WarPlan Pacific - Carrier Strikes (Video)

[*Lava*]

In this video I discus how WarPlan Pacific simulates carrier vs carrier strikes and offer some recommendations to improve the interaction between CVs.

Agree? Disagree? Don't be afraid to have your say.

https://www.youtube.com/watch?v=kpr5NQljPEw

[*Lava*]

So I would like to give some perspective to the video I did above.

The video is actually the result of tests which I conducted on CV vs CV battles. I conducted over 100 CV vs CV battles in 5 different categories. These categories were (1-4 Japanese moving to contact):

1 - Japanese (4 CV, 2 BC, 2 CA, 1 DD) vs American (3 CV, 3 CA, 3 DD) - US disadvantage
2 - Japanese (4 CV, 2 BC, 2 CA, 1 DD) vs American (4 CV, 2 BB, 2 CA, 1 DD) - Equal forces
3 - Japanese (4 CV, 2 BC, 2 CA, 1 DD) vs American (4 CV, 7 BB) - American advantage
4 - Japanese (6 CV, 9 BC, CA, DD - 15 Total) vs American (4 CV, 16 Support - 20 Total) - Japanese CV advantage vs American support ship advantage
5 - An addition scenario using the forces of #3 with the Americans moving to contact.

The tests consisted only of the initial strikes (first operational point) and second strike (2nd operational point) as I consider 2 strikes from both CV groups to be a good representation of a CV battle.

My impression from the battles (because I didn't have the massive amount of time to record all results) was that in approximately 95% of the cases, the Japanese gained a clear victory over the Americans. In 5% of the cases, the Americans received a clear victory over the Japanese. The Japanese appeared in the majority of the tests to focus their attacks on the CVs, while the Americans had a tendency of attacking other ships such as BBs. The amount or type of support ships with additional AA appeared to make no difference to the results. In about 15 to 20% of the battles, the 2nd strike failed to find the opposing fleet.

These were the conclusions and recommendations I reached from my testing:

1. Search: When a CV enters into a potentially hostile area, it is continuously launching search aircraft (without ordnance giving it a far greater range) at altitude to find the opposing naval forces. These search aircraft are not fighters, they are attack aircraft. In WWII these aircraft will be assisted by seaplanes from CAs and BBs. The search effort is meant to find the opposing force before it enters strike range.

In a CV vs CV encounter there is also a probability that the search aircraft from both fleets will actually come into contact, thus limiting the search sectors and increasing the odds of finding the opposing fleet.

Recommendation: Any opposing fleet should be spotted automatically at 3 hexes (attack range) from the CV. The report should have a fairly high probability of being inaccurate. This also applies to long range, ground based aircraft. The report should not be 100% correct as it is now in the game.

As an example of a worse case real life erroneous report, during the search phase of the Battle of the Coral Sea a report was made of 2 US CVs which turned out to be a DD and an oiler. Both were sunk by airstrike.

2. CV vs CV Airstrikes: Both the Japanese and American commanders knew that the key to victory in the Pacific was the destruction of the opposing forces CVs. During strikes, aircraft went to extraordinary means to find the CVs and attack them. We see this at the Battle of Midway where the US torpedo bombers flew through heavy Japanese anti-aircraft and were attacked by fighters to get to the Japanese CVs. In one case, an entire torpedo bomber squadron was destroyed. Later in the war, as the Japanese became more desperate, they adopted a strategy of attacking picket ships to limit exposure to US fighters and of course, Kamikazes to attack the CVs.

What we also find happening here, when two opposing CVs groups would attack each other, is that aircraft losses were not only limited to the point of attack, but would also occur as the returning air wings would fly past each other and air-to-air combat would occur. This is why air strikes had such appalling losses of aircraft.

Recommendation: A battle between 2 opposing CV fleets should see all damage applied to only CVs. The commanders understood how to achieve victory in the war and the aircrew who fought in that war understood what their only target was. I would also increase the number of aircraft losses during these battles. These are desperate fights and many aircraft would return to their CVs only to be pushed into the sea because of battle damage.

3. Airstrike Results: In the main, the vast majority of my tests resulted in clear victories by the Japanese over the Americans, no matter what the scenario. In this early stage of the war, the Japanese aircrew were battle hardened veterans who were clearly superior to the Americans.

As I chatted about in the video, was the Battle of Midway a dice roll of six 6's in the American's favor? Pure luck? The Americans in this battle attack piece meal, with the torpedo bombers attacking first. The Japanese zeros concentrate on these folks and a slaughter ensues. Then, out of the blue, the American dive bombers appear from another direction and attack the Japanese, striking 3 CVs. Sounds like luck to me.

However, let us change the tables and instead of the Americans attacking let us say that is the Japanese who make this attack. Would the same result have occurred? Nope. And the reason is clear. An Air Wing strike would transit at a medium level such that they could get a clear picture of where the CVs were to attack them. The difference, and this a huge tactical difference, is that the Americans have radar. So the Americans are going to see aircraft coming at them from 2 different directions and are going to be able to vector fighters to intercept both groups and have time to maneuver their ships so as to place as many pickets between them and the carriers. Do the Americans take damage, you betcha, but not the kind of carnage that occurred in the actual battle.

So overall, I believe that the Japanese are a tad over powered. Yes, they should, in comparison to the Americans, do more damage. But from my tests, they are doing a tremendous amount of damage and that is not realistic.

Recommendation: Dial back Japanese damage rates a bit to give more realistic results. That doesn't mean that 6 Japanese CVs vs 4 American CVs aren't going to be more powerful. But you shouldn't see the Japanese wiping out the entire American CV force with the Americans, sinking only 1 Japanese CV. 3 to 2 seems a bit more realistic, IMO.

So... sorry for the wall of text. I know most folks won't be bothered with my video as I am a pretty poor YouTuber. As an old guy I enjoy making videos as a hobby and they don't have flash intros, quick wit, and the precise language of a pro. Because of this they are also much longer then they should be.

Nevertheless, this video took a lot of hard work with tons of testing and was the best presentation I could make given my abilities.

[ncc1701e]

Wow - this is an impressive test. 100 CV battles.

I would also add something but wonder if at this scale it works. This is the weather. Sometimes CV were undercover of clouds. For example, in Midway, the Japanese should have spotted TF-16. But they missed it due to clouds cover.

What would be great is to have random sea hexes in a given weather zone where there are some clouds or some rain. This way, spotting can fail in these hexes protecting an given CV fleet.

[ncc1701e]

What would be great is that you are re-running your testing with the new incoming patch. Now, fleets are losing effectiveness.

Warplan was just updated. I foresee that Warplan Pacific will be officially updated soon too.

[*Lava*]

Yes, I will re-run my tests with any patch where it seems appropriate and report the results and the next time I will record my results.

And yes, weather is the big game changer. I would say that if you have clear weather CV has automatic 100 percent spotting with some erroneous reporting. But as you have light rain that starts to drop off and reporting gets worse until you have none with blizzard conditions. In fact, I was once in the Sea of Japan in the winter and we encountered blizzard conditions. That was scary as hell due to ice on the flight deck.

That would be a great addition to the naval game, but probably a huge pain in the ass to code.

[ncc1701e]

23:53 in your video. Just look at the reconnaissance level as Japanese of the hex where the enemy fleet is. The hex is stated as "Reconnaissance very low". This is why Japanese has not detected the US fleet.
To improve reconnaissance, you must use COMINT.

[ncc1701e]

29:54 in your video. The US fleet is detected. Yes but this is not because of your planes. This is because hex 128, 60 is stated as "Reconnaissance high". And it is "reconnaissance high" because the hex, just south of it, is Japanase own. You must pay attention to this variable. This is the basis for detection.

[*Lava*]

ORIGINAL: ncc1701e

23:53 in your video. Just look at the reconnaissance level as Japanese of the hex where the enemy fleet is. The hex is stated as "Reconnaissance very low". This is why Japanese has not detected the US fleet.
To improve reconnaissance, you must use COMINT.

And that is why I explained above that a CV TF entering into a hostile area will be flooding the area with search aircraft facing the threat axis to ensure it finds any enemy before it gets into strike range.

Thus, from my POV, the simulation is flawed.

ORIGINAL: ncc1701e

29:54 in your video. The US fleet is detected. Yes but this is not because of your planes. This is because hex 128, 60 is stated as "Reconnaissance high". And it is "reconnaissance high" because the hex, just south of it, is Japanase own. You must pay attention to this variable. This is the basis for detection.

That is true AFTER the CV moves. Notice, however, at 29:26 when I am placing my movement "dodad" that hex is marked "Reconnaissance very low." So I have no way of knowing when I am actually moving that in fact that hex is a high reconnaissance area until after I move there.

As for COMINT, the documentation is pretty poor in how it works. In my tests, I placed a COMINT site on Midway Island to give the US as much advantage as possible. When I placed it there it said that the COMINT was very low. Afterwards, there was no indication of a COMINT site being there, so I am now under the impression that creating a COMINT site is treated as an expendable that only has one turn of value.

While obviously it has some utility, it appears (and please correct me if I am wrong) that first you need to break the enemy code and then when you think it might yield good information, expend your COMINT consumable to see what you can see.

[ncc1701e]

ORIGINAL: *Lava*

That is true AFTER the CV moves. Notice, however, at 29:26 when I am placing my movement "dodad" that hex is marked "Reconnaissance very low." So I have no way of knowing when I am actually moving that in fact that hex is a high reconnaissance area until after I move

At 29:26, you are in your US turn. So you are moving your DD fleet in a hex stated as "reconnaissance very low" to the US side. This hex is just near Japanese territory with plenty of coast watchers. This is when you switch to Japanese side that the hex is then marked as "reconnaissance high" to the Japanese side.

[ncc1701e]

ORIGINAL: *Lava*

ORIGINAL: ncc1701e

23:53 in your video. Just look at the reconnaissance level as Japanese of the hex where the enemy fleet is. The hex is stated as "Reconnaissance very low". This is why Japanese has not detected the US fleet.
To improve reconnaissance, you must use COMINT.

And that is why I explained above that a CV TF entering into a hostile area will be flooding the area with search aircraft facing the threat axis to ensure it finds any enemy before it gets into strike range.

Thus, from my POV, the simulation is flawed.

I see what you mean. Land spotting planes are simulated by the fact land ownership is determining the reconnaissance level of nearby hexes. Thus, a small island owned by one side is giving some reconnaissance.

Now, in WW2, CV and BB were having planes or floating planes (for BB) in order to carry their own searches. I don't know if CV stack is acting as a "mobile" COMINT. But, if it is, its capability should be tied to Carrier Operations advancement level.

[ncc1701e]

And, just to complete, on first turn, you will see that Honolulu is reconnaissance very high for Japan. But ALL the hexes around are reconnaissance very low.

Looks like a trick to allow Japan to sink something on turn one in Pearl Harbor.

[*Lava*]

ORIGINAL: ncc1701e
Looks like a trick to allow Japan to sink something on turn one in Pearl Harbor.

It's just that the simulation needs adjustment and shows us how to do it by making all the area within 3 hexes from the CVs as Reconnaissance high.

In the great scheme of things I don't think that will unbalance the game. It will allow American CVs to do hit and run tactics in the early part of the war (which they can do already) but more importantly, ensure that when the CVs are committed to battle each other, this will be a conscience decision of both players.

[ncc1701e]

ORIGINAL: *Lava*

It's just that the simulation needs adjustment and shows us how to do it by making all the area within 3 hexes from the CVs as Reconnaissance high.

No, all the hexes aroung the CV fleet is completely ignoring the search arcs of spotting planes.

[AlvaroSousa]

Point #1 - The system isn't a only a hex based system. It is a sea-zone hex based system that uses hexes to create artificial sea zones for actions. The turns are 14 days not 1-3. All ships are considered patrolling the area. It doesn't mean they always sit at one point. It's a good idea but at this point I am keeping the searching system the same unless I see a glaring flaw in it.

Point #2 - In my recent games I did notice a flaw in targeting. It is an exploit I found in the system I designed. Sadly I only realized it after the new patch was put out. I won't say what it is so it isn't abused. But it directly relates to your 2nd point. CVs will be adjusted in the next patch to have a much higher targeting profile in air battles.

Point #3 - The scales you are talking about are in a single day. The scale here is two weeks. There is luck involved in combat. If we take the Midway battle redo it. Say the US fleet isn't in fleet mode. They Japanese attack the island that has a couple AA points. Now on the US turn a tactical bomber there attacks the Japanese fleet. They don't do damage but suck up effectiveness. The Japanese attack the island again. Now the US fleet goes into fleet mode and attacks the Japanese. Try that scenario. Next is that naval battles have so much variance it is hard to adjust for. Next is that the Japanese planned foolishly. If they brought a 6 CV fleet and a less complicated plan they wouldn't have been killed at Midway. In every board game I played like World in Flames neither player does what was historically done. It is a cat and mouse game until 1944 where there is one massive CV battle to determine who controls the Pacific. Kind of boring actually. Lat note.... Japanese optics were as good as American radar during the day. So if you have the Japanese play historically yea you will get historical results. If they play intelligently you won't. But regardless if the USN gets wiped out in the Pacific when 1944 hits everything changes. They could have lost every battle until 1944 not sinking a Japanese CV and the Japanese will lose. Right now I am in a game where the Allied navy got thrashed. it lost 3x as many CVs as the Japanese. It doesn't matter. 1944 I will have such a massive armada of carriers the Japanese can't stop me. Which is how the game generally should roll. It's my first two games of WPP vs another human who is very good and uses all the exploits. He does everything that the most proficient players do every game we play. I always play alternative slower strategies. This way I get a good feel for how the game is and know where to balance it out. We have played a good 20 games so far of WPE and this is our 1st 2 of WPP.

The most important flaw you pointed out was the targeting. Which I only realized after the patch was deployed. The other stuff I will think about and see how and if they can improve the game play. So just because I don't agree with them now doesn't mean I might not employ them or add them to WP2. WP2 will have a sea zone system BTW. The flexible hex-zone system I have here has proved problematic. But it has helped me write a better sea-zone system for upcoming games.

Thank you very much for the video and the input. it is ALWAYS appreciated. Sometimes I don't agree with an idea someone has but I still put it on my list of considerations and ponder it for a while. I have to run through all the possibilities and potential exploitations before I implement a change that is large. Somethings you recommend are big changes even though they seem small.

[*Lava*]

ORIGINAL: AlvaroSousa

Thank you very much for the video and the input. it is ALWAYS appreciated. Sometimes I don't agree with an idea someone has but I still put it on my list of considerations and ponder it for a while. I have to run through all the possibilities and potential exploitations before I implement a change that is large. Somethings you recommend are big changes even though they seem small.

I absolutely understand.

The reason I was testing the CV battles is because I saw a number of comments which I would characterize as frustration. And indeed, when I was testing the battles I also got this sense of frustration. Many Americans are intensely proud of the War in the Pacific and especially the carrier battles that occurred during that conflict. However, from a computer games POV I understand that it is probably impossible to duplicate.

And I also understand that the time for making big changes is not after publication, because God knows how that might effect game play.

Keep up the good work. I am really liking your games and wish you the best of luck.

[*Lava*]

I have just started my first campaign as the allies vs a normal Japanese AI.

As I attacked the Gilbert Islands, the Japanese CV fleet showed up and we started having engagements. These engagements lasted from October 1942 to February 1943.

I sank 11 Japanese CVs and CVLs to my 3 US. I had strategic bombers in close support covering the area and changed my COMINT code.

So at least against the AI the CV battles are not as lopsided as my testing showed. In fact, given the results, the reverse seems to have been true.

[AlvaroSousa]

For a game as complex as WPP it was incredibly difficult to code the A.I..

But still I need to change the targeting mechanisms and from this last post how the A.I. adds escorts to CV forces.

[*Lava*]

ORIGINAL: AlvaroSousa
But still I need to change the targeting mechanisms and from this last post how the A.I. adds escorts to CV forces.

I believe the 1st time the Japanese showed up they had 2 groups with something like 4 or 5 CVs in a group and 5 or 6 escorts each. They withdrew without battle.

In our first CV vs CV fight the Japanese returned with one group of 5 CVs, 1 BC, 2 CA and 3 CL/DDs. I sank 1 CV with ground based air. I then sank 2 CVs and a BB with my first Carrier Combat and then another CV with my second Carrier Combat.

In our second fight they returned with 5 CVs and 7 escorts. For some reason that battle group ended up with the Japanese in 3 groups. I sank a CV with land base air to start the battle. I next attacked them with my CVs to no success, then detached my CVs and attacked them with my surface groups. My main body (15 escorts including BBs) sank 2 Japanese BBs and they then retreated. The Japanese were now left in 3 separate groups. One reported as 1 ship with 1 CV and 0 BBs, the other as 3 ships with 1 CV and 0 BBs and the third as 3 ships with 2 CV and 0 BBs. I attempted to attack one group with a surface group of 5 ships but was stopped by a carrier interdiction and then attempted to attack with another surface group and it too was stopped by carrier interdiction.

Overall, my impression was that the Japanese were using far too few escorts. My main body had 5 CVs, 6 BBs, 2 CLs, and 4 DDs. I had a UK group with 1 CVL, 5 BBs, 4 CAs, and 6 DDs. In the second main battle I also had a 3rd group of 2 BBs, a CA and a DD to escort my marines onto the beach.

[*Lava*]

Sorry for the size of the photos...

Nov 22, 1942:



Feb 28, 1943:

[*Lava*]

Are you sure the Japanese are returning to port to repair damage and replace aircraft?

Whenever I have a major battle, I return all damaged ships to Pearl Harbor until they are fully repaired and the air wings restored to 6.