I've been out of the loop for a bit, but I know that Mailanka has been working on rules for building Psi-Wars vehicles and, most recently, dogfighting. I've read his analyses, and having seen his comments about the strangeness of using Action chase rules to simulate a dogfight, I have to wonder if starting with the abstract rules is the best method. Sure they work, but it seems like we're trying to make an existing part work with two different standards using a mess of adapters instead of just making a part that uses those two standards to make the connection.
So, I'm going to try and work my way through making that part for Psi-Wars.
EDIT: Realized it might help if I link the posts I'm referring to. Oops.
In reading through Mailanka's two posts, I noticed that he's using modern fighters with modern missiles, and I think this is probably the biggest flaw in the approach he's chosen. He has much more experience than I do with GURPS mechanics and is much better at that side of the game than I am, so I don't think this is a fatal flaw, but I feel that it is one nonetheless. It ignores the feel of Psi-Wars and the Star Wars and adds in an element that doesn't get much use in the Star Wars media that I have seen. Worse, it's an element that breaks the dogfighting element of space combat. That is why I feel that it is a flaw.
If it is a flaw, then how do we fix it?
Instead of trying to get an abstract set of rules to fit our paradigm, we start with the paradigm and work towards our rules. Naturally, our next step is to establish what our combat looks like. It's clear from the Millenium Falcon chase in A New Hope that Lucas was heavily inspired by clips of WWII aerial combat. The Falcon is in the role of a bomber while the TIEs are our Bf-109s. I remember seeing a clip of that scene alongside the clips of footage that Lucas used for reference, but I can't recall where. In any case, it's clear that Lucas was going for an almost frame for frame recreation of those WWII clips.
But is it really WWII dogfighting? Sure, planes would sometimes get ridiculous close, but how close was a typical dogfight? Well it seems that it's remarkably difficult to get hard numbers on that, but we can make some inferences. Based on some information about the Thach Weave that I found, a F4F Wildcat could make a 200 foot radius turn at speed. It's also been noted that the F4F Wildcat and other contemporary American fighters were not the most maneuverable of planes. This means that planes could be rolling, diving, and turning with less than a couple hundred feet between them, and probably had no more than a couple thousand feet when the dogfight started. This jives pretty well with what we see in the Star Wars movies. They may appear closer, but the fighters are also bigger than their WWII counterparts.
So how do we model this in GURPS? Well, let's look at what a hypothetical dogfight might be like. A four-man formation is flying around looking for targets of opportunity on the ground. Another formation of four enemy fighters has been sortied after reports of a bombing attack on a weapons depot. The four pilots in our first formation, formation alpha for sanity's sake, are looking out for trouble as well as prey. The four pilots of the enemy formation, formation beta, know generally where formation alpha is heading thanks to the report and they also know the terrain well. There's little cloud cover, but plenty of hills and valleys in the area to help hide formation beta's approach.
We have the setup and our first sets of rolls to make. First, formation alpha makes two Perception rolls, the first is done to look for targets of opportunity. The second is to see if they spot formation beta. If this were a game and formation alpha the PCs, I'd say that this one needs to be a secret roll. Next, formation beta's commander makes an Area Knowledge roll to see if he can find a way to sneak up on formation alpha. Success means that formation beta can make a piloting roll to fly nape of the earth and stay hidden. Failure means that formation beta is stuck with trying to get into a position that they can come at formation alpha with the sun at their backs. Arguably, this should be a Tactics roll by the commander with possible bonuses for coming in from unexpected directions or from knowing the land better than the other formation thanks to a successfull Area Knowledge roll.
All of this is setup. How the first phase goes depends on the rolls. If formation alpha misses the incoming fighters, they're caught flatfooted as bullets come zipping in followed by formation beta. Formation alpha has to make Will rolls to snap out of the surprise and react, then make piloting rolls to evade or control their fighters while formation beta just makes Piloting and Gunner rolls to position their aircraft for the kills and pour fire in the formation Alpha. If formation alpha spots the incoming fighters, then we have a quick contest between the commanders Tactics skill. alpha can turn the tables on formation beta with a success while beta can maintain their advantage and the first strike.
After that, it's rolls to see if wingmen manage to stick together, perfrom maneuvers and notice that someone's managed to get on their tail. I'm not about to detail the fight since it is involved and repetative, but I can go over the first round.
Let's say formation alpha doesn't spot formation beta until the rounds are flying, what now? Well, now formation alpha knows beta is there, so the commander rolls leadership to keep his men from panicking and get them to go through the proper response to this. Of course this is assuming that we aren't rolling for Will due to something like Combat Reflexes, or that he's already succeeded. The rest of alpha also needs to roll Will if they don't have Combat Reflexes then the whole formation rolls Piloting to see if they can follow through the commander's orders. Beta then rolls Piloting to see if they stay on the tails of alpha, then Gunner to see if they hit, rolling for damage like normal. This goes on with each pilot rolling to see if they can gain the upperhand on another or just escape. Pilots should also get a Perception roll to see if they notice when an enemy is trying to get on their tail.
We can basically divide our dogfight into two phases. The first is set up where we determine the conditions for the first exchange of fire. This also neatly settles who has the advantage in the first round. The main issue with this is that it requires the GM to have and commnicate a fairly detailed view of what the setting of the dogfight is like. If it's just in deep space with two fighters heading at each other head on, then one or both of them are going to break and try to get out of the other's line of fire while also trying to get the other pilot into his line of fire. Basically, it turns into a really lethal game of chicken. But odds are you're not going to run into that outside of a thought experiment. The most likely scenario of that occurring is when you have an incursion by parties unknown and the other pilot is trying to safely get a look at the intruder.
You may notice that I haven't mentioned a word about radar or missiles. While WWII saw both of these come into play, they were very limited in ability. Radars could only show returns. they couldn't tell an operator what craft it was seeing, or how many, or how big, at least not directly. Large objects and formations returned larger signatures, but it ws crude and you had to close in to know what you where looking at anyway. Most of the craft that mounted radar systems were night fighters who needed it to find the bombing formations that were their prey.
Missiles present an entirely different issue. In Star Wars, we don't really see missiles, but from a comment I read, it seems like there's an expectation that they'll be a part of space combat. To that, I would like to answer with a short little story. Back in 2011-2012 there was a game called Moon Breakers that was released as an open Alpha. It was basically WWII dogfights in space around asteroids and moons and it was a lot of fun. The game wound up going belly up after it launched on Steam and while many would point to the lack of content or the variation in the maps, I think there's another issue that cropped up at the same time that helped it along. You see, about the same time as the Steam launch, they introduced a slew of new fighters, several of which had homing missiles.
Now, at the time I had gotten fairly good at using a small, swift little fighter that was a rough analog to the Zero. I stopped playing because I got tired of dying to missile after missile after missile. The game was built as a dogfighting space sim, and the introduction of missiles that even a speedy little fighter couldn't dodge broke the game. It became more about how fast you could spam missiles out and thanks to the Free to Play model they adopted, it became a pay to win thing. You had to buy into the in-game economy in order to get the fighters that had the homing missiles because getting them in game would be a crushing grind of dying over and over again.
The moral of the story is that you have to be careful when adding weapons and elements to an established paradigm or else it can and will break, and this is the issue with missiles as they are in GURPS at high TLs. They are too accurate and lethal to allow for the kind of hectic dogfighting we want. So what is the solution? Star Wars has missiles, the famous Trench Run needed them to hit the exhaust port and you see them chasing Anakin and Obi-Wan in Revenge of the Sith's opening scenes. But are they really missiles? They are only slightly faster than the fighters and not much more maneuverable, the 90-degree turn on a dime from the aforementioned Trench Run notwithstanding. I'd argue that this makes them different from real missiles and thus in need of their own rules.
Capital ships, another area that Mailanka was worried about, could be modelled by basing combat with them on attacks on bomber box formations. The formation would be the ship and the bombers move slowly enough that I think it'd work rather well. I'm not going to do that here because this has gone on long enough and it wouldn't be that different from the previous example.
The last thing I wanted to cover was how to abstract things so that they move swiftly in gameplay. I'll be the first to admit that the scenario above would take a long time to play through, especially with options like stunts in the mix. You are basically rolling for every pilot's maneuver, and that's not very conducive to speediness. So, what do we do? Well, this is just a first guess, so this is going to be rough. You have four states, On Enemy's Tail, Enemy on Tail, Pursuing X, and Pursued by X. These are for keeping track of who you're fighting and what your status is relative to them. On Enemy's Tail means that you have an advantage on the enemy that you are pursuing. Generally this means that you won the last contest between you and your target. Enemy on Tail means that an enemy has an advantage on you. This means that you lost the last contest between you and the guy targeting you. Pursuing is the enemy you are, well, pursuing. If you are On Enemy's Tail, then this is who you have in your sights. It's also possible for the enemy that you are pursuing to be on your tail. Pursued by is basically a list of all the enemies going after you. Combat works by picking a target, then rolling to see who gets an advantage. Everyone rolls once, unless you're doing a stunt, then you roll that separately and modify the Piloting roll that you are going to use to see how you did against the enemies you are pursuing and being pursued by. If you are On Enemy's tail, then you can roll Gunner to see if you hit. This will require some modification to make work since beam weapons are far too accurate as is for having an Enemy on tail not to be a death sentence. Having the margin of success of a stunt applied as a penalty to the pursuer's Gunner roll could work.
Anway, there you have it. My two bits about dogfighting. Hopefully some of you will find this down and dirty foray helpful. If not, I at least hope it wasn't boring. :)