High Guard - You only get one pass

[Linwood]

What if we extend the weapon ranges? That might at least provide a few extra rounds of engagement.

 

[phavoc]

There are some rather large holes in Lanchester's laws. His linear equations (for factoring in ancient style fighting like phalanx vs phalanx) works at only the absolute simplest level - it does not include archers, chariots, cavalry, terrain, weather, leadership, etc. Yes, it's a model, but with severe limitations. His square law takes into account massed fire (to a degree), but not technology. And it's only attempting to model attrition. So somewhat useful in things like artillery bombardments, though as a former redleg (artillerist) I can tell you there are a host of things to blunt or remove the effects of shelling.

Gaming none of this can be modeled very well, and shouldn't as it really slows things down and, in my opinion, makes it far less fun. Understanding the underlying aspects (and limitations) of things helps shape theories. But the first thing to be remembered is that modeling things like this is not necessarily a good reflection on how it will go in reality - especially when it's people who are being modeled.

There's nothing wrong with models, but problems arise when people mis-use models and get them to do things they're not designed for. These aren't holes or errors in the model.

To use the example of artillery. Predicting where artillery shells will land when fired is a model. Do you trust that model?

Predicting the effect of that shell landing is a whole different model, and is less accurate.

Regarding High Guard, if you line up equal capital units and fire intelligently, Lanchester will work just fine. If you don't, it's a user error.

Well, I guess we can agree to disagree. As someone who used to be in artillery I can tell you that knowing what your rounds are capable of does impact your process. If you know the effects of what your weapons can do and factor in other items such as the amount of time you have to bring your rounds to bear, the type of target you are attacking, etc, that heavily influences the choices you make. Technology has also changed in that you can paint a target with a laser and you can have the rounds impact that target(s) specifically.

Of course it's warfare so no matter how much you model and predict, you won't be 100% right all the time.

I'm not sure you are looking at the model the same as I am. His linear model doesn't scale up well and ignores multiple other factors. Models are useful as predictors, but only when you have applied as much knowledge as possible and the models continue to be refined to reflect feedback and changes. His have not, thus they remain mired in the timeframe they were created.

 

[Anonymous]

I don't know what you're looking at. I abstracted High Guard to a binary ko/not ko and was therefore able to use the square law for a large 400 ship combat at which size statistical variance will be low. I can't tell whether your objection is to what I've done, or modelling as a whole.

If it's to what I've done, what's wrong with it? If it's the other you seem to agree with me that all models are wrong, but some are useful. Or is Lanchester wrong?

Where we seem to disagree is that I think if someone is using the model wrong, thsts that's human error, and blaming the model is like blaming a firearm for a murder instead of the user.

I mean you know good models work because you use one everyrtime you try aim that artillery. Laser guidance still needs to use a model (called physics) to steer the round by doing calculations based on how they expect it to steer when they move the fins.

So I guees we cone back to why Lancester is or isn't valid for classic high guard capital fleets?

 

[phavoc]

My objections are specifically oriented towards Lanchester's laws and their use in the modern era. If you look at his work the theory doesn't scale well. That, and only that, is what I was commenting on. I did not, nor do I currently, have an opinion on whatever model you have developed to fight large fleet engagements.

It is difficult, I think, to model such a thing, especially for gaming purposes. In the past when I've tried to play large fleet battles (though never 400 ships) we simply applied off/def values to ships and each had a point value. At the conclusion of a round each defender made their own decision on how many points of ships to remove from the board. It made it very simple, but not necessarily a good representation of what combat might look like. However from a gaming point of view it was easy to resolve things.

Once, for a smaller fleet battle using Star Fleet Battles, a friend and I fought with a Federation battlegroup vs. an Andromedan battlegroup. Due to the points of the ship types, the Federation had more overall ships (11 I think?). It took us hours to fight through just a few turns because we had to allocate power each turn to individual ships and resolved damage per the dice rolls per ship. In the end a lone Fed CL survived as the two groups closed to knife range the Andromedan battleship had a cascade overload of it's power absorbstion panels starting blowing up, the ship blew up, it blew up the ships around it, and since pretty much everything was damaged all the ships in the battle zone followed suit. The Feds only won because their little cripple was too far away. Was it fun? Yes. Would I do it again? Mmm, probably not. Games like Federation and Empire did fleet combat at a very high level. It's not perfect, but it works for the gaming model. I have forgotten the mechanics of Starfire (Imperial Starfire was the strategy game I think... gotta go open boxes to recall).

 

[Anonymous]

I would not use this model for Star Fleet Battles as the player actions and statistical variance are too significant at this scale, and the model lacks enough detail to simulate the ships. It was not designed for this and using it would be an error.

A simplified summary of Federation & Empire: ships are assigned a combat rating. You make a fleet of about 10 ships (if you have more they don't participlate this battle) and total up the combat rating giving a single numerical value for each side. You then determine a percentage by dice roll. A side inflicts the percentage times its combat strength to the other. You then allocate the enemy's damage by cripping or destroying that many points of your own ships. Crippling a ship turns its counter over to the damaged side (half rating) and soaks up points equal to the undamaged rating. Destroying removes the ship and soaks up points equal to the crippled rating

It can be seen that the lanchester square model correctly predicts the casualty ratio in damage points between fleets in F&E because their rules are actually a gamified version of the model.

However this isn't the whole story because that model doesn't take stategic considerations into account.

Let's say you have lots of 8 point ships and take 24 damage. You can cripple 3, or you can destroy 2 (8+4 doubled). The long term consideration is to cripple 3 as you can repair those in the next strategic turn. The short term consideration is to destroy 2, as your number of usable ships for this straetgic turn drops by 2 instead of 3. Lanchester doesn't model this additional level, and using it would be inappropriate for a full engagement.

So we see that we have to know when we can or can't use a model. If you are working at a scale where individual unit tactics aren't that relevant, your units are similar, don't degrade well, and you have a lot of them, Lanchester's OK. However if one side achieves dis-similarity (for example they disable the enemy's only radar) then your units are no longer similar and it won't work anymore. One has to assume senior army officers and their planning staff know when they can and can't use a model.

Starfire was prerry terrible. You assigned ship systems in order from left to right at design time: shield, armor, engine, weapon, engine, weapon, life support, engine. Systems were lost from left to right, with some exceptions. You therefore had to choose at design time if you protected engines with weapons, or weapons with engines.

The main issue with starfire is that it generated huge battles, and most people resolved those using range only instead of the hexmap it was designed for, and therefore there was a single best design strategy: 13-15% passives, 10-15% point defense, all missiles, all engines bar 1 up front to protect the weapons. There was no diversity between fleet designs when played out in this manner without a hex map. At least for open space. Warp point assaults (they had fixed jump gates and no jump drive on ship) allowed the defender to shoot first and you needed some heavy protection and close range beam weapons there to survive the "gate camp".

 

[locarno24]

I disagree that this will happen. They want to drive-by our planet and drop their bombs on our factories. They don't really want to fight anything they don't have to. We don't want to let bombers near our planet. I can't see us co-operating. (A landing might be different, of course; they may want to stick around to make sure nothing happens to their marines).

It basically depends on how effective the high-speed pass on a target that's not the enemy fleet is.

If you can do your damage in a single pass without slowing down then you can afford to make a 'fast firing pass'.

That in turn means you can afford to accelerate to high speed on the way in, which means you can afford to pick some balance point between jumping in close (which gives the defending fleet less time to intercept you) and jumping in further away (which gives you more time to build up speed so whilst they will be able to intercept you they will be going so slow relative to you they'll get a miniscule engagement window).

Which, as you say, boils down to the defender getting 'one pass'.

But the better you can engage your target effectively in one quick pass, then the better (logically) they can engage you.


If your objective basically boils down to "BFR them and who-cares-about-subtlety-or-collateral-damage?", that's not the case, and the defending fleet has to roll the dice on a passing engagement and hope.

If you're trying to land and conquer the planet, you have to do a zero/zero intercept with the planet, meaning a defending fleet can almost certainly make a zero/zero intercept with you.

If you're trying to deliver precision strikes to erode the planet's industrial capacity or fixed defences with an eye to conquering it later....it's going to be somewhere between the two, and the point at which 'match vectors' replaces 'strafing run' as the preferred defensive engagement is going to be a judgement call based off assumed enemy intentions, technology level, and the luck of your relative starting positions.

 

[Condottiere]

It's going to come down to some form of abstract fleet management, possibly World of Warships, programme.

If one side is the aggressor, has a specific target such as planetary conquest, he might just keep sending ordnance down range until the defenders are forced to attack him, and he has his fleet optimized for attrition. Strafing runs could be broken up through the use of extensive minefields.

Friendly fire should only be considered that if you're positioned behind the artillery.