TIE Defender spoiler article is up

By Danthrax, in X-Wing

Those faults in the model come from the things the Lanchester model doesn't account for fully(Upgradeability, the incomplete manuever dial understanding, alternate arcs, etc.) The Defender is the least adaptable ship in its point class...

I've raised those issues in the past, but that's not what I'm addressing now. What I'm saying here is that the most atypical thing about the Falcon and Firespray, from the standpoint of the model, may be that they cost at least 50% more than the next most expensive ship in the game. The jousting model doesn't fit those ships very well, and MJ is currently dealing with that in his "fair value" model using two unique weights. That's shaky methodology from my standpoint, but I understand what he's doing and why.

But with the release of Wave 4 we suddenly have two starfighters in the same neighborhood as the Firespray, which aren't relevant to MJ's rationales for those weights. What I'm saying is that those weights could be obscuring more general model fit issues, with residuals proportional to cost.

Shouldn't we be using the "Salvo Model"?

Probably; one of the principal issues with a model based on Lanchester's Square Law is that Lanchester proposed a set of relationships that operate in continuous time, and X-wing operates in discrete time. In fact, MajorJuggler has expressed a desire to put together such a model when he has the spare time...

But MJ and I are both doc students, and while it's not really hurting anything for me to spend a few minutes every hour checking around message boards, taking 30 hours to put together a simulation using a combat salvo model would definitely have an impact on my real work. I presume he has the same issue. :unsure:

Yeah I agree, the rear arc is great.... but only for one round. Then you usually have to K-turn and you're out of the fight since you're so far out. Occasionally you can 1 turn instead to get something in arc.

I tried writing in a 1-turn on my Firespray dial, but for some reason my opponent still wouldn't let me do it. :ph34r:

And with respect to the jousting efficiency of the Defender, I'd like to point out (again, and gently) that I think the Lanchester's model suffers when applied to very expensive ships--by which I mean I think it's penalizing them more than the game itself does. The Firespray's 85% efficiency in the model doesn't do a very good job of explaining its very acceptable level of performance, and while you've used a unique weight for the rear arc to improve the fit in your fair-value model, that weight (and the similarly unique weight for the Falcon) might be acting to obscure a problem with the overall shape of the curve.

Ha ha ha, good catch. I must have meant 2 turn. :blink: I have been doing a good job at making stupid typos lately.

Re: model fit for expensive ships: yes, there's some room for improvement there at the upper end. The error ranges for Defender / Firespray / YT-1300 are larger than the other ships. It could be solved better by using a higher order equation. It would make for a rather messy equation, but as you say below, its a mostly time issue.

Those faults in the model come from the things the Lanchester model doesn't account for fully(Upgradeability, the incomplete manuever dial understanding, alternate arcs, etc.) The Defender is the least adaptable ship in its point class...

I've raised those issues in the past, but that's not what I'm addressing now. What I'm saying here is that the most atypical thing about the Falcon and Firespray, from the standpoint of the model, may be that they cost at least 50% more than the next most expensive ship in the game. The jousting model doesn't fit those ships very well, and MJ is currently dealing with that in his "fair value" model using two unique weights. That's shaky methodology from my standpoint, but I understand what he's doing and why.

But with the release of Wave 4 we suddenly have two starfighters in the same neighborhood as the Firespray, which aren't relevant to MJ's rationales for those weights. What I'm saying is that those weights could be obscuring more general model fit issues, with residuals proportional to cost.

The biggest issue with pricing the Defender and Firespray, is that they both have something that no other ship has: a rear arc and a K-turn. Inventing coefficients isn't really that meaningful because there is nothing to correlate them to. If another ship had a rear arc or white K-turn, then you would have another data point to sanity check against. A post I made in a different thread is pertinent here.

Yes, my Lanchester's based formulas have always only worked best on ships that do NOT have unique capabilities. I would break down the "certainty" into categories something like:

Very high degree of certainty (no unique capabilities):

X-wing

A-wing

B-wing

E-wing

Z-95

TIE Fighter

TIE Advanced

TIE Interceptor

Good degree of certainty (minor unique traits)

Y-wing (turret with 2 base attack)

Medium degree of certainty (significant unique traits)

Millennium Falcon (360 degree arc)

Firespray (rear arc)

TIE Bomber (basically requires ordnance to be useful)

Low degree of certainty (absolutely unique traits, not even remotely similar to any other ship)

TIE Defender (white K-turn requires playtesting)

Phantom (cloak)

Shuttle (no white turns; red 0-stop)

HWK-290 (turret with 1 base attack)

Once we get enough play testing on the TIE Defender, and I update my dial scoring methodology, I'll probably be confident enough in moving the Defender into the "medium" category. But anything less than the first tier fundamentally involves guesswork, since there are no other ships that you can correlate them to. Thankfully, half the ships should have a high degree of confidence, so overall the approach should still have utility.

Shouldn't we be using the "Salvo Model"?

Probably; one of the principal issues with a model based on Lanchester's Square Law is that Lanchester proposed a set of relationships that operate in continuous time, and X-wing operates in discrete time. In fact, MajorJuggler has expressed a desire to put together such a model when he has the spare time...

But MJ and I are both doc students, and while it's not really hurting anything for me to spend a few minutes every hour checking around message boards, taking 30 hours to put together a simulation using a combat salvo model would definitely have an impact on my real work. I presume he has the same issue. :unsure:

That pretty much hits the nail on the head. Working 40 hours a week at my 9-5, and then working on my doc nights and weekends doesn't leave much time. The extent of my X-wing analysis these days is basically keeping the Store Championship data updated, and more useful. That's probably far more useful than theoretical cost predictions anyway, since it's hard data.

I have a couple of quick questions for the more experienced. I apologize if these were covered earlier, 18 pages is a lot to look through lol.

1) predator - I see a lot of talk about this being primary weapon only, but I thought when that was the case, the card would specify "when attacking with a primary weapon" while this card simply says "when attacking". What am I missing?

2) colonel vessery - does his ability mean that, since it's after rolling attack dice. He can gain a target lock then immediately use that lock on that roll, or is it a lock for next turn?

Thanks in advance!

It can be used on secondary weapons. I believe what you're actually seeing is that it doesn't "work" on turrets. Which isn't actually true either.

For Outmaneuver, the defender has to be in your firing arc, and you have to be out of the defenders firing arc. So that means that you can use your secondary ion/blaster turret with it, but only if the defender was in your arc... so the other 280 degrees that the turret works, outmaneuver does not work.

Predator works regardless of primary/secondary/firing arcs.

Further question then, if I use captain Jonus in tandem with it, then would I be able to reroll 3 dice or reroll 2dice then 1 die after that?

You would be able to reroll 3 separate dice. Dice can only be rerolled once, no matter the source of the rerolls.

You would be able to reroll 3 separate dice. Dice can only be rerolled once, no matter the source of the rerolls.

Just to clarify, if Han had a target lock, he shoots, uses his ability to re roll the dice, he then cannot re roll any more dice with the target lock, correct?

Yes.

You would be able to reroll 3 separate dice. Dice can only be rerolled once, no matter the source of the rerolls.

Just to clarify, if Han had a target lock, he shoots, uses his ability to re roll the dice, he then cannot re roll any more dice with the target lock, correct?

Correct.

You would be able to reroll 3 separate dice. Dice can only be rerolled once, no matter the source of the rerolls.

Just to clarify, if Han had a target lock, he shoots, uses his ability to re roll the dice, he then cannot re roll any more dice with the target lock, correct?