Matrix Games Forums

hmm, OK maybe one more combat upgrade to get approved by the troops.

I think that the CAG counterstrike trap exploit is really not acceptable. I find that my knowledge of it greatly reduces my enjoyment of the Pacific, because I am so careful to avoid the gamey exploit and/or I look for how to use the gamey exploit. I'd like to fix it.

Reminder of the nature of the exploit: you can load up massive fighter CAP over a single CV, then send a single sacrificial CAG to strike the opponent's CVs. The opposition will counterstrike right into your hornets nest, generally losing most or all of the CAGs. Then your own remaining CAGs can go to work on the opposing CVs.

Here is what I have already coded, I'll get it out unless there is a big opposition.

1) when the cstrike is "organized", the quantity of cstrike air units (all types of air) is counted, call it ca (cstriking air)

2) the quantity of defending (active player) fighter and CAG units in the cstrike target region is counted, call it da (defending air)

3) the "observed" quantity of defending air is computed as the sum of da rolls of die(3) plus one, divided by two, rounded down. That is, oda = floor((1 + sum_over_da(die(3)))/2).

4) the counterstrike is aborted if oda>ca.

5) if the counterstrike is aborted, then all of the defending air in the cstrike zone is is marked as "cannot undo move", so that a player cannot simply load up the air, prevent the cstrike, then undo the move and use the air some other way.

The die rolls give some uncertainty as to whether the cstrike will occur. The range of the "observed" quantity of defending air will be 1/2*da<= oda <= 3/2*da, but it is not uniformly distributed. The larger da is the "tighter" the distribution is in relative terms.

This appears to me to work pretty well.

On a related note, when we discussed this we discussed limiting the range of the counterstrike to the range of the air units. I looked at the code, and I was surprised to see that this is already done. I tested it, and confirmed that air does not move further than its range in order to take part in the counterstrike.

One subtlety is that the range is computed from where the unit is, on CAP, not from it's airbase. So it could still get unnaturally far from its airbase. But on the whole it seems that the implementation of this is OK.

Sounds like a great change to me!

Thanks for your dedication to improving AWD, I'm sure there's a number of people who appreciate it. [8D]

ORIGINAL: GKar

Thanks for your dedication to improving AWD, I'm sure there's a number of people who appreciate it. [8D]

Agree with GKar.

Sounds like a good change to me as well.

I really hope I'm not coming off like a retard (it would hardly be the first time!) but I'd just like clarity here.....
I'd like to take what could be a pretty common scenario like:
1 CV with attached CAG and 5 fighters on CAP sends out the 1 CAG against 4 CVs with 4 attached CAGs.....
the attacking CAG is (almost certainly) destroyed leaving 4 CAGs to counterstrike....

then your formula kicks in....

5 (now they are DEFENDING against the counterstrike) defenders equal 5da, so
5da (now we will check for o(bserved)da) multiplied by a three sided die (let's say the roll is a 1) is 5*1=5. 5+1 = 6 and 6 divided by 2 is 3 (no need to round down here.) Soooooo oda is 3 which is inferior to 4, meaning the 4 CAGs will still counterstrike.

the same situation with a roll of "2" on the 3 sided die results in: 5*2=10. 10+1=11. 11/2=5.5 rounded down to 5 which is > 4 (ca)

So, Brian, I'd like to know here if
1) I've gotten this right
2) if not, where I've misunderstood.

The main point I'd like a bit of help on is whether each individual da gets its' own 3 sided die roll or whether a single die is rolled for all da..

In the case of a group roll, the odds of getting the "1" result and hence, an opportunity for the ca to still make their strike are greater, an individual roll for each will tend to favor an average result of "2" more strongly.........this is really the main point since honestly, I like the idea of allowing people to try this tactic (ruse, whatever you want to call it - hey after all, war DOES involve trickery!) in the hope of achieving a "1" group roll - this entails great risk, of course, the sacrifice of the CAG and the use of all those fighters for potentially no gain.....

This would also encourage keeping down the number of fighters on CAP with the originating force and so, less damaging in the case of a oda > ca result.

If, on the other hand, the individual roll is used, the odds are that the result oda < ca will almost NEVER happen, particularly if the "exploit" is used in a traditional sense (i.e.: massive fighter CAP present)

I like the idea of allowing this type of thing to exist on a very low chance level (group roll for oda)keeping in mind the significant costs paid by the perpetrator in the case of failure - it makes the game just a bit trickier and "tactical" if you'll allow that term to be applied to AWD; carrier battles in WWII had strategic impact but were largely tactically decided anyway.......

Your thoughts?

Christian,

It looked like you had it right. One 3-sided die is rolled for every defending (active player) air unit in the cstrike region.

So, the smallest possible "observed defending air" value occurs if every die is "1" and is (1+num_defending_air)/2 (rounded down), and the largest possible observed defending air value is if every die is "3" and is (1+3*num_defending_air)/2 (rounded down).

The possible range is actually round_up(defending_air/2) to round_up(3/2*defending_air).

It is true that war is full of trickery, but I'd rather have ways of tricking the player than ways of making his units do things no player would ever order. The cstrike stands out as the one place where a player can force the opposing player's units to do things, sometimes nonsensical things.

Maybe there are better schemes. I thought of just comparing defending air to cstriking air. But it does seem better to leave some uncertaintly in it. Just depends on how much uncertainty you want to adopt.

how about this:

4 cags strike 4 cag defenders, losing 1 cag in the process and hurting one defending cag, now they are 3da vs 3ca

if the da's get a 1, 2, and 3 respectively, this will result in (1*1+1)/2 = 1
(1*2+1)/2= 1
and
(1*3+1)/2=2
which total
4 oda
so oda > cs
so no counterstrike..........
this seems like a fairly standard result.......
am i missing something here?
You're basically suggesting that a standard 4 to 4 cag battle with pretty standard rolls is going to result in no counter strike. Don't get me wrong, I don't want to see this exploit used anymore than you do, but this seems like a "fix" that breaks as much as it repairs (if I understand your suggestion correctly.)
Please correct me here if I'm wrong somehow.

If you have 3 defending CAGs and they roll 1, 2, 3 respectively then
oda = rounddown( ( 1 + 1+2+3)/2) = 3

You do the summing over dice before you add 1, divide, and round.

Also, the counterstrike occurs BEFORE the originally attacking CAGs return to their CVs, so in your example the number of defending air da=0.

ORIGINAL: WanderingHead

Also, the counterstrike occurs BEFORE the originally attacking CAGs return to their CVs (...)

This is a crucial detail, which isn't mentioned above unless I missed it.

Also, the counterstrike occurs BEFORE the originally attacking CAGs return to their CVs, so in your example the number of defending air da=0.

Wow, I'd never noticed that, I was certain the counterstrike happened after the attacking CAGs had returned and basically could have sworn there were cases where counterstriking CAGs duked it out with the original attacking CAGs a second time (I must be wrong about that!?)

If that's the case, then of course your formula will work just fine! [:)]

Also, thanks for the clarity on the +1 to the dice rolled to determine oda!

Brian, I can't help but wonder if my recent attempt to sucker Dave into this trap got you stirred up on this? Dave noticed the danger of coarse and moved his carriers into Pearl harbor and placed his air on Cap over the sea zone, thus denying me a chance to trigger a counter attack. In our 4 player game the US has the advantage in this case because he's got a port to back into, and I don't (Wake), so I have to always maintain fighter superiority to keep Dave from trying the same on me. What really seems strange to me is that the cituation changes so much just by removing the carriers from the sea zone. All them juicy heavy fleets just sitting there and I can't pull this trick to even hit a single one, just because there is no CV present to trigger the counter. All the same air is present, only the CV backed into Pearl Harbor. Don't get me wrong, I'm not complaining that I couldn't force a counter attack. I'm pointing out the obsurdity of the outcome over a minor difference in placement of a single CV. You guys can keep going round and round with complicated suggestions to try and address this game flaw, and never come up with a good and simple to understand mechanic to solve it. Which is why I pointed out before that the counter attack should just be removed from the game. It's only there to allow the defending air a crack at the attacking fleet. But the way it is implemented leads to all this funny business. If on the other hand, the attackers CV were always attached to their carrier air, then the defending air could fire back on them in a single battle. There would only be two complications to overcome.
1. Define the difference between when the attackers CV's are actually in the defenders sea zone or outside it.
2. If inside the defenders sea zone then the CV's take part in normal fleet combat. Otherwise if outside, only air can strike at them.&nbsp;

In normal cases, no attacking fleets can enter the defenders sea zone until air combat has cleared out all his carrier air. But the above rules would still be needed for the rare cases. Using a system like this is easy to understand because everything can take place in one battle and it would be clear in the detailed battle report what fired on what. In a simple case of 2 CV attacking 2 CV and 1 HF you could see with just one battle report the air combat and the torpedo attacks on the opposing fleets with the 1 HF not firing at any ships because the attacking CV in this case was not inside the sea sone. When fighters are present for either side nothing gets funny. It only results in a larger air battle.

ORIGINAL: Lebatron
Brian, I can't help but wonder if my recent attempt to sucker Dave into this trap got you stirred up on this?

Not really, I am not actually that observant on other people's turns in our game, I didn't realize you tried to pull it off.

No, I was just thought it might as well get cleaned up.

I can put a hook in easily enough to mod-ably remove it from the game, but I can't rewrite the combat mechanics to give it an entirely different form.

Giving us the choice to remove the counter attack in mods without the other change would lead to immune CV's. The defending air would have no chance to hit back, so if you can't find a way to include attacking CV's into the first battle you may as well not bother adding a hook to enable that.

There might be a way to get the combat engine to recongnize and include attacking CV's into the primary battle. Could you use the carrier-to-carrier air link to make the engine recongnize which CV's outside the enbattled sea zone are actually participating in the battle? Say the chain link says 3 in the icons corner and that carrier air#3 attacks a fleet with CAP over it, couldn't the engine be programmed to automatically include CV#3 into the battle so the defender CAP can strike back at the CV?

I would really think it would be worth it to pursue this programming challenge because it would make things logical and easy to understand. Also notice that in my example a defending CAP was present and not necessarily a loaded carrier. Because this way of resolving fleet combat does not require a counter attack, one does not need the presence of a loaded carrier to trigger it. Air simply does what it should do.

Jesse,
What you suggest seems reasonable but it appears that you may have overlooked the fact that the attacking carriers will not have any of the protection afforded by the rest of the fleet they sometimes cruise around with........Have I misunderstood you?

For my money, Brian's original suggestion to determine the "observed defending air" will prevent the exploit and does not seem overly complex to me (the player does not really need to understand the EXACT dynamic, only that the exploit will not work, it could be worded like this:
"In the event that air units on CAP over the CVs from whence come CAGs that strike an enemy fleet outnumber the air forces of the defending CV fleet when they are about to make the counterstrike, the counterstrike will be aborted to avoid having the counterstriking air forces lured into a trap."

ORIGINAL: christian brown

Jesse,
What you suggest seems reasonable but it appears that you may have overlooked the fact that the attacking carriers will not have any of the protection afforded by the rest of the fleet they sometimes cruise around with........Have I misunderstood you?

Yes and no. I failed to mention all the details. Other defensive warships would become targets too, using the same percentages in place now. When defensive air strikes back, any ship in the attackers carrier fleet could become a target. But not a target in ship to ship combat since they may be in another sea zone. That's the key issue where the battle engine could be improved. I'm sure that if a sequel ever gets made, the battle engine may end up working like this, but Gary and Joel I'm sure would want to limit how much Brian can improve the game engine. There is no money to be made in a constantly improved game that leaves nothing good for a sequel.