Dice vs. Strategy

Damage is an all or nothing thing in Magic; if you don't do enough damage, the creature is not defeated and does not carry the damage to the next round.

Note: Sadly, I was thinking of this out as I was typing it, and I'm afraid it's turned into a table top version of the video game FTL...
Do your starships move across a field? If your ship moves like a minitures game, it could be represented by a single card. Suppose your ship "schematic" is a tableau made up of different cards, like Weapons, Thrusters, Cockpit, etc. You could also have a room for Shields, or Hull.

To track damage, you can have the damaged parts of your ship turned face down (you either do enough damage in one attack or not), or since it's laid out in front of you each part can track damage with tokens. Depending on the number of components, this idea could support a small number of ships (fighters), or perhaps one large one (battle cruisers).

As for removing dice from your game, were you planning on another form of number generation, like attack cards, or did you want to keep numbers static?

MtG is probably one of the most accessible card games out there, why don't you try playing it? You can play physically, online, and it's even free on certain platforms.

From the looks of your post it seems like you spend a lot of time designing games instead of playing games looking for good design.

There are many ways of resolving conflict. You can dice-off (what you've mentioned), "shitloada" dice (a la Warhammer), lottery method (put variables in a tumbler and take out randomly), table method (roll dice, consult a table with answers), physical determination (like marbles), etc.

You must look at games you've played before and see what works in other games. If you've played Monopoly all your life, you're going to make a Monopoly-like game. If you're asking for game designs, you may not fully realize how it works when people ask cause you haven't seen it work in a good game.

One thought for multiple combat is to use coordinated attacks as the concept. That would sum all the attacks to a single number applied to a single target. Maybe they can take a penalty (like lose 1 attack point) for each additional target.

Banding

There was also an ability called banding that had players group up their attacker into one attack. It's been phased out but was handled in much the same way as multiple blockers is.

I think that removing luck from games is difficult to do.

Dice and cards both rely on luck (unless you start off with access to all of the cards).

I have looked at this a bit and I can pass on what I have found.

For a game to be purely strategic you need a few things:
Limited space
Limited movement
limited pieces that are in play from the beginning

Games like chess, checkers/Chinese checkers/even stratego all fit.

The one thing that all of these have in common is they have very simple "attack/defend" mechanics. The game play is strategic because it is about board positioning.

I am working on two separate games right now, a dungeon crawl and a space station defense game. I was unable to eliminate "luck" or "chance" from either of them. I think the only real to do to it is to make a throw back game. I just don't know if there is much of a market for a game like that anymore. There already are some great games that are very old that work fine.

If you are making a "starship game" you should just accept that there will be some chance/luck involved. Try to mitigate it by making somethings static. Damage could be static. Or you could make all attacks hit. Have different parts of the ship be given different numbers and the attacker rolls to see what they hit. When something is hit it is destroyed. If you hit the same thing twice it is basically a miss.

Crews could have the ability to work on specific parts of the ship. If a crew member is in a location that is hit they die.

I guess you could have blank ship designs. Have little cutout pieces that go into rooms. The rooms have numbers on them. The cutout pieces can be weapons, life support, controls ect.

I don't really know what the overarching idea of the game is. Just throwing out some ideas.

I'll use A/B for power/toughness. No special abilities.

Player 1 has 3 creatures: 1/2, 3/1, 2/4
Player 2 has 2 creatures: 1/1, 1/3

Player 1 decides to attacks he simply says: "All my creatures are attacking." That is all he declares.

Player 2 now has decisions to make. First he can block with 0,1, or 2 creatures. Then he has to decide whether to block together or separately.

Scenario 1:
P2 decide to focus on the 1/2 creature. He declares that both of his creatures will block that one creature. The two creatures combine their attack to make a 2 power, but their toughness stays separate. They deal 2 damage to the 1/2 creature.

The attacker will be dealing 1 and decides to apply it to the 1/1 creature. The 1/2 creatures dies because the two damage is enough. The defender's 1/1 also dies. The remaining two attackers deal 5 damage to the player.

Scenario 2:
P2 decides to block multiple creatures. His 1/1 blocks the 3/1, his 1/3 blocks the 2/4. The 1/1 and 3/1 each deal enough damage to kill the other. The 1/3 only deals one damage vs. 4 toughness and it only takes 2 damage. So both of those survive. The remainin 1/2 monster deals 1 damage to the player. At the end of the attackers turn, all damage is forgotten.

Once you're ready to crunch numbers:
You want to avoid the scenario where all the player is doing is building a ship just to suck up hits. I don't know how fast ships take to be "built", but a player won't have fun if they get stuck building a ship that gets destroyed in order to build a ship that gets destroyed, etc.
Even if you net one ship a turn, it shouldn't be too static that a player can determine that if things continue as they do and perform the same number of actions (some sort of loop), then they'll win in X turns.

Keeping the derelict at bay in the beginning while you get the ball rolling would be fine, as long as you could do that, but you'd want a back-and-forth kind of play, not a snowballing one.

The static part I'm concerned about would play out like this:
I spend X turns building one ship to lose to the derelict while I get my deck fine tuned, then...
Build two ships, lose one to Derelict (My fleet is 1 ship at the end of Derelict's turn)
Build two ships, lose one to Derelict (My fleet is 2 ships at the end of Derelict's turn)
Build two ships, lose one to Derelict (My fleet is 3 ships at the end of Derelict's turn)
Build two ships, lose one to Derelict (My fleet is 4 ships at the end of Derelict's turn)
I attack with 4 ships and defeat Derelict
That would be no fun. If it took 3 turns or 10, I wouldn't have any fun if all I had to do was rinse and repeat to victory. I ESPECIALLY wouldn't be having fun if I was losing to an opponent like this.

If a balanced deck lets you build one ship a turn, then you'll be locked in a stalemate as the derelict destroys your ship every turn it attacks. Assuming the cost to build a ship is just playing the three components, then the luckiest hand a Monarch can draw lets you build two ships (assuming a Monarch gets to draw up to seven cards each turn). I'm not a probability expert, my gut tells me this is going to be a tremendous uphill battle with a lot of my turns consisting of being locked in a holding pattern and making zero headway while I feed a ship to the derelict.

You might be able to build a solo mission/tutorial, but I don't think you can base it on one, large ship. I could be completely wrong and there's something I'm not seeing though. :-)

Edit: to be fair, that wouldn't be fun for ME; I would poll the forum here for more opinions.