Jump to content
Sign in to follow this  
Ayleron

Inefficiencies in new ships

Recommended Posts

 

SithBorg - huh? You have multiple ambiguous pronoun references there. Are you trying to say that the generic X-wing is OK?

 

No. The issue with the generic X-wing and B-wing is fairly obvious by this point. But, that hasn't stopped people from making a good amount of the unique X-wings work. And apparently even the generics. A freaking Red Squadron (not even the rookie) made the cut at Nova Open. So, I fail to see why people won't be able to make the T-70 work. Especially in a TLT heavy enviroment and access to Autothrusters. 

 

Because it is the gospel truth.

All it says is that in a joust amongst equally skilled players, certain ships are more points efficient and likely to come out on top. People that don't like this like to pretend that all ships are balanced equally and you just, 'need to fly ship X right, it doesn't suck'.

Also, the Alpha Squadron Pilot with Autothrusters slapped on it for 20 points is a good deal. 5 of them at once is a decent squad. The issue is that not a lot of people have 5 individual Interceptor packs to run it, and that Imperial Aces doesn't include Alpha and Avenger Squadron Pilots. The other issue is that squads of generics either get blown away by stuff like Corran, or they're perceived to (and rightly so).

I don't think MJ ever proclaimed that it can account for fat turrets having hypermobility.

 

 

shrug

 

The math is an important facet. But, I still maintain that it isn't the full story of the game. I look at some of the weird squads that have done well (Europeans remain really crazy in their builds, from what I've seen), and I can't help but think there is more to the story. 

 

It isn't the full story, but it's the one we can have extended discussions about on the forum.  "YEAH I OUTFLEW THIS GUY" or "I ROLLED SOME REAL GOOD GREEN DICE" don't make for much interesting discussions.

Share this post


Link to post
Share on other sites

 

 

SithBorg - huh? You have multiple ambiguous pronoun references there. Are you trying to say that the generic X-wing is OK?

 

No. The issue with the generic X-wing and B-wing is fairly obvious by this point. But, that hasn't stopped people from making a good amount of the unique X-wings work. And apparently even the generics. A freaking Red Squadron (not even the rookie) made the cut at Nova Open. So, I fail to see why people won't be able to make the T-70 work. Especially in a TLT heavy enviroment and access to Autothrusters. 

 

Because it is the gospel truth.

All it says is that in a joust amongst equally skilled players, certain ships are more points efficient and likely to come out on top. People that don't like this like to pretend that all ships are balanced equally and you just, 'need to fly ship X right, it doesn't suck'.

Also, the Alpha Squadron Pilot with Autothrusters slapped on it for 20 points is a good deal. 5 of them at once is a decent squad. The issue is that not a lot of people have 5 individual Interceptor packs to run it, and that Imperial Aces doesn't include Alpha and Avenger Squadron Pilots. The other issue is that squads of generics either get blown away by stuff like Corran, or they're perceived to (and rightly so).

I don't think MJ ever proclaimed that it can account for fat turrets having hypermobility.

 

 

shrug

 

The math is an important facet. But, I still maintain that it isn't the full story of the game. I look at some of the weird squads that have done well (Europeans remain really crazy in their builds, from what I've seen), and I can't help but think there is more to the story. 

 

It isn't the full story, but it's the one we can have extended discussions about on the forum.  "YEAH I OUTFLEW THIS GUY" or "I ROLLED SOME REAL GOOD GREEN DICE" don't make for much interesting discussions.

So, what. Sable's flight school things aren't interesting

Share this post


Link to post
Share on other sites

Then we as posters should work harder, figuring out ways to use a ship's intangibles to maximize the situations where it has the advantage is the essence of outflying someone, and is substantially more important than getting an extra 2% efficiency.

Or 20% as is the case with the t70. We arent talking about 1 or 2 percent shift, that would indeed be able to be covered with creative work, but a significant gap in efficiency.

Edited by Ayleron

Share this post


Link to post
Share on other sites

I think the T-70 is 3% less efficient than the t-65, the FO is much less efficient being 7.5% behind the ln.

I meant between the b-wing and t70 you are most likely correct in those accountings. But the again why is there such a gap between tie and FO

Edited by Ayleron

Share this post


Link to post
Share on other sites

 

I think the T-70 is 3% less efficient than the t-65, the FO is much less efficient being 7.5% behind the ln.

I meant between the b-wing and t70 you are most likely correct in those accountings. But the again why is there such a gap between tie and FO

 

It's not my maths just repeating what i remember from MJ's work.

Share this post


Link to post
Share on other sites

MathWing assesses statline efficiency: the four numbers on the side of the card, the point cost and the combat actions. It's fairly sophisticated maths, but it's still hamstrung as a direct measure of ship quality by the inability to quantify the dial.

What MathWing measures is jousting efficiency: you remove maneuvering from the equation by assuming all ships are equal, then you average how good they are in the dice war over all the situations that are likely to come up.

MathWing also doesn't account for the different forms of defence in the game: Once per round (OPR) and Once per attack (OPA). Health and Agility are good examples, health is similar to an OPR defence and Agility is OPA. OPA is stronger against lots of attacks, whereas OPR will probably serve you better against fewer, larger strikes. MathWing only gives you a single efficiency number, it doesn't tell you which targets a ship is best against.

Jousting Efficiency doesn't tell you how good a ship is, it tells you how much you've got to get out of the ship's unquantifiables (dial, the more complex pilot abilities and the reposition actions) to match or exceed a more statline efficient ship. Turrets have awful statline efficiency, but they're not overpriced: they get a lot more favourable shots than a flat jouster like a Headhunter would. The TIE interceptor is another example of a ship which is lacking in the statline efficiency department but very effective because it can dictate an engagement to its advantage, usually by getting shots with no return fire.

MathWing only detects overcost when you have two ships with comparable unquantifiables and wildly different statline efficiencies.

Mathwing very much _does_ recognize the difference between, say, 6hp behind 3 green dice and 6hp behind 2 green dice.

Mathwing very much _does_ attenp to quantify how good one dial is when compared against another.

Mathwing even attempts to quantify how valuable repositioning actions/abilities are, and how valuable it is to not have to account for such.

I don't think- I could be wrong- that mathwing accounts for how much player skill is required to get maximum efficiency from a given list.

What Mathwing can't do is tell me how I can play better. But it can help me figure out if I'm playing a bad list, or if I'm playing a good list poorly. It's always one or the other in my case. :)

Share this post


Link to post
Share on other sites

Maybe the X-Wing T70 is overcosted on purpose.

Deliberately breaking a ship so you can sell fixes later? I've a higher opinion of FFG than that.

I think Davies said that the T-70 was deliberately created with the same efficiency as the T-65 because anything they do to help the T-65 woukd also apply to the T-70. They don't want to accidentally make the T-70 OP when making Wedge merely Good.

Share this post


Link to post
Share on other sites

 

I don't think- I could be wrong- that mathwing accounts for how much player skill is required to get maximum efficiency from a given list. 

Skill can't be accounted for by maths it's a variable not a fixed integer, it can be affected by how you slept the night before if your having health problems etc.

 

The Skill can give a player the ability to make even sub par ships do wonders, it's a powerful force that defies the odds.

Share this post


Link to post
Share on other sites

The Phantom has the exact same statline and thus Mathwing predictions pre and post-nerf (for those new to the game, the Phantom nerf only changed when it's decloak happens).  So, Mathwing alone (at least, the public model we know of) predicts that the Phantom would perform just as well before and after this change.

 

Not true - and this is an important distinction: mathematics predicts that the statline efficiency of Whisper is the same before and after the nerf. What has changed is the ship's positional capability, which directly affects its ability to attain the absolute "required efficiency" to break even with a bunch of low PS mooks like TIEs or B-wings. MathWing in that thread won't tell you exactly how much value to expect out of the post-nerf decloak, but it does tell you how effective you need to be at it to break even.

 

Does this mean the shield upgrade is actually over costed. If I take a T65 X-Wing and add shield upgrade for 4 points I actually pay too much, ignoring the free engin upgrade for now.

 

As answered upthread -- yes, for the overwhelming majority of cases. It may or may not have been intended that way, but it is the reality.

Edited by MajorJuggler

Share this post


Link to post
Share on other sites

Then we as posters should work harder, figuring out ways to use a ship's intangibles to maximize the situations where it has the advantage is the essence of outflying someone, and is substantially more important than getting an extra 2% efficiency.

Or 20% as is the case with the t70. We arent talking about 1 or 2 percent shift, that would indeed be able to be covered with creative work, but a significant gap in efficiency.

My point is that instead of trying to optimize the last few efficiency points of a list because of its the easiest thing to discuss, we should spend more effort discussing intangibles, because that has more bearing on who wins games.

As an aside, what is the T70 20% less efficient than?

Share this post


Link to post
Share on other sites

 

 

Then we as posters should work harder, figuring out ways to use a ship's intangibles to maximize the situations where it has the advantage is the essence of outflying someone, and is substantially more important than getting an extra 2% efficiency.

Or 20% as is the case with the t70. We arent talking about 1 or 2 percent shift, that would indeed be able to be covered with creative work, but a significant gap in efficiency.

My point is that instead of trying to optimize the last few efficiency points of a list because of its the easiest thing to discuss, we should spend more effort discussing intangibles, because that has more bearing on who wins games.

As an aside, what is the T70 20% less efficient than?

 

The only generic 20% more efficient than the 83-85% efficient i think i heard MJ say is the PS2 Z-95. The two ships are hardly comparable, though.

Share this post


Link to post
Share on other sites

 

The Phantom has the exact same statline and thus Mathwing predictions pre and post-nerf (for those new to the game, the Phantom nerf only changed when it's decloak happens).  So, Mathwing alone (at least, the public model we know of) predicts that the Phantom would perform just as well before and after this change.

 

Not true - and this is an important distinction: mathematics predicts that the statline efficiency of Whisper is the same before and after the nerf. What has changed is the ship's positional capability, which directly affects its ability to attain the absolute "required efficiency" to break even with a bunch of low PS mooks like TIEs or B-wings. MathWing in that thread won't tell you exactly how much value to expect out of the post-nerf decloak, but it does tell you how effective you need to be at it to break even.

 

Does this mean the shield upgrade is actually over costed. If I take a T65 X-Wing and add shield upgrade for 4 points I actually pay too much, ignoring the free engin upgrade for now.

 

As answered upthread -- yes, for the overwhelming majority of cases. It may or may not have been intended that way, but it is the reality.

 

The gist of FFG's reasoning from my perspective is that they price defensive upgrades more heavily than offensive ones in general.  Palpatine's defensive use is what makes him cost 8 points, for example.  C-3p0 is an outlier, but the power of his ability is that it's totally action independent- rec spec+jan ors or simply doing an evade action with the MF title is actually more effective defense, but c-3p0 works no matter what.

 

Notice how flight instructor, ysanne, chewie, r2-d2 clock in over 3 points, while powerful offensive upgrades like Vader and K4 security end up being only 3.  Gunner's easily the most expensive but its action independence is what makes it powerful.  Threepio's cost is a bit of an outlier.  I don't think FFG really considered people guessing zero, if I had to guess.  There are very few defensive EPTs in the game at all.  Lone wolf is an outlier here as well.

 

It even comes down to the dice design.  There are more hits/crits than evades.  FFG wants offense to beat defense, and I think the game bears that out.  This is why hull/shield are so expensive given their effects.

Share this post


Link to post
Share on other sites

The T-70 X-wing is actually less efficient from a stat line efficiency perspective (jousting value).

 

...How?

The T-70 pays fewer points per HP (24/6 < 21/5), and it's got the same offensive stats.

 

Stat line efficiency only cares about 5 variables: cost, attack, agility, hull, shields. The T-70 gains 1 shield but it costs 3 more points.

 

3/2/3/2 is worth just under 18 points. T-65 jousting efficiency is therefore about:

18 * (1 + 1/24) / 21 = 89.3%

 

3/2/3/3 is worth about 20 points. T-70 jousting efficiency is therefore about:

20 * (1 + 1/24) / 24 = 86.8%

 

Boost on the T-70 is nice, but will not be enough to make the PS2 viable. This is easy to predict because the PS1 TIE Interceptor also has boost, and has a better jousting efficiency of ~90%, but is basically extinct now.

 

Poe is another story. I haven't done the math on him yet.

But with the numbers you've run, it seems like the 25-point T70 with IA and an R2 Astromech should be closer to efficient. It's basically 25 points for a 3/2/3/4 statline, along with all of those greens and the Tallon Roll.

Share this post


Link to post
Share on other sites

 

The Phantom has the exact same statline and thus Mathwing predictions pre and post-nerf (for those new to the game, the Phantom nerf only changed when it's decloak happens).  So, Mathwing alone (at least, the public model we know of) predicts that the Phantom would perform just as well before and after this change.

 

Not true - and this is an important distinction: mathematics predicts that the statline efficiency of Whisper is the same before and after the nerf. What has changed is the ship's positional capability, which directly affects its ability to attain the absolute "required efficiency" to break even with a bunch of low PS mooks like TIEs or B-wings. MathWing in that thread won't tell you exactly how much value to expect out of the post-nerf decloak, but it does tell you how effective you need to be at it to break even.

 

What you are saying in the first part of your last sentence was what I was trying to say above.  Mathwing says that the required efficiency of pre-nerf Whisper is 168.6%.  Mathwing says that the required efficiency of post-nerf Whisper is 168.6%.  Mathwing says nothing about how the nerf affects how well Whisper can achieve this required efficiency of 168.6%.    

 

That is, Mathwing can tell us how much a ship's non-statline abilities need to be used to achieve its required efficiency, but tells us nothing about how likely those non-statline abilities can help it achieve that required efficiency.  Thus, Mathwing only tells us part of the story when considering ships that have significant non-statline abilities.  And clearly, with the example of the Phantom nerf, the part of the story that Mathwing misses can have a large effect on the real-world effectiveness of the ship.    

Share this post


Link to post
Share on other sites

But with the numbers you've run, it seems like the 25-point T70 with IA and an R2 Astromech should be closer to efficient. It's basically 25 points for a 3/2/3/4 statline, along with all of those greens and the Tallon Roll.

 

25 points @ 7HP behind 2AGI is still only around 90% - 91% jousting efficiency though. We're back to the "but the TIE Interceptor has the same jousting and boost" argument. The T-70 is a little better off than the TIE Interceptor if you put autothrusters on both because it's a smaller percentage of overall cost on the T-70, but it's still a little underpowered.

Share this post


Link to post
Share on other sites

Oh geez, mathwing debates again.

 

The thing to remember is that MathWing doesn't say whether a ship is good or not on the table. It says how well it needs to do compared to its cost for a given price. A poor efficiency doesn't make a ship in and of itself bad, it asks how much you need to get out of it for its price.

 

The Defender is a phenomenal ship, with a great statline and a funky dial, the likes of which we've never seen before or since. It's just too expensive.

 

The Z-95 is pretty rubbish as a ship. It's just cheap enough that no-one cares. ;)

Share this post


Link to post
Share on other sites

Not being susceptible to one horrible defensive roll is pretty nice, though, as is an increased resistance to extremely powerful attacks that a high HP ship provides.

Edit: that was in regards to the T70.

Edited by Biophysical

Share this post


Link to post
Share on other sites

 

Mathwing is useful, but I have to agree that a lot of people forget that just slamming the most efficient things they could find on the table isn't enough. Dials are very difficult to properly value, since the same maneuver on two different dials isn't equally useful--would you care all that much if your Headhunter got green sharp turns? Would you care if your Interceptor had white turns at all speeds? Since stress-shedding isn't equally useful to every ship, and different actions are better/worse depending on base stats, Mathwing can give misleading results. That's before we factor in pilots like Corran, who are weak when taken as they are but get scary as hell with the right upgrades. 

 

There are some ships with action bars, dials, and statlines that are made to go straight at something and shoot it out- that's your TIE fighter, Z-95, T-65 X-wing, B-wing and BTL Y-wing.  In this type of ship, jousting is very important to them.

 

I understand that, I'm just saying that a lot of ships and pilots are harder to sort out. The T-70 is still an open question, since it can probably outmaneuver several of the other jousters. It looks like it should joust but more so than a lot of the jousters, it can choose not to. 

 

Mathwing also has a harder time accounting for ships like the Interceptor that live and die by their ability to avoid getting into a straight fight. For ships like them comparing dials and actions if much more relevant to what they want to do. 

Share this post


Link to post
Share on other sites

Mathwing also has a harder time accounting for ships like the Interceptor that live and die by their ability to avoid getting into a straight fight.

As I understand it, Mathwing predicts that the Interceptor generics will not be played a lot, and this prediction is borne out by the statistics. So there does not appear to be a particular problem here.

Even the true 'anomalies', i.e. ships that should not be popular according to mathwing but are chosen a lot, can often be explained by things that are obviously outside the scope of the mathematical model. This shouldn't be a problem either, any model has limits.

Share this post


Link to post
Share on other sites

There is more to the story, but you've constantly got people looking for a single truth that they can throw at others and say, "No, THIS is right!".

 

No, you have people with a bizarre hatred of mathematics and an incessant need to believe in magic and the power of the unseen.

 

These people see numbers, numbers that are explained quite clearly and verifiable by all, and inexplicably refuse to believe what they are seeing. They then allege that the numbers are actually a misguided trick meant to represent an all-encompassing scalar value for all ships in the game where higher-order ships necessarily defeat lower-order ships in any and all circumstances.

 

It's maddening to read over and over again.

Share this post


Link to post
Share on other sites

 

Let me be a bit more specific. While I understand they want the "elite" pilots to be better enough to be desirable over the fenerics, but I feel like the generics continue to be overcosted. Poe is a monster, as is Corran, but the e-wing and the t70 generics are definitely points heavy.

Not nearly as badly as the TIE Defender is. But i understand your point. E-Wings aren't really viable in squads that are more than 3 ships. T-70s are a good in-between point between an E-Wing and a normal X-Wing, however. Not prohibitively expensive for the generics, enough to fit 4 in one list with enough for an astro and soon to be IA, the PS4 generic has an EPT slot,  and the nameds are well priced for the role they can serve. 31 points for Poe isn't bad at all, especially considering that he's only 2 points more than Wedge, and 1 point less than Etahn A'baht, a PS5 E-Wing pilot. T-70s can be outfitted for jousting or arc-dodging, your choice. They seem like a more general ship than either one is. X-Wing isn't quite powerful enough, E-Wing is too expensive to fit in the list, T-70 is juuuuust right. 

 

I'm going to be calling T-70's the Goldilocks ship now.

 

Except defenders' generics are still viable in multiple lists

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...