I'm not one of the BIG guys around here when it comes to Starmada. Heck, I still use a bastardized version of the original version.
But I've utilized a different way to handle fighter combat recently and I thought I'd throw it out here for others to debate its relative merits.
It started because I wanted to play out some large space battles having to do with my Star Wars RPG currently running. To faithfully represent some of the Star Wars ships, I need to incorporate a LOT of fighters.
I have squadrons of 12 fighters rather than 6 for example.
To keep track of the number of fighters in a squadron I use stacks of small change. A 12-fighter Squadron is a dime and a couple pennies for instance.
This is all tertiary to the main mechanic though. Each squadron I rate with an accuracy and an agility score.
A v-19 Torrent Squadron is Accuracy 5+ and Agility 2 for example.
When a squadron rolls an attack, they roll 1D6 and try to roll higher than the accuracy rating.
For every die that is successful, you try to roll over the ships shields (if attacking a capital ship) of above the fighter agility (if attacking an enemy fighter squadron.)
In the original Starmada, you'd lose a fighter for every '1' you roll against fighters (and Against ships if the ship had AFB) Instead, I adopted this system:
When two enemy squadrons meet, they are in a dogfight. Whichever side has MORE fighters in the dogfight gets to attack. Every '1' rolled on the attack allows 1 enemy fighter a counter-attack opportunity. If no '1's are rolled then the enemy doesn't get to fight back at all that turn.
A squadron that attempts to leave a dogfight hex is immediately attacked without a chance for retaliation. (even if '1's are rolled.)
During fighter movement, other fighters may join the dogfight hex, to increase their sides number, and allow them to initiate the attack, rather than hope for the enemy to roll '1's.
This makes sense to me because when fighters are outnumbered, they are spending most of the time trying to shake fighters off their tail, and possibly take snap shots when an enemy passed in front of them.
In a dogfight you want to have the numerical advantage, so you tend to group squadrons together, and throw them into the furball. This ties up lots of fighters, in a realistic manner, in several major hot spots without having them stalking all over the map.
You can also intercept incoming fighters, because once you engage an incoming squadron, they are pretty much fixed in place until they clear up your fighters, or they take the hefty penalty for showing your fighters their afterburners.
In the last game I used this in, we had 16 capital ships on both sides ranging from hull 2 to hull 30, and we had 168 fighters on one side (Grand Army of the Republic), and over 500 on the other (Confederate Independent Systems).
When there were really big dogfights, we'd roll an attack die per SQUADRON rather than per fighter, but resolve damage per fighter.
Some of the dogfighting hexes had a fighter counter sitting atop 3 quarters, a pair of dimes, a nickel and a couple pennies (representing 102 fighters) versus a fighter counter atop 3 dimes and 4 pennies (representing 34 fighters)
The 102 fighters, (8.5 squadrons) would roll 8 attack dice. Their accuracy being 5+.
If they rolled a 1, 1, 3, 3, 4, 5, 6, 6, then 3 squadrons would be in a good tailing attack position. We'd roll 12D6 three times against the opposing fighters agility (3) which would filter out roughly half of those hits.
After the destroyed fighters were removed, the opponents were allowed to attack back with 2 surviving squadrons (because two '1's were rolled on the attack.) If no '1's had been rolled, they wouldn't get an attack at all that turn, being entirely defensive.
It worked well for us.