Ground bombing is borked, part II

[Bullwinkle58]

ORIGINAL: Andy Mac

I dont believe bombing against ground targets is out of kilter now v bases maybe I would agree but if anything CAS is nerfed as FB's die in droves if they try it.

I agree. FBs were a key element in the late war, particularly USMC CAS. After Tinian napalm was used extensively and changed the whole equation, especially in hilly terrain. I've used P-47s extensively in strafe mode since late 1943 and get nothing like the results I know they were capable of from ETO tank buster data. And so on.

In clear terrain level bombers ought to devastate ground forces. Even dug in to slit trenches massed area bombing should hurt LCUs quite a bit in good weather. Level bombing against ports and airfields? I think the airfield damage code works better than the ports. Ports were quite varied in construction, but they're pretty fragile infrastructure, and a lot of results take out supply more than repair facilities. And going back to WITP I tear my hair out that there's no good way to impact CD except to walk up and shoot the gunners in the head with infantry.

[Andy Mac]

Arguably on Iwo Jima the base fell with the fall of hill 365 on D DAy6 art that point 2 of the 3 AF's were under US control......equivalent to taking the base.

The mopping up took far longer but the key point is securing the base so dependfing on your interpretation of when an island falls actually Iwo can be used to argue the point either way

ORIGINAL: Andy Mac

Just remember after 7 or 8 Days I have not secured Christmas Island with total air and naval superity and 6 Divs atackign basically 2.

Folks need to remember that all I have done is secure a badly damaged Air Field and Port the game is slightly all or nothing on combat as its a base based system but all I have done so far is secure the easiest part of the island (the flat bit witt the Air field

How long did it take the allies to secure the AF on Iwo is a better comparison what 10 - 12 days max ?? less....certainly the AF was repaired and in opweration in under 20 days

in my view 400 casualties is light with only 2 squads destroyed for what I brought

[frank1970]

Maybe it would help to simply damage each plane which was used in a low level attack (3000 feet?)? It should be quite easy to hit one of those large planes with a rifle.
The damage should be added after the plane landed so the pilots won´t be lost. Planes which then reach the critical damage should be "permanently disabled".

As there are too many planes in the game this would reduce the number of planes per raid and therefore help the model work better.

[frank1970]

BTW, does artillery fire destroy supplies and planes as well as the runway in a base?
Else it should not be possible to use a airfield on an atoll before the very last enemy artillery/mortar unit is destroyed or disabled.

[rader]

Ok, here's some numbers from the past 3 turns (June 7-12, 1945).

This table shows the number of squads, support units (+ engineers), and guns/vehicles disabled or destroyed by allied aircraft. Note that once devices are disabled, they are subsequently destroyed on the next hit, so we can combine these together provided we double the number of devices for the "div equivalent" row (assumed at 1500 squads per division).

The first three columns show the number of devices disabled/destroyed by 4E bombers. The next 3 colums show the number of devices disabled/destroyed by all other a/c (including 2E bombers and light attack aircraft). The last colum "div equivalents" shows how many division equivalents this relates to being completely destroyed exlcusively from the air.

Some observations:

1. Even though probably 3x or 4x the number of 2E and light bombers are involved here, they are doing less than half the combined damage of the 4E bombers. I.e., per plane, 4E bombers are at least 6 times (and more like 10 times) as effective as ground attack aircraft at killing ground units from the air.

2. The main 4E bomber attack (~200 B-29s in Manchuria) flew 5 out of the 6 turns represented here. This is a very high tempo of operations, and this strike alone was responsible for entirely eliminating over 2 entire full strngth divisions in 5 days.

3. The average rate of loss for these 3 turns is higher than 1 division equivalent eliminated entirely from the air (starting at full strength). Given the size of the Japanese army, if this rate of loss were to continue, Japan's entire army would be eliminated in around 100 days, without any ground combat. If this was possible, why did the allies bother with land units?

4. Damage was significantly modified by terrain, but even in clear terrain, the damage inflicted by real ground attack aircraft (like the IL-2) was moderate to small (~10 squads max per bombing run). By contrast, attacks of 3-6 B-29s (the stragglers), typically did at least double the damage that would be inflicted by 50+ IL-2s. In fact, in this ground bombing model, I suspect that B-29s inflict around 10 times the damage per plane, simply because they carry 10 times the bombload.





Does anyone here actually think a loss rate of over half a division of troops a day due to air attack alone is reasonable (especially in the Pacific theater)?

[Erkki]

Rader, can you put them in relation to sortie numbers, both 4E and 2E? It may be that there are just as many or nearly as many 4Es as 2E sorties which of course effects the results.

[rader]

No, were talking at least 3x the number of 2E and 1E sorties compared with 4Es. However, it is true that most of the 4E sorties were vs. better terain on average (more clear, less jungle). But still, I don't know how to describe the results other than "borked".

ORIGINAL: Erkki

Rader, can you put them in relation to sortie numbers, both 4E and 2E? It may be that there are just as many or nearly as many 4Es as 2E sorties which of course effects the results.

[herwin]

Given the willing to fix things like bombing, what would it take to get ground combat fixed?

[Andy Mac]

How much AA was present, what mode were they in, what height were the bombers, what level of forts did the defenders have ?

[rader]

ORIGINAL: Andy Mac

How much AA was present, what mode were they in, what height were the bombers, what level of forts did the defenders have ?

Generally a couple of AA units are present, but in some cases, just what the divisions had instrinsically. I think the bombers were between 6K and 10K (maybe 8K)?

Combat mode, no forts. But it makes very little difference. While it is true that it is generally better to defend rough terain, you absolutely have to defend clear terrain in some places. For example, it is unreasonable to suggest that you can't put LCS in Japanese home country clear base hexes. If you don't defend them, paratroopers will just come and take them. If you do defend them, you will lose all the LCUs doing so and the paras will still take them. Right now, I am faced with the prospect of abandoning all bases in clear terrain and giving them up to the paratroopers, or getting the defenders killed from the air.

I would suggest that ground bombing troops that aren't in contact with the enemy should be dramatically reduced in potency. Without being able to spot the location of enemy troops for the aircraft, I'm not sure it makes sense that 4E bombers in formation can destroy entire divisons in rear areas each day...

[herwin]

Combat in OCS Burma

Background: Allied divisions have typically 5 steps, while Japanese
divisions have 10. Casualties per step are more comparable at 800 or so.
Turns are 3-4 days and hexes are 5 miles.

Terrain: this is open (rare in Burma), close (jungle or villages), very
close (swamp, town, rough, mountain along roads), and extremely close
(mountain away from roads). River crossings halve the attacker.
Armour/tank/mech units are limited to attacking in the clear and along
roads/tracks/railroads, and their attack/defense is halved or worse in
non-clear terrain.

A 2-level fortification is a hedgehog. Fighting unsupplied halves that
force.

Bombardment: artillery bombardment is expensive--1000 tons of ammunition
for a Commonwealth division bombardment or for a Japanese corps
bombardment. These bombardment levels have about a 50-50 chance of
suppressing a target, which halves its firepower and reduces its
effectiveness, and about an 8% chance of damaging it (a half step). Air bombardment
has about a 30% chance of suppressing the target and a 3% chance of
damaging it.

The attacker takes its losses and chooses to convert options into hex
retreats or step losses. Label an attacker retreat a 'probe', and a
choice of step losses an assault. Then the defender makes the same
decision. Step losses are a 'defence', and a retreat is a retrograde.

Here are some typical battles:

3-1 in clear, probe. 0.313 expected attacker step losses. 10% chance of
a forced defender retreat and attacker exploitation. 0.67 expected
defender step losses.

3-1 in clear, assault, defender defends in place. 16.3% chance of a
forced defender retreat and attacker exploitation. 0.987 expected
attacker step losses. 1.7739 expected defender step losses.

3-1 in clear, assault, defender retrograde. 16.3% chance of a forced
defender retreat and attacker exploitation. 0.987 expected attacker step
losses. 0.67 expected defender step losses. 1.43 hexes retreated.

3-1 in jungle, probe. 0.427 expected attacker step losses. 10.67% chance of
a forced defender retreat and attacker exploitation. 0.426 expected
defender step losses.

3-1 in jungle, assault, defender defends in place. 16.3% chance of a
forced defender retreat and attacker exploitation. 1.07 expected
attacker step losses. 1.507 expected defender step losses.

3-1 in jungle, assault, defender retrograde. 16.3% chance of a forced
defender retreat and attacker exploitation. 1.07 expected attacker step
losses. 0.426 expected defender step losses. 1.08 hexes retreated.

3-1 in rough, probe. 0.572 expected attacker step losses. 6.7% chance of
a forced defender retreat and attacker exploitation. 0.31 expected
defender step losses.

3-1 in rough, assault, defender defends in place. 9.7% chance of a
forced defender retreat and attacker exploitation. 1.32 expected
attacker step losses. 1.287 expected defender step losses.

3-1 in rough, assault, defender retrograde. 9.7% chance of a forced
defender retreat and attacker exploitation. 1.32 expected attacker step
losses. 0.31 expected defender step losses. 0.97 hexes retreated.

For comparison, I would expect a defender retreat (9 OCS hexes in a day)
in WitP-AE for 3-1 in the clear with 2.5 attacker steps lost and 7.5
defender steps lost. For 3-1 in the jungle, I would expect a defender
retreat (9 OCS hexes again) with 2.5 attacker steps lost and 5 defender
steps lost. For 3-1 in the rough, I would expect a defender hold with
7.5 attacker steps lost and 2.5 defender steps lost.

[rader]

I think it's a good idea to have partial hex control. Ground combat in WITP is way too all-or-nothing (unlike what it specifically says in the manual) and dosen't take enough time. If you just have a hex control counter for all hexes (bases/non-bases), and you can attack to increase your control level. Once you get to a certain threshold you capture the base (could have histeresus, so that you need to get to 70% or so, and the other side would then need to get back to 30% to retake it). This way you could have both sides attacking in a hex to try to regain (partial control). Once you get to 100%, the other side retreats. It would take several very successful assaults to capture a hex, and therefore several days of fighting - e.g., your control % would go up by double your assault value or something like that (e.g., a 3:1 would give +6% control).

You could even have partisans decrese your partial hex control, and therefor your supply movement through hexes. E.g., hexes in China could slowly swindle in control from the Japanese, and the Japanese would have to "attack" rear hexes now and then to re-establish control.

[Jzanes]

ORIGINAL: rader

Ok, here's some numbers from the past 3 turns (June 7-12, 1945).

This table shows the number of squads, support units (+ engineers), and guns/vehicles disabled or destroyed by allied aircraft. Note that once devices are disabled, they are subsequently destroyed on the next hit, so we can combine these together provided we double the number of devices for the "div equivalent" row (assumed at 1500 squads per division).

The first three columns show the number of devices disabled/destroyed by 4E bombers. The next 3 colums show the number of devices disabled/destroyed by all other a/c (including 2E bombers and light attack aircraft). The last colum "div equivalents" shows how many division equivalents this relates to being completely destroyed exlcusively from the air.

Some observations:

1. Even though probably 3x or 4x the number of 2E and light bombers are involved here, they are doing less than half the combined damage of the 4E bombers. I.e., per plane, 4E bombers are at least 6 times (and more like 10 times) as effective as ground attack aircraft at killing ground units from the air.

2. The main 4E bomber attack (~200 B-29s in Manchuria) flew 5 out of the 6 turns represented here. This is a very high tempo of operations, and this strike alone was responsible for entirely eliminating over 2 entire full strngth divisions in 5 days.

3. The average rate of loss for these 3 turns is higher than 1 division equivalent eliminated entirely from the air (starting at full strength). Given the size of the Japanese army, if this rate of loss were to continue, Japan's entire army would be eliminated in around 100 days, without any ground combat. If this was possible, why did the allies bother with land units?

4. Damage was significantly modified by terrain, but even in clear terrain, the damage inflicted by real ground attack aircraft (like the IL-2) was moderate to small (~10 squads max per bombing run). By contrast, attacks of 3-6 B-29s (the stragglers), typically did at least double the damage that would be inflicted by 50+ IL-2s. In fact, in this ground bombing model, I suspect that B-29s inflict around 10 times the damage per plane, simply because they carry 10 times the bombload.





Does anyone here actually think a loss rate of over half a division of troops a day due to air attack alone is reasonable (especially in the Pacific theater)?

I am Rader's oppoent in this match and I have a few points to add.

I agree ground bombing vs. units in clear terrain is about 2x too deadly and bombing vs. units in cover is about ½ as effective as it should be. However, I think the situation in Russia reflects a near optimal situation for ground bombing;

1. There has been absolutely 0 Japanese fighter opposition.
2. Japanese AAA is either non-existent or ineffective. Bombers are flying at 6000 feet.
3. Allied bombers are flying out of large (level 8 or 9) airbases with excellent supply, plentiful aviation support, and air HQs are present.
4. Japanese forces are moving and are either in MOVE mode or COMBAT mode with 0 forts when bombed. Having some forts dramatically reduces the effectiveness of ground bombing even vs. units in clear terrain.
5. Generally the weather in Russia has been “good” for the last few weeks. Good weather means anything but thunderstorms basically.
6. Even with optimal supply and support, all heavy bombers are not flying every turn. Most turns, 150-250 B29s fly sorties while another 100-200 bombers stay home for various reasons.
7. Some of the ground units being bombed have already been routed in previous land battles and have lots of disrupted components ripe for being destroyed.

[herwin]

The largest number of defender casualties I would expect to see in a day for a corps versus division battle is about 400.

I have calibration data for the Burma campaign, and they are consistent with the OCS combat model. They are not consistent with the WitP-AE ground combat model.

There's been some interest in calibrating air attacks on troops. In OCS, a Sally or a Beaufighter unit (somewhere in the range of 20-45 aircraft) in Burma has an expectation of causing about 10 casualties per day. A fighter-bomber unit (same size range) can be expected to cause about 3 casualties a day. A Vengence unit, perhaps 15 per day. A B-25 was good for about 20/day, and a B-24 about 3 casualties a day (high-level bombing). The game engine produces about ten times that number.

[rader]

ORIGINAL: herwin

The game engine produces about ten times that number.

And in completely opposite proportions [8|]

[rader]

ORIGINAL: Jzanes
I am Rader's oppoent in this match and I have a few points to add.

I agree ground bombing vs. units in clear terrain is about 2x too deadly and bombing vs. units in cover is about ½ as effective as it should be. However, I think the situation in Russia reflects a near optimal situation for ground bombing;

1. There has been absolutely 0 Japanese fighter opposition.
2. Japanese AAA is either non-existent or ineffective. Bombers are flying at 6000 feet.
3. Allied bombers are flying out of large (level 8 or 9) airbases with excellent supply, plentiful aviation support, and air HQs are present.
4. Japanese forces are moving and are either in MOVE mode or COMBAT mode with 0 forts when bombed. Having some forts dramatically reduces the effectiveness of ground bombing even vs. units in clear terrain.
5. Generally the weather in Russia has been “good” for the last few weeks. Good weather means anything but thunderstorms basically.
6. Even with optimal supply and support, all heavy bombers are not flying every turn. Most turns, 150-250 B29s fly sorties while another 100-200 bombers stay home for various reasons.
7. Some of the ground units being bombed have already been routed in previous land battles and have lots of disrupted components ripe for being destroyed.

Yes, it does reflect a near-optimal situation. And just to clarify, I am by no means implying that you are "doing anything wrong". I would be doing exactly the same thing you are. But even given optimal conditions, these results are out of whack to the point that it is impossible to defend (or even occupy) any clear hex without a lot of fighters/AA (even with forts 4 or less, as observed in Thailand). I cannot hope to defend many of the base hexes in Southern Korea or Japan. In fact, I see no possible way to prevent allied airforces from destroying the Japanese army (apart from hunkering down only in rough hexes, which would abandon many critical hexes in the home islands.

[rader]

ORIGINAL: Jzanes
6. Even with optimal supply and support, all heavy bombers are not flying every turn. Most turns, 150-250 B29s fly sorties while another 100-200 bombers stay home for various reasons.

Haha, only 150-250 flying each day? [8|]

Being able to send any 4Es every day is kind of insane compared with historical operations. Of course, air op tempos are too high for both sides - but the way 4Es are modeled, the high tempo of ops is a bit one-sided.

[rader]

Forts help, but don't allow you to hold clear terrain without a LOT of AA. I had level 6 forts in Hailar, and 3-4 units including a big fortress were trashed in a matter of days by B-29s. You could have destroyed all the units in Hailar (about a division equivalent) in about a week. It's really clear terrain that's borked (and the fact that 4E bombers are so much better at bombing ground units than ground attack planes are).

[Bullwinkle58]

ORIGINAL: rader

Forts help, but don't allow you to hold clear terrain without a LOT of AA. I had level 6 forts in Hailar, and 3-4 units including a big fortress were trashed in a matter of days by B-29s. You could have destroyed all the units in Hailar (about a division equivalent) in about a week. It's really clear terrain that's borked (and the fact that 4E bombers are so much better at bombing ground units than ground attack planes are).

500 lb iron bombs being dropped on clear terrain:

http://www.youtube.com/watch?v=sks6D2l8erA

http://www.youtube.com/watch?v=dGLgZ8ht ... re=related

Yeah, A BUFF can carry more bombs that a B-29. But that was one B-52 in those vidoes. Why would you think a division in the open with no forts would survive a week?

[JeffroK]

What about nerfing the effect and accuracy of the bombs??
 
PS I love the talk inferring the use of Caves on Christmas Island, apart from a few sand dunes the highest point is about 10ft, the island may dissapear with Climate Change.
 
I think an important change would be to limit the number of troops you can base there as its clearly not capable of "Unlimited"