Jump to content



Photo

New Player question - Explain the TIE Advanced to me?


  • Please log in to reply
58 replies to this topic

#41 Mikael Hasselstein

Mikael Hasselstein

    Member

  • Members
  • 2,231 posts

Posted 24 July 2014 - 11:59 AM

Hey,

 

Again, mad respect on the math. Especially because if you (or somebody else weren't doing it) I'd be tempted to give it a go myself, and I can't afford that kind of time-sink. Also, I haven't spent enough time with this type of mathematics for many years, and it would take me too many hours (days?) to brush up on it all. Also, in the end, I would probably not get as far s you've gotten.

 

Question: do you use specialized software beyond just a spreadsheet to work this stuff out?

 

Yes it matters, but I think the two main reasons that red dice are better than green dice are:

  1. There are 4 hits on attack dice vs 3 evades on defense dice
  2. The attacker is more likely to have a focus token than the defender, making the red eyeballs worth more than green eyeballs.

Between these two factors it is common for the red dice have 6/8 good results vs green dice only have 3/8 or 5/8. Answering your original question is actually more complicated, but the scenario you are describing is essentially that green dice can have "wasted" evades, but red dice can never have "wasted" hits, except when over-killing a target. To answer/quantify that original question, I would have to run the numbers with evade dice having 4 evades vs. 3, and giving both sides the same likelihood of having focus tokens. I might go do that at some point just because I'm curious, but it's purely a theoretical exercise.  :)

 

Yeah, now that I've seen the math for myself, I accept that the whatever factor that might be is simply outweighed by the reality of your point 1. I guess I've just been fixated on the notion that 25% of the time with 2 reds and 12.5% with 3 reds, it doesn't matter how many greens you have. With that in mind, the green dice is conditional on the red dice coming first, and you should therefore not equate die sides on the green as having probability as those on the red. Does the concept of the "wasted green" cover that notion?

 

The math is slightly more involved (I use an exponent of 0.52, not 0.5), but it doesn't really change anything. It's just horribly overcosted. The next closest ship is probably the A-wing, which is, not coincidentally, the #2 least-used ship.

 

I'm sure you've mentioned it somewhere, but does the Chardaan Refit fix the A?

 

Nope, I don't take into consideration the tactical roles, which is why you still need to interpret the results when ships are specialized. One of the best examples of this is the TIE Bomber. Mathematically, its jousting value and overall value are quite good. The problem is that it's jousting value is still slightly lower than the TIE Fighter, so you generally only ever take it to use Missiles/Torpedoes or Bombs. Missiles/Torpedoes are very poor performance for their cost, so that just leaves Bombs. Bombs are... OK but not enough to make you take TIE Bombers at a top Regionals table. So as a result the TIE Bomber doesn't really get used.

 

I'm guessing you would first need an adequate taxonomy of roles. I once did that taxonomy in a forum thread for the roleplaying game, but that was based on the lore rather than a ground-up analysis of stats and implications.

 

Also, as a lover of the bomber, has the introduction of epic and huge ships changed its value at all? From a lore perspective, the bomber shouldn't be an efficient ship in a dog fight, given that (lore-wise) it wasn't intended to be a space superiority vessel, but.. well, a bomber. Conceptually, it should shine against huge ships. Do you know if it does?

 

...uh oh, I see you've replied... and I just got this done. Okay, I'm posting this and then going to read your other stuff.



#42 Mikael Hasselstein

Mikael Hasselstein

    Member

  • Members
  • 2,231 posts

Posted 24 July 2014 - 12:18 PM

Vader is good. Especially with an Engine Upgrade.

 

I agree, that's how I play him too.

 

Does he also make 'Expose'  worthwhile card, given his ability to barrel roll out of arc before using an action to activate expose, or am I just better off barrel rolling and focusing? Focus is just better, right?

 

To add my 2 cents: I ran the damage numbers taking critical hits explicitly into account. weighting double damage crits as 2x, and all other crits as 1.33x. It turned out that shields are worth about only 15% more than hull for a predominantly focus-based economy.

 

Good to know. Thanks!

Could you explain the focus-based economy a bit more? It sounds intriguing.

 

The coefficients would only change if the underling action economy or the kinds of ships remaining (representing a different "meta" of base attack and defense dice) is far different at the end game than in the start of the game. Even if it did, I calculated the ranges for the ship jousting costs based on the meta (not action economy yet), and generally the values don't change all that much. In this case, the change is basically zero, because it's relative to a TIE Fighter, and both have 2 attack and 3 agility. The durability is basically linearly scaled because it has 2 more shields, so the TIE Advanced's relative durability increase over the TIE Fighter is almost completely independent of what's shooting at it.

 

Actually, if anything, it seems to me that your base equation should hold better at the end game than at the beginning, given that your calculations are based on dyadic pairings, right? Or am I misremembering the method you used? The end game is a simpler system as long as we don't try to inculcate the degree to which ships come into it scathed by battle (which they usually do).

 

I'm glad they are keeping everything in print, it's not a collectible game. I'm sure at some point they will do something to buff the Advanced.

 

Well, there's clearly no factoring for taste. :P



#43 dvor

dvor

    Member

  • Members
  • 1,102 posts

Posted 24 July 2014 - 12:19 PM

I'm sure you've mentioned it somewhere, but does the Chardaan Refit fix the A?

No it does not. Because the A-Wing was hardly broken to begin with.

 

 

Proton Rockets might be the first step towards fixing the TIE Advanced.

 

proton-rockets.png


X-wing is played over a series of game rounds. Turn is a type of maneuver.


#44 Mikael Hasselstein

Mikael Hasselstein

    Member

  • Members
  • 2,231 posts

Posted 24 July 2014 - 12:27 PM

I'm sure you've mentioned it somewhere, but does the Chardaan Refit fix the A?

No it does not. Because the A-Wing was hardly broken to begin with.

 

Care to elaborate?

 

MJ has both theory (the abstract math) and empirical data (the tournament data) to back up his point. What do you have to support your argument?

 

Proton Rockets might be the first step towards fixing the TIE Advanced.

 

Hm, it's possible. Is it worth the 3 points?


Edited by Mikael Hasselstein, 24 July 2014 - 12:29 PM.


#45 MajorJuggler

MajorJuggler

    MathWinger

  • Members
  • 3,073 posts

Posted 24 July 2014 - 12:35 PM

Question: do you use specialized software beyond just a spreadsheet to work this stuff out?

 

I use MATLAB because I have it handy for my engineering endeavors (at the moment primarily dissertation simulations), and it is far more convenient to program loops than copy / pasting a bazillion cells in Excel. Not to mention I can make custom functions to my hearts delight, to calculate anything. And the plotting functions are far better.

 

 

Yeah, now that I've seen the math for myself, I accept that the whatever factor that might be is simply outweighed by the reality of your point 1. I guess I've just been fixated on the notion that 25% of the time with 2 reds and 12.5% with 3 reds, it doesn't matter how many greens you have. With that in mind, the green dice is conditional on the red dice coming first, and you should therefore not equate die sides on the green as having probability as those on the red. Does the concept of the "wasted green" cover that notion?

 

Actually, if anything, it seems to me that your base equation should hold better at the end game than at the beginning, given that your calculations are based on dyadic pairings, right? Or am I misremembering the method you used? The end game is a simpler system as long as we don't try to inculcate the degree to which ships come into it scathed by battle (which they usually do).

 

Short answer: I brute-force every single possible attack vs. defense scenario in the game. Then each one of them gets weighted by how likely it is to occur. Range, base dice, action economy to modify dice. I.e. 75% chance to have focus on offense and 50% on defense, range 1-3 bins at something like 25% / 50% / 25%, and base # of attack and defense dice about the same as the Regionals / Store Championship results. So no, it's not a time-variant formula.

 

 

 

Vader is good. Especially with an Engine Upgrade.

 

I agree, that's how I play him too.

 

Does he also make 'Expose'  worthwhile card, given his ability to barrel roll out of arc before using an action to activate expose, or am I just better off barrel rolling and focusing? Focus is just better, right?

 

Expose is terrible. Although in this case you can +1 attack +focus, vs TL+F because Vader has 2 actions. If you could change your dial from 2/3/x/x to 3/2/x/x then the ship value would scale proportional to:

 

(1.75*0.71)^0.52 = 1.12x

 

But it costs you an action. Blah.

 

 

To add my 2 cents: I ran the damage numbers taking critical hits explicitly into account. weighting double damage crits as 2x, and all other crits as 1.33x. It turned out that shields are worth about only 15% more than hull for a predominantly focus-based economy.

 

Good to know. Thanks!

Could you explain the focus-based economy a bit more? It sounds intriguing.

 

I just mean that usually people focus for an action, not target lock. And focus can't generate critical hits, so there's not as much point in having shields. Extra hits do help existing crits get through, and that's inherently calculated.

 

 

 

I'm sure you've mentioned it somewhere, but does the Chardaan Refit fix the A?

 

No it does not. Because the A-Wing was hardly broken to begin with.

 

On the contrary, the A-wing at 17 points is mathematically terrible. At 15 points it will be about as cost efficient as the B-wing, but at a lower point value, and with different mechanics: Boost vs. System Upgrade, and lower stat line and cost.


Edited by MajorJuggler, 24 July 2014 - 01:05 PM.


#46 Vorpal Sword

Vorpal Sword

    O frabjous day!

  • Members
  • 2,759 posts

Posted 24 July 2014 - 12:59 PM

Good points on the dial. I trust you're right on the need for actions for defense to make it more tanky, but does that mean that you've worked out the relative value of shield points to hull points, and shield points are less than 1.33x as valuable as hull points?


One of my default mathematical assumptions (when I say something like "a ship with 3 Attack and focus does an average of 1.53 damage per round to a ship with 2 Agility") is that all damage results are equal. And as a consequence of ignoring the difference between hits and crits, I can't meaningfully represent the difference between hull and shields.

It's not true, of course, but it's close--and, moreover, it's necessary if you want to handle things like average damage numbers without running simulations using a lot of other, more complex assumptions.

The other reason not to do it, though, is that the effect of critical damage can range from 0 to 1, and the effects are highly variable. Crits are relatively unusual events in the first place, and they can easily fail to cause any additional damage: if you have shields, if it's a "bad" crit (like Munitions Failure for a ship with no secondary weapons), or if you would simply be destroyed by the number of damage cards you take, regardless of their additional effects. So the crits that "count" are a relatively small proportion of an already unusual event--so I'm comfortable saying shields are much less valuable than 1.33 hull. (MJ's estimate of 15% is at least a decent ballpark.)
 

I've been reading discussions about 'end game' and how certain ships might be more troublesome than others. I suppose durability and ignorability both play into a ship's probability of making it to the end game. And, as a guy who played a lot of chess when he was younger, I understand how an end game is different from the beginning and middle. But how do the virtues and vices of a ship change as a game grows old? Do the coefficients in MJ's equation slightly alter per round? (MJ, since you're here, what do you think of this, or is that beyond the desirability of parsimony?)


It's probably beyond meaningful quantitative modeling. Picture trying to model the value of chess pieces based on possible board positions in different phases of a match, and then add in the massive increase in complexity of X-wing as compared to chess...

I'd say values do differ substantially, however, even if the precise determination of their magnitude is impossible. To take an obvious example, the Defender is overcosted by MJ's estimate, and even setting that aside I think its dial is too conservative for its current cost (at least on the generic pilots). But in a one-on-one situation, its value almost certainly grows: the white K-turn offers an advantage in the action economy that relatively few ships can match without upgrades. The Phantom has a similar advantage, in terms of forcing an opponent to sacrifice attack opportunities rather than actions, and the Falcon has another kind of end-game advantage due to its turret.

I hope that they don't do another fix, and not just because I'm determined not to get any more TIE/x1s, or that I don't want to face an opportunity cost. I'm a thematic player. I like the Advanced being 'just Vader'.


Suppose, then, that all you have to go on is the dogfight in the last act of A New Hope. Darth Vader is clearly flying a different fighter than other Imperial fighter pilots, but where did it come from? It has advanced technology, including a targeting computer that Vader relies on despite his Force abilities, and it appears just as polished and "complete" as the more common version. So the only reason to believe that it's unique is that we only see one... but of course it's reasonable to infer that the Imperial military machine is massive, and we only see a tiny fraction of it on-screen.

Once you admit the EU, there are a lot of things that describe Vader's fighter as a prototype--but there are enough of them floating around that it's clear that it saw at least limited production, even if it was never widely adopted by the Imperial Navy, and that Vader was not its only pilot.

I don't want to step on anyone's right to adopt whatever constraints he or she wants with regard to headcanon, but there's no reason you can't have a whole squadron of Advanced x1 on the table while remaining perfectly thematic.

#47 Sithborg

Sithborg

    Member

  • Members
  • 2,426 posts

Posted 24 July 2014 - 01:12 PM

The Refit was needed, but I sort of disagree with the why. The Z-95 is what made the Refit necessary. Before it, the A-wing was the main Rebel filler and/or cheap torpedo option. Both of which is sort of trumped by the Z-95. The Refit allows the A-wing to compete with the Z-95 for filler options.



#48 Winner

Winner

    Member

  • Members
  • 55 posts

Posted 24 July 2014 - 01:30 PM

Is this ship sub-par?  Why?

 

Will anything be done to make the ship more appealing?

 

J

1 word to describe my feelings towards the Tie-Advanced:
Nope

If you play Darth Vader, (in the Tie Advanced x1) then that's acceptable, but normal Tie Advanced? Nope.

You know, if FFG released a new model for the Tie Avenger, and added a boost action and reduced the cost, I'd be absolutely down with that.


Rebel Alliance: 6 [X-Wing], 3 [Y-Wing], 3 [A-Wing], 4 [B-Wing], 2 [HWK-290], 2 [YT-1300], 1 [GR75 Transport], 3 [CR90 Corvette], 7 [Z-95], 2 [E-Wing]

Empire: 10 [Tie Fighter], 2 [Tie Advanced], 10 [Tie Interceptor], 4 [Tie Bomber], 2 [Firespray-31], 3 [Lambda Shuttle], 2 [Tie Phantom], 3 [Tie Defender]


#49 MajorJuggler

MajorJuggler

    MathWinger

  • Members
  • 3,073 posts

Posted 24 July 2014 - 01:34 PM

The Refit was needed, but I sort of disagree with the why. The Z-95 is what made the Refit necessary. Before it, the A-wing was the main Rebel filler and/or cheap torpedo option. Both of which is sort of trumped by the Z-95. The Refit allows the A-wing to compete with the Z-95 for filler options.

 

A-wing usage in Wave 3 Regionals is very low. It could have been given Refit earlier, or even released 2 points cheaper and it would have been fine.


  • Winner likes this

#50 Mikael Hasselstein

Mikael Hasselstein

    Member

  • Members
  • 2,231 posts

Posted 24 July 2014 - 01:51 PM

I use MATLAB because I have it handy for my engineering endeavors (at the moment primarily dissertation simulations), and it is far more convenient to program loops than copy / pasting a bazillion cells in Excel. Not to mention I can make custom functions to my hearts delight, to calculate anything. And the plotting functions are far better.

 

Hehehe, yeah, I fell back into Star Wars during my dissertation-writing. It was a great escape, but I hope you're not letting it get the better of you. YMMV, but the dissertation was a big motivational struggle for me, and it took me too many years to complete.

 

Now that I'm teaching, I have the same motivational problems, but at least I've got a paycheck now. On that note, I won't engage in this enlightening discussion as deeply as my inner nerd would really like me to, so you'll forgive me if I don't comment on every bit you posted, even if it does authentically fascinate me.

 

I also won't hold out hope for Expose.

 

 

It's probably beyond meaningful quantitative modeling. Picture trying to model the value of chess pieces based on possible board positions in different phases of a match, and then add in the massive increase in complexity of X-wing as compared to chess...

 

Quite so, certainly if you're committed to quantitative modeling from the ground up. I don't mind macro-level modeling, and as a social scientist I may be more comfortable with accepting assumptions bout the micro level as you and MJ might be. However, I'm not really sure where I would start on that. I do like the rock<paper<scissors<rock assumption that seems to be going around about builds. I wonder if one might construct an asymmetric network model around that.

 

Once you admit the EU, there are a lot of things that describe Vader's fighter as a prototype--but there are enough of them floating around that it's clear that it saw at least limited production, even if it was never widely adopted by the Imperial Navy, and that Vader was not its only pilot.

I don't want to step on anyone's right to adopt whatever constraints he or she wants with regard to headcanon, but there's no reason you can't have a whole squadron of Advanced x1 on the table while remaining perfectly thematic.

 

I totally see where you're coming from. I also assume that you're right that there were squadrons of them - even if it was just to do the testing of their value. Now, given what we know about the TIE/x1 from this simulation that we love to play, it's clear that many TIE pilots died to bring us this information, that the TIE/x1 just doesn't cut it for the cost of its production. :D

 

Here's where I'm coming from, and it's strictly a personal orientation that may be idiosyncratic. I like stories that don't involve other people's characters. I also like representativeness. So, for me, the Imperial Navy is a huge bureaucracy and it relies on its constraints of availability. TIE/x1s are not in its usual arsenal, nor do expensive hyperspace-capable fighters fit Imperial naval doctrines. My fascination (as a social scientist) is in modeling the Imperial Navy as an institution, and playing that out. It's how I like to nerd out in my mind. It's also why I'm annoyed by the Phantom and the Defender, given how I feel they transgress the notions in my headcanon as you put it.

 

Then again, I like flying Vader, so... I'm not all that dogmatic about the above either. ;)



#51 iPeregrine

iPeregrine

    Member

  • Members
  • 960 posts

Posted 24 July 2014 - 02:11 PM

Short answer: I brute-force every single possible attack vs. defense scenario in the game. Then each one of them gets weighted by how likely it is to occur. Range, base dice, action economy to modify dice. I.e. 75% chance to have focus on offense and 50% on defense, range 1-3 bins at something like 25% / 50% / 25%, and base # of attack and defense dice about the same as the Regionals / Store Championship results. So no, it's not a time-variant formula.


How did you come up with that weighting? Do you have objective data for things like how often a focus is available, or is that just your personal opinion of what the weighting should be? This is a rather important question given how often your numbers are treated as absolute fact instead of just your own analysis.
  • MajorJuggler likes this

#52 MajorJuggler

MajorJuggler

    MathWinger

  • Members
  • 3,073 posts

Posted 24 July 2014 - 02:40 PM

 

I use MATLAB because I have it handy for my engineering endeavors (at the moment primarily dissertation simulations), and it is far more convenient to program loops than copy / pasting a bazillion cells in Excel. Not to mention I can make custom functions to my hearts delight, to calculate anything. And the plotting functions are far better.

 

Hehehe, yeah, I fell back into Star Wars during my dissertation-writing. It was a great escape, but I hope you're not letting it get the better of you. YMMV, but the dissertation was a big motivational struggle for me, and it took me too many years to complete.

 

I'm all but dissertation, and if I can get through just-moving mode, then I can be done in a few months.  :)

 

 

Short answer: I brute-force every single possible attack vs. defense scenario in the game. Then each one of them gets weighted by how likely it is to occur. Range, base dice, action economy to modify dice. I.e. 75% chance to have focus on offense and 50% on defense, range 1-3 bins at something like 25% / 50% / 25%, and base # of attack and defense dice about the same as the Regionals / Store Championship results. So no, it's not a time-variant formula.


How did you come up with that weighting? Do you have objective data for things like how often a focus is available, or is that just your personal opinion of what the weighting should be? This is a rather important question given how often your numbers are treated as absolute fact instead of just your own analysis.

 

 

Short answer: this is the weakest link! I need more data. Looking at vassal logs could be one area. Do you know any volunteers?  :D

 

  • Range data was taken from Worlds 2013 final table (not the above numbers). I need more data.
  • Action economy at 75% attacker focus, 50% defender focus, needs real data.

The next iteration will at least include different meta (base attack / defense dice), and action economy, to widen the possible range of values. If we have more confidence in the action economy then we can narrow down the results.



#53 Red Castle

Red Castle

    Member

  • Members
  • 1,107 posts

Posted 24 July 2014 - 09:47 PM

 

To add my 2 cents: I ran the damage numbers taking critical hits explicitly into account. weighting double damage crits as 2x, and all other crits as 1.33x. It turned out that shields are worth about only 15% more than hull for a predominantly focus-based economy.

 

Good to know. Thanks!

Could you explain the focus-based economy a bit more? It sounds intriguing.

 

Just a little something else to take into consideration regarding shield vs hull. While the damage the shield can take over the hull by avoiding a Direct Hit or Minor Explosion is not that great, there is other critical result that, while not lowering your health, can reduce the ship efficiency: Structural Damage, Weapon Malfunction, Damage Sensor Array, Blinded Pilot, Stunned Pilot, etc, etc...

 

So while it might not save you from destruction, it might help your ship stay in the game. Getting a blinded pilot critical hit on your scimitar right before unleashing your missiles with the help of Jonus feels... bad. Having all your hard turns become red on an PtL Interceptor can screw you up.


I'm a french native player so sometimes, some expressions or meanings might be lost in translation. I mean no disrespect.


#54 MajorJuggler

MajorJuggler

    MathWinger

  • Members
  • 3,073 posts

Posted 24 July 2014 - 09:54 PM

Yes, that's why I weighted "normal" critical hits as 1.33 a regular hit.


  • Red Castle likes this

#55 Red Castle

Red Castle

    Member

  • Members
  • 1,107 posts

Posted 24 July 2014 - 10:01 PM

Yes, that's why I weighted "normal" critical hits as 1.33 a regular hit.

Sorry, somehow missed that part.


Edited by Red Castle, 24 July 2014 - 10:01 PM.

I'm a french native player so sometimes, some expressions or meanings might be lost in translation. I mean no disrespect.


#56 MajorJuggler

MajorJuggler

    MathWinger

  • Members
  • 3,073 posts

Posted 24 July 2014 - 10:58 PM

 

Yes, that's why I weighted "normal" critical hits as 1.33 a regular hit.

Sorry, somehow missed that part.

 

 

No worries, lots of text to read through.  :)



#57 Magnus Grendel

Magnus Grendel

    The Empire Needs You!

  • Members
  • 1,435 posts

Posted 25 July 2014 - 02:20 AM

 

Proton Rockets might be the first step towards fixing the TIE Advanced.

 

Hm, it's possible. Is it worth the 3 points?

 

 

In practice, we shall have to see.

 

That said, it's damn good. when fired off a TIE advanced, it stacks up damn well firepower-wise against the dear-holy-got-what-did-you-just-fire advanced proton torpedo, for half the cost, on a platform that I trust to survive into range to fire them.

 

Because you only need a focus token on your ship, you don't need to spend it, you can spend it to modify the results - kicking the advanced's damage potention from about one-and-a-half hits wiith a focused primary weapon to just shy of four. Vader (or Maarek Stele with Push The Limit) is even scarier due to his focus-and-target-lock shennanigans, dropping a faintly ridiculous, getting closer to four and a half damage - which is 'blow-rebel-fighter-out-of-the-sky' territory.

 

Secondly, focus-enabled rockets are incredibly tactically friendly for low-skill pilots; take a focus action, and use it for defense if you need to or thump whichever poor goon appears in front of you if you don't - unlike a target lock, you aren't pre-designating who you want to shoot at, so (taking a classic rebel squad as an example) you can put one proton rocket into biggs, then, if he's dead, shoot luke, if he's not shoot biggs again.

 

The pairing is a traditional one, anyway - as Maarek Stele in TIE fighter, you spend quite a lot of time in a TIE avenger (essentially the Advanced Mk2) firing proton rockets at people.

 

 

WRT making use of the Advanced's tankyness - taking an entire squad of overcosted fighters seems a bit backwards, but since you have no equivalent of biggs to force people to shoot at you, it does at least avoid presenting a weak spot....



#58 any2cards

any2cards

    Member

  • Members
  • 966 posts

Posted 25 July 2014 - 06:34 AM

1) Yes.

2) Because.

3) Answer unclear. Ask again later.

 

Damn ... this reminds me of playing with my Magic 8 Ball.  Provided the same valuable answers to my life altering questions.  :P


Edited by any2cards, 25 July 2014 - 06:34 AM.


#59 mazz0

mazz0

    Member

  • Members
  • 1,081 posts

Posted 25 July 2014 - 09:10 AM

Let's not forget, of course, that while the TIE Advanced may not be a competitive ship to fly in tournaments, it doesn't mean it's not fun to play with casually!  Plus it may come with cards that you want.  Also I agree that the Proton Rocket will be a good one to stick on a Tie Advanced, particularly Vader (maybe with an Engine Upgrade to make R1 easy to get into)






© 2013 Fantasy Flight Publishing, Inc. Fantasy Flight Games and the FFG logo are ® of Fantasy Flight Publishing, Inc.  All rights reserved.
Privacy Policy | Terms of Use | Contact | User Support | Rules Questions | Help | RSS