Attempts to reverse engineer the ship point formula

By khan48, in Star Wars: Armada

X-post from reddit!

https://www.reddit.com/r/StarWarsArmada/comments/3edpcp/attempting_to_reverse_engineer_the_point_formula/

Basically i just used Hull values, total shield values, and damage and ran a regression against points. The resulting coefficients for hull, shield and damage give you a formula.

I used two types of damage values in the regressions. One was the total ship damage probability of all the dice summed (Its .75 for red and blue and 1 for black). This in effect, doesn't weight the dice at all.

The other was the inch distance for each range times the damage probability for the dice color times the number of dice of each color for a ship. Then i divided it by 10 to put it near the same scale as hull and shield values. This is an an effort to weight the red and blue dice which do indeed seem more valuable on ships due to their greater distance and versatility.

Spreadsheet is here. https://docs.google.com/spreadsheets/d/1SJAZ92a-zZule1eeKagKYk1iw-F8rHdiS9LQowGjLpA/edit?usp=sharing

TLDR; Formulas i came up with:

  • Points=1.13+(Shields x .72)+(hull x 1.96)+(dmgDst x 5.15)

  • Points = -5.6+(shields x 3.55)+(hull x 5.33)+(dmg x .694)

  • ISD estimated cost 99 to 107

  • Home one estimate 92 to 99

Resources

Other ramblings: There is so much that goes into the ships that i haven't accounted for that its probably pretty off base. The defense tokens, ship size, anti-squadron, and ship speeds are probably the biggest factors i'm not accounting for that have a significant effect on ship costs.

Other factors include the ship's arcs and the distribution of the shields and attack dice. I theorize though that they don't effect point cost much, rather just how the ship is played.

There is also probably a baseline ship that they use and tune point costs from there. My bet is its either at 25 points or 50 points. Using the sheet 3 graph, a 50 point ship would have 5 hull, 7 shields total, about 8-10 dice distributed around.

The other big thing to keep in mind are size of arcs and place!emt for shields and dice. A die in a large arc would be worth a lot more than one in a tiny arc. See Nebulon B for an example. Or the VSD and its huge front arc.

That being said we know exactly how much 2 Black die are worth in a huge VSD front arc.

Braces are expensive.

Think we also need to work out the dice vs tokens effectively first before we figure out what every ship's point value is.

*pokes head in* yup smart folk

even if you accounted for defense tokens and size, you still wouldn't arrive at an accurate formal because interactions with Squadrons completely throw costs out of whack because they are very difficult to quantify (could either add a tie fighter's worth of damage, or a vader's per point)

still, I think it'd be a fine endevour to try and introduce a form of "mathwing" to Armada, if only to see how cost-effective ships are at a VERY base level before factoring in the complexities of ship + squadron interactions just so we could see what distinguishes an exemplary ship-to-ship combatant

but if I may, I love this game's ships + squadron interactions because even an inefficient combat ship would instantly find a way to contribute to the game rather than go the way of the Scyk or X-wing in...X-wing

I don't even understand why anyone would do this. But then again I despise math. I just have fun blasting enemy ships to bits, and I've yet to need any mathematical formula to do so. But to each his own!

I don't even understand why anyone would do this. But then again I despise math. I just have fun blasting enemy ships to bits, and I've yet to need any mathematical formula to do so. But to each his own!

Best set up? Never needed any math for that. Only lost once so far, and never used math to figure this out. Just read the cards, it's obvious what works and what doesn't. You don't need math to figure it out. To me this would totally kill any enjoyment of the game, had I to figure all this out. But as I said, I know some people enjoy this, so more power to them. To me it's wasting time, instead of working up all these numbers they could be getting in more games....lol.

I wish it was that simple.. "Just get in more games." :P

Well I always enjoyed knowing that my X space marines dealt an average of X damage and we're worth X amount of points.

Same with hero's and such. I then knew how worth it they are.

Knowing that my ship is X points because of its weapons and what I am wasting not using its other arcs and such is information I like.

I also like knowing what future wave ships could reasonably cost

Hehe...I will know what future wave ships cost when they come out. Patience, my friend.

As for how worthy something is, like I said, you don't need math for that. Play it a couple of times and that will tell you all you need to know. Plus all these formulas can be deceiving, as they do not take player skill into account. You can give the most awesomely cost effective ships to a newb, and he will still get his butt kicked by a veteran, even if the vet is using a less than optimal build.

I never build for optimum performance, I build a list I think will be fun to play. Just so happens I'm really good with this game...lol.

I know this I am giving you an explanation on why people like doing such things.

I can roughly estimate points for ships. Such as the MC30, while I think it will be about the cost of a Gladiator BUT it has no brace so it is likely more like a Nebulon-B.

Oh I know why people do it. I just think it's wasting valuable gaming time, hehe. As I said, some people love their math, more power to them. I'll continue to not use math, and still kick butt. Because the Empire rules!

Not everyone gets to game all the time sometimes they do side projects like this on their breaks and such.

I also don't know if I should feel offended. . . Are the things I do then a waste of time since it is not gaming?

If you have a (more or less) working formula you can get own(custom) ships into the game, ships that will not (so soon) be produced by FFG. Plus some people want to spend time with their game even if there's no opportunity to game, so they do the math thing.

If you have a (more or less) working formula you can get own(custom) ships into the game, ships that will not (so soon) be produced by FFG. Plus some people want to spend time with their game even if there's no opportunity to game, so they do the math thing.

not sure if anybody said it already but currently there are more variables then ships so you will be left with degrees of freedom.

Wave 2 should help with that.

Let's see if I can nail the base things down.

Speed values

Top speed

Command value

Squadron value

Engineering value

Hull strength

Shields

Anti-squadron dice

Arc sizes

Battery dice

Base size

Did I miss anything?

Things like interactions with other pieces on the board are tested through play testing and then effects base values which then impacts points

Best set up? Never needed any math for that. Only lost once so far, and never used math to figure this out. Just read the cards, it's obvious what works and what doesn't. You don't need math to figure it out. To me this would totally kill any enjoyment of the game, had I to figure all this out. But as I said, I know some people enjoy this, so more power to them. To me it's wasting time, instead of working up all these numbers they could be getting in more games....lol.

Its hard to get in games between work, and other commitments!

Braces are expensive.

I did something similar for squadrons, and since they had less stats i included tokens for hero squads. Braces were indeed quite expensive. (3-5 points)

The other big thing to keep in mind are size of arcs and place!emt for shields and dice. A die in a large arc would be worth a lot more than one in a tiny arc. See Nebulon B for an example. Or the VSD and its huge front arc.

That being said we know exactly how much 2 Black die are worth in a huge VSD front arc.

The arc portion is a great point. My attempt weighting die value by "area damage" could be further refined by breaking it down for each arc. You could calculate the "area damage" of each arc based on the dice, and angle of the arc and add them together.

I think total shields is fine for that calculation as redirect lets you pretty readily distribute damage.

Edited by khan48

not sure if anybody said it already but currently there are more variables then ships so you will be left with degrees of freedom.

Interesting to know. Next wave should definitely help lock some of these down with more certainty.

Wave 2 should help with that.

Let's see if I can nail the base things down.

Speed values

Top speed

Command value

Squadron value

Engineering value

Hull strength

Shields

Anti-squadron dice

Arc sizes

Battery dice

Base size

Did I miss anything?

Things like interactions with other pieces on the board are tested through play testing and then effects base values which then impacts points

What upgrades you can equip.