with other ISR and integrate to give a complete picture.  China is clearly focused on the maritime domain, which also means the ISR they are using may not be optimized for land battle.  China like Russia still has pretty slow refresh rates, as such they are not real-time.  Repositioning those assets toward Ukraine would mean holes in the rest of their system - this is not simply flying one satellite like a balloon over Ukraine.  It would mean shifting entire constellations or launching new one.

Ironically, and definitely not-funny, the use of high altitude balloons in Russia as ISR into Ukraine would be a possible solution if they could control them.  Regardless, China is not there yet with respect to global ISR.  The US is because projecting military power globally is a strategic objective, but even it has limitations.
 

 

 

 

Once you’re in sun-synchronous orbit, global coverage is literally free. It’s actually kind of impossible to make non-geosynchronous satellites that don’t have complete coverage of whatever latitudes they cover- it’s the nature of orbit.  Sun-sync orbits are set so you get the same illumination at the same latitude on every pass.  And most of China’s satellites are in mid altitude high inclination orbits.  Including some SAR that don’t care what the illumination of the earth is- they just use the sun for power.

There are a few that have the highly elliptical orbits that are used for high res imaging, and those probably have perigee over the latitudes of the South China Sea, but most of their stuff is flying orbits that have global coverage at altitudes that can give you reasonable performance.  They aren’t at a level comparable to the US on performance, but they do have global capability that doesn’t suck.

Yes, everybody flies constellations, but they’re really about revisit rate, particularly if you’re doing things that have narrow field of view or can’t look sideways.  

Link to comment
Share on other sites

10 minutes ago, Grigb said:

There are signs about UKR offensive preparation. Unfortunately, they are not credible enough to report them. We have to wait. 

I think, too early. "Western-trained&equipped brigades" not operational yet and mosltly not in Ukraine, though first Leo as if already here. Unless we will start with available forces and "westerners" will attack in second echelone later. 

Link to comment
Share on other sites

3 hours ago, danfrodo said:

I am going to stick to the most obvious answer, RU

2 hours ago, Battlefront.com said:

Anyway, the logic that Russia did the attack is pretty straight forward

Sorry gents, I don't buy it. Nothing is obvious or straight forward about this and frankly it sounds like a very crude rule of three: Russia = evil, blowing up pipelines = evil => Russia = blowing up pipelines. If we are looking for rational motives, Russia really has a weak motive at best. Economically they had exactly nothing to gain from it. Those contracts had pretty large minimum payment clauses, so even if no more gas would have been taken, they would still have gotten a lot of money. Prior to the explosions Russia never just switched the gas off, they always looked for excuses - weak ones but still enough to not be able to legally accuse them of breaking contracts. Given that Europe in general and Germany in particular insistet on honoring contracts, we couldn't very well just cancel payments and short of an official boycott the (privat) buyers would still have to pay. Also, blowing up the pipelines before the winter (when it was still very much in the realm of possibilities that there would be a cold winter and Germany would come back groveling) didn't make sense.

Just going by cui bono US, Ukraine or even the German government (the whole discussion about sanctions pretty much ended at that point) stood to gain much more from a sabotage, however at very high risk. Now, we know Russia doesn't always do the rational or smart thing so there is that. But still, lots of players with a variety of motives here and by no means straight forward.

I think the best and most honest thing we all can say at the moment is: We simply don't know.  

 

Edited by Butschi
Link to comment
Share on other sites

Just now, Haiduk said:

I think, too early. "Western-trained&equipped brigades" not operational yet and mosltly not in Ukraine, though first Leo as if already here. Unless we will start with available forces and "westerners" will attack in second echelone later. 

Yes, it is too early. Rasputitsa is in full swing. Also, if I properly interpret Mashovets hints UKR command believes RU offensive is not done yet. We have to wait.

Link to comment
Share on other sites

The pipelines thru Poland and Ukraine still exist and flow. Makes little sense to silence NS1 and NS2 when easily Russia can reallocate to those two. False flag operations and misinformation also abound and whatever incompetence in the Russian military, i think does not translate the same to their intelligence services and most certainly the public will not get the truth during the war.

Link to comment
Share on other sites

22 minutes ago, Butschi said:

Sorry gents, I don't buy it. Nothing is obvious or straight forward about this..

Clipping out the rest, I can agree with this statement.  There is no clear finger pointing to do in this case.  Everybody had something to gain (yes, even Russia) and everybody had something to lose.  So if we're talking about state actors, SOMEONE did some math and determined the advantage was worth the downside.  Which is why Russia can not be ruled out just because it's not all happy kittens and puppies for it.

There's a bunch of countries I don't think had anything to do with it.  Germany, for one, and the US for another.  Germany was hoping things would blow over and they could go back to cheap gas, this would get in the way of that.  I highly doubt the US did it because it wasn't necessary to get what it wanted and the risk of long term relationship damage being caught doing it would be massive.  Plus, anybody that knows anything about Biden's character would not see this as compatible.

Other nations have similar horribly bad costs to benefit ratios, like Poland for example.  Others don't have the interest level to do something like this even if they had the means.

It seems to me that the non-state actor theory is the one with the strongest legs.  It is exactly the sort of thing a rogue element might do, though it seems unlikely without some form of state assistance.

Steve

Link to comment
Share on other sites

45 minutes ago, The_Capt said:

So we still use force ratio/rules of thumb in planning but it is a start point, not the final answer.  For example, if we are planning to attack an opponent 3 times your size with rough parity of combat power, well you had better have some offsets.

I think what is breaking the calculus in this war is the level of impact information advantage is having.  It is followed up by how that information is being generated. 

Another key aspect is qualitative factors.  I spoke to an old op research guy from back in the day and he noted that the US calcs for the Gulf War did not take into account qualitative factors, while the UK did...and the UK numbers were far more accurate in the end (he was a Brit himself, so there is that).   Analysis of this war did take into account some qualitative aspects (the mighty BTG) but failed to really grasp information quality as a thing.

Back to hard force ratios.  Yes, I went through War By the Numbers  as a phase back a few years ago.  But what became apparent is that large sampling of battles and their force rations/losses is really just one apples-to-oranges-to-Tuesday exercise.  For example, the only way to do some math on Bakhmut is to take historical battles similar to Bakhmut - but there really aren't any.  One cannot look at the Somme, or Verdun or even Mosul as they all had very different conditions at play.  So you can take 600 engagements and do some math but everyone of those engagements is unique and can skew the numbers for all sorts of reasons.  Bakhmut can only really be compared to similar battles in this war (e.g. Severodonetsk) but as we have seen even in these conditions are really different.  So people built some pretty complex models, and still are but I have yet to see one that can do the job complete justice - outside CM, of course. 

We can see trends of attackers in similar situations and derive some deductions that inform but this all get down to "every battle is the same, every battle is unique" paradoxes.  It is possible that the RA has pulled off 1:1 casualty ratios but given their shortfalls of effective combat power compared to those of their opponent, and the fact they are attacking over prepared ground, it does not really add up.  At best one comes up with some "gut feelings" of how things should be happening and compare to results.  Then spend a decade trying to figure out what happened. 

We see occasional news articles about the U.S. helping the Ukrainians run command staff exercises. I suspect theses are more at the operational level than the tactical. It really does seem to me though that there should be a a whole building full of people working on tactical level stuff using Combat Mission Professional Edition, or similar if such exists to sort out tactics for the vast array of stuff the Ukrainians are getting. It is not at all clear to me that the best tactics for a unit equipped with Leopards and CV-90s is the same as the best tactics for a unit equipped with Bradleys and PT-91s. And there are probably five other kit combinations out there. Do the Brits have a worked out tactical scheme for using Challengers with Marders?

Combat Mission Professional Edition is also probably capable of simulating something like the fighting around Bakmuht at very high fidelity, triply so if the people setting it up had access to the AFU General Staff hard drives The_ Capt mentioned. Or is all this just too resource intensive? And staff colleges across the  Western world are going to be doing this for the next five years just trying to figure out what actually happened after the fact?

Link to comment
Share on other sites

SBU detained former commander of 73rd Maritime Special Operation Center 1st rank captain (equal of colonel) Eduard Shevchenko. He was a famous veteran of ATO, personally participated in many combat operations, for example in liberation of Dzerzhunsk (now Toretsk) in 2014, where he and other 33 SOF fighters could hold the town for eight hours of fight with DNR forces, having the armor. He was awarded by state orders and even unofficial award People's Hero of Ukraine. Though, since 2017, between him and part of his unit there were some conflict situations.

Now SBU detained him for betrayal. He turned out agent of GRU and tried to enlist town mayor of Ochakiv (he was his advisor) to work on Russia. Despite the mayor had reputation of moderate pro-Russian man, he told SBU about Shevchenko's attempts to enlist him. Now SBU issued phone calls between Shevchenko and his GRU curator. 

Very likely Shevchenko transmitted information about UKR forces in Ochakiv, so Russian often shell the town from Kinburn spit area.

Former commader of "Donbas" battalion Semenchenko already blamed Shevchenko in elimination of pro-Ukrainain Chechen commander Isa Munayev, attempts on him personally, bringing into ambush units of his battalion etc. As if he many time sent reports to SBU, but Shevchnko as if had strong protection among former UKR authorities and all his appeals just were ignored. Though, since Semenchenko gained scandal reputattion, it's  unknown have his words a true or not.

Зображення

 

Зображення

Edited by Haiduk
Link to comment
Share on other sites

6 minutes ago, Haiduk said:

SBU detained former commander of 73rd Maritime Special Operation Center 1st rank captain (equal of colonel) Eduard Shevchenko. He was a famous veteran of ATO, personally participated in many combat operations, for example in liberation of Dzerzhunsk (now Toretsk) in 2014, where he and other 33 SOF fighters could hold the town for eight hours of fight with DNR forces, having the armor. He was awarded by state orders and even unofficial award People's Hero of Ukraine. Though, since 2017, between him and part of his unit there were some conflict situations.

Now SBU detained him for betrayal. He turned out agent of GRU and tried to enlist town mayor of Ochakiv (he was his advisor) to work on Russia. Despite the mayor had reputation of moderate pro-Russian man, he told SBU about Shevchenko's attempts to enlist him. Now SBU issued phone calls between Shevchenko and his GRU curator. 

Very likely Shevchenko transmitted information about UKR forces in Ochakiv, so Russian often shell the town from Kinburn spit area.

Former commader of "Donbas" battalion Semenchenko already blamed Shevchenko in elimination of pro-Ukrainain Chechen commander Isa Munayev, attempts on him personally, bringing into ambush units of his battalion etc. As if he many time sent reports to SBU, but Shevchnko as if had strong protection among former UKR power and all his appeals just were ignored. Though, since Semenchenko gained scandal reputattion, it's  unknown have his words a true or not.

Зображення

 

Зображення

The fact the bridges weren't blown at Kherson, and Nova Khahovka before the Russians got over them has never been adequately explained...

Link to comment
Share on other sites

4 hours ago, The_Capt said:

The bottom line is that if you look at highly attritional battles against prepared defences losses ratios at the tactical level can get very high - the the opening of the Somme.  However, over time those ratios tend to settle into around 2-1.5 to 1 losses agains attacker…until/if the attacker achieves break out, then the ratio will flip pretty fast. The major weakness of defence is that it is more rigid system, more tied to owning terrain in land battle. Once that system is cracked it can fall apart pretty quickly.  However clearly the UA has not suffered this yet.

What would you say is the reason that loss ratios tend to become less favorable for the defender over time and what can be done to avoid or mitigate this?

A few hundred pages back I brought up the example of the Somme and gave my suggestions for the reasons to be roughly a combination of the following:

1. defensive fortifications degrade/get shot up and lose their effectiveness.

2. defenders have to (big question mark) have to conduct increasingly more risky and potentially desperate counter-attacks to regain shot-up positions which are then easily lost again

3. attackers gain familiarity and experience with the positions they are attacking, the defenders behavior, etc

My own - admittedly simplistic - thoughts on how to mitigate these would be to cede ground but slowly, a few kilometers at a time - unless you are fighting for some key terrain like Kyiv, of course, and not some generic part of french countryside (Somme) or some small city somewhere in Donbas. Instead of keeping on feeding reserves into an ongoing defensive battle and losing them at increasingly worse rates, use them to erect or reinforce new fortifications, a few fields/treelines/ridges to the back and force the attacker to assault fresh and unfamiliar terrain and defensive networks, thereby hopefully keeping your own advantages up.

Of course, giving up Bakhmut now (or even back in December) would mean to cede some rather favorable urban terrain, but as we have seen, most of the fighting of the last weeks took place in much less favorable, smaller villages and fields around the city itself - all fortified positions themselves, but generally not as good defensive terrain as a fighting for the city itself.

Link to comment
Share on other sites

1 minute ago, dan/california said:

The fact the bridges weren't blown at Kherson, and Nova Khahovka before the Russians got over them has never been adequately explained...

In Nova Kakhovka and Kherson cases explaination is one - collapse of defense and control. Nobody could give order, nobody know where to receive such anount of HE, nobody knows where to find sappers who could do this. Nova Kakhovka was attacked by helicopter landing and then arrived VDV on armor. Small National Guard and some army units fought desperately, but they already hadn't time to blow up the bridge. 

Link to comment
Share on other sites

12 minutes ago, Battlefront.com said:

Clipping out the rest, I can agree with this statement.  There is no clear finger pointing to do in this case.  Everybody had something to gain (yes, even Russia) and everybody had something to lose.  So if we're talking about state actors, SOMEONE did some math and determined the advantage was worth the downside.  Which is why Russia can not be ruled out just because it's not all happy kittens and puppies for it.

There's a bunch of countries I don't think had anything to do with it.  Germany, for one, and the US for another.  Germany was hoping things would blow over and they could go back to cheap gas, this would get in the way of that.  I highly doubt the US did it because it wasn't necessary to get what it wanted and the risk of long term relationship damage being caught doing it would be massive.  Plus, anybody that knows anything about Biden's character would not see this as compatible.

Other nations have similar horribly bad costs to benefit ratios, like Poland for example.  Others don't have the interest level to do something like this even if they had the means.

It seems to me that the non-state actor theory is the one with the strongest legs.  It is exactly the sort of thing a rogue element might do, though it seems unlikely without some form of state assistance.

Steve

When it comes to this sort of thing, I'm a big believer in looking at practical questions to test extraordinary claims. For example, is it very likely that non-state actors would be able to work up fake passports, have clear knowledge of what level of boat would require AIS, know something of NATO/Russian patrol schedules, include skilled divers and *underwater explosives experts*? More unbelievably...suppose you've got this cinematic dream team of freelancing, stick-it-to-Putin fire eaters together will all the skillz...are they then the kind of folks to conveniently leave around explosive residue on the yacht they rented?

I don't know who did it. I strongly suspect motive, opportunity and available, checkable OSINT suggests Russia. We'll see.

Link to comment
Share on other sites

Just now, chrisl said:

Once you’re in sun-synchronous orbit, global coverage is literally free. It’s actually kind of impossible to make non-geosynchronous satellites that don’t have complete coverage of whatever latitudes they cover- it’s the nature of orbit.  Sun-sync orbits are set so you get the same illumination at the same latitude on every pass.  And most of China’s satellites are in mid altitude high inclination orbits.  Including some SAR that don’t care what the illumination of the earth is- they just use the sun for power.

There are a few that have the highly elliptical orbits that are used for high res imaging, and those probably have perigee over the latitudes of the South China Sea, but most of their stuff is flying orbits that have global coverage at altitudes that can give you reasonable performance.  They aren’t at a level comparable to the US on performance, but they do have global capability that doesn’t suck.

Yes, everybody flies constellations, but they’re really about revisit rate, particularly if you’re doing things that have narrow field of view or can’t look sideways.  

But it only gives one coverage of one slice of the planet at a time.  You get "global coverage" but only of the same spot once every 12 hours.  Assuming you load that orbit up with enough satellites you get a continuous slice of the planet as it passed beneath.  In order to get real-time continuous, one needs to either load up on geosynchronous, or build multiple solar orbits to get full coverage - I do not even think the US is there yet.

So in order for China to be able to get ISR on Ukraine, it will get a snap shot every 12 hours with those solar orbits, but a lot can happen in 12 hours.  I would bet all the peanuts in the bar the US has got eyes on that part of the world 24/7 using a bunch of platforms, including geosynchronous.

China also has most of its assets pointed regionally so in order to build up a full picture, we are still talking re-vectoring assets away.  So how much does China really love Russia in all this because even trying for half decent parity over the AO is going to cost them.  We know this is not happening, or at least there is no evidence it is - such as Russians suddenly getting a lot better at targeting.   

Link to comment
Share on other sites

Just now, Rokko said:

What would you say is the reason that loss ratios tend to become less favorable for the defender over time and what can be done to avoid or mitigate this?

A few hundred pages back I brought up the example of the Somme and gave my suggestions for the reasons to be roughly a combination of the following:

1. defensive fortifications degrade/get shot up and lose their effectiveness.

2. defenders have to (big question mark) have to conduct increasingly more risky and potentially desperate counter-attacks to regain shot-up positions which are then easily lost again

3. attackers gain familiarity and experience with the positions they are attacking, the defenders behavior, etc

My own - admittedly simplistic - thoughts on how to mitigate these would be to cede ground but slowly, a few kilometers at a time - unless you are fighting for some key terrain like Kyiv, of course, and not some generic part of french countryside (Somme) or some small city somewhere in Donbas. Instead of keeping on feeding reserves into an ongoing defensive battle and losing them at increasingly worse rates, use them to erect or reinforce new fortifications, a few fields/treelines/ridges to the back and force the attacker to assault fresh and unfamiliar terrain and defensive networks, thereby hopefully keeping your own advantages up.

Of course, giving up Bakhmut now (or even back in December) would mean to cede some rather favorable urban terrain, but as we have seen, most of the fighting of the last weeks took place in much less favorable, smaller villages and fields around the city itself - all fortified positions themselves, but generally not as good defensive terrain as a fighting for the city itself.

I think it is actually more simple than this.  If I could boil down the problem for the defender it is to become unsolvable.  Every defence is a military puzzle whose biggest problem is human learning.  If attacked, poked and probed enough, even the most vigorous defence can be solved given enough time.  The game is to make the cost of solving it beyond the bank account of the attacker. 

Even back in WWI the extensive trench systems, communications, rapid firing artillery and railways all conspired to make trench warfare unsolvable.  Sides adapted, inventing airpower, tanks, tunneling and storm troops, all as way to solve for this defensive warfare.  In the end one side simply was exhausted but someone would have solved for that type of defence eventually - we know this because the Germans did in 1940.

So in order to become unsolvable, a defence must become non-linear, adaptive and dynamic.  Big problem in land warfare is that terrain does not work that way.  Problem in air and maritime is that the physics of fluids work that way too well.  I strongly suspect that unmanned systems, particularly ground systems could change this.  Defence/Offence on land may start to look more like that in the sea - at best one can gain temporary control.  Land warfare may be evolving towards denial of ground, the trick will be the right peice of ground at the right time.

Within small wars we see this sort of thing "amongst the people", similarities between people and oceans are interesting.  But unlike oceans, people appear to be able to suddenly freeze from the inside out as opposed to external factors. But this is another topic entirely.

As to Bakhmut, well currently the UA defence is still not solved.  It is solvable, however, the question is how long and at what cost. The RA bizarre inability to learn is very much helping the UA right now, but no party last forever.

Link to comment
Share on other sites

1 hour ago, billbindc said:

When it comes to this sort of thing, I'm a big believer in looking at practical questions to test extraordinary claims. For example, is it very likely that non-state actors would be able to work up fake passports, have clear knowledge of what level of boat would require AIS, know something of NATO/Russian patrol schedules, include skilled divers and *underwater explosives experts*? More unbelievably...suppose you've got this cinematic dream team of freelancing, stick-it-to-Putin fire eaters together will all the skillz...are they then the kind of folks to conveniently leave around explosive residue on the yacht they rented?

I don't know who did it. I strongly suspect motive, opportunity and available, checkable OSINT suggests Russia. We'll see.

But you have to admit, Zelensky hiring Mr.T and the rest of the crew. Would make for an interesting episode of The A-Team. 😀

p7892304_b_v8_ab.jpg

Edited by Harmon Rabb
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...