NOVA Squadron Radio – Episode 3 “Alex Davy from FFG, Building X-Wing Communities, and Running Tournaments”

By EvilEd209, in X-Wing

I really like Alex's design approach to game balance, it is exactly the same that I would have, and is the same philosophy I have with house rules. (Granted House Rules obviously allows for much greater flexibility.)

I'm still convinced that I have better tools at my disposal (MathWing) for getting a good first-cut estimate of ship value, especially looking older wave releases. It has accurately predicted the low performance of the TIE Advanced and A-wing, and more recently: the high performance of the ACD Phantom and Falcon + Z-95s, low performance of generic E-wings and Defenders, and sharp drop in use of generic X-wings. E-wings and Defenders still have some time to gain traction, but I strongly suspect that even after Worlds we won't see much generic E-wing and Defender use, they're priced just a little too high. The model's certainty on E-wing cost in particular is quite high, since the ship has common functionality with other ships, making it easier to baseline against.

What's interesting is that they are on their 3rd team for design / balance, and it sounds like they took over for wave 5. Hopefully balance continues to improve, while still increasing the overall depth and ship capabilities. It is a difficult tightrope to walk...

So I have a question: Your math is base on 100 point squads? Is it possible the generic E's and TIE/D's fit better in Epic? I ask because I haven't had a chance to play Epic yet, but I'm dying to. I've been coming up with a few squad lists for when I can make this happen, which have included some of those fighters.

The math shouldn't significantly change going from 100 point to larger squads. The basic underlying assumption is that a ship's value is proportional to the square root of its: (attack power times durability). At 100 points, the curve starts to diverge from (A*D)^0.5 for ships valued at 30+ points, because the ships keep dealing out 100% of their damage until they are completely dead. For smaller point value ships (like TIE Fighters) you can knock them off faster, and immediately decrease your opponent's damage output. For this reason I actually use an exponent of ^0.52, and I plan to revisit this again later with a more precise solution based on an analytical solution of a differential equation with non-continuous first derivative. (Technical way of describing the above scenario of a ship still doing 100% damage while it still only has 1 hull left.)

Anyway, onto your question, there are multiple factors that come into play with larger point values that can change the curve from simply being a square root:

  • It is easier to consolidate all your forces together if your ships are more expensive. In a 100 point game this isn't so much an issue, but in a 300 point game with 12+ ships your focus fire starts to become less than optimal since the ships are physically spread out. Advantage: more expensive ships.
  • In a 300 point game, a 30 point ship is now only 10% of your squad, not 30%. So the actual value is going to be much closer to the pure square root ratio. I.e. a 3/3/3/3 ship is going to last at least one round in a 100 point match, but in an epic match enough stuff can focus fire it at once that it can die in one round, so the ship loses the "boost" that it gets in a 100 point match. Advantage: less expensive ships.

I haven't run numbers on comparing these two factors, but I speculate that they about cancel each other out, and if anything, slightly favor the less expensive ships once it degenerates into a furball after the initial pass.

Edit: Of course, certain pilot abilities (like Rexler Brath) are obviously better in Epic play, but I'm just talking about the base ships.

Edited by MajorJuggler

I really like Alex's design approach to game balance, it is exactly the same that I would have, and is the same philosophy I have with house rules. (Granted House Rules obviously allows for much greater flexibility.)

I'm still convinced that I have better tools at my disposal (MathWing) for getting a good first-cut estimate of ship value, especially looking older wave releases. It has accurately predicted the low performance of the TIE Advanced and A-wing, and more recently: the high performance of the ACD Phantom and Falcon + Z-95s, low performance of generic E-wings and Defenders, and sharp drop in use of generic X-wings. E-wings and Defenders still have some time to gain traction, but I strongly suspect that even after Worlds we won't see much generic E-wing and Defender use, they're priced just a little too high. The model's certainty on E-wing cost in particular is quite high, since the ship has common functionality with other ships, making it easier to baseline against.

What's interesting is that they are on their 3rd team for design / balance, and it sounds like they took over for wave 5. Hopefully balance continues to improve, while still increasing the overall depth and ship capabilities. It is a difficult tightrope to walk...

So I have a question: Your math is base on 100 point squads? Is it possible the generic E's and TIE/D's fit better in Epic? I ask because I haven't had a chance to play Epic yet, but I'm dying to. I've been coming up with a few squad lists for when I can make this happen, which have included some of those fighters.

The math shouldn't significantly change going from 100 point to larger squads. The basic underlying assumption is that a ship's value is proportional to the square root of its: (attack power times durability). At 100 points, the curve starts to diverge from (A*D)^0.5 for ships valued at 30+ points, because the ships keep dealing out 100% of their damage until they are completely dead. For smaller point value ships (like TIE Fighters) you can knock them off faster, and immediately decrease your opponent's damage output. For this reason I actually use an exponent of ^0.52, and I plan to revisit this again later with a more precise solution based on an analytical solution of a differential equation with non-continuous first derivative. (Technical way of describing the above scenario of a ship still doing 100% damage while it still only has 1 hull left.)

Anyway, onto your question, there are multiple factors that come into play with larger point values that can change the curve from simply being a square root:

  • It is easier to consolidate all your forces together if your ships are more expensive. In a 100 point game this isn't so much an issue, but in a 300 point game with 12+ ships your focus fire starts to become less than optimal since the ships are physically spread out. Advantage: more expensive ships.
  • In a 300 point game, a 30 point ship is now only 10% of your squad, not 30%. So the actual value is going to be much closer to the pure square root ratio. I.e. a 3/3/3/3 ship is going to last at least one round in a 100 point match, but in an epic match enough stuff can focus fire it at once that it can die in one round, so the ship loses the "boost" that it gets in a 100 point match. Advantage: less expensive ships.

I haven't run numbers on comparing these two factors, but I speculate that they about cancel each other out, and if anything, slightly favor the less expensive ships once it degenerates into a furball after the initial pass.

Edit: Of course, certain pilot abilities (like Rexler Brath) are obviously better in Epic play, but I'm just talking about the base ships.

So hopefully this is telling me I need to play more and gather anecdotal evidence.

In all seriousness, thanks for the answers

but I'm just talking about the base ships.

To add to this...

A ship that's 2 points too expensive is still 2 points too expensive regardless of how many points you have to spend. It may be smaller % of your total points, but it's still an issue.

Unless you're already using 12 of a given ship, and need to fill in a few points, you're still going to be better off saving those 2-3 points to buy more ships then pay for a overpriced one. In some ways in a large game it's perhaps even more noticeable.

Lets take Academy Pilots vs Obsidian Pilots. 12 AP's cost 144 points, 12 Obs cost 156. If you don't need/want that extra 2 PS then you just spent 12 points that could be used on upgrades or better pilots on other ships.

Or in other words, just because you have more money to spend, doesn't mean you should throw some of it away.

Major Juggler = Math Wizard! Very interesting points my friend!

Major, this is great stuff. Can you provide a link to, or post, the efficiency of the various ships?

Major, we are gong to have to get you on the show sometime to talk about the Regional breakdowns you have been running. If that's cool with you.

So hopefully this is telling me I need to play more and gather anecdotal evidence.

Playing more is ALWAYS the answer! :D

But actually it's true: play testing is the ultimate test of game balance. The issue is getting a large enough sample size, which quite honestly is extremely hard to do as a developer before release. We have several weeks worth of Regionals data, and the general consensus is that we still don't have enough data. Now imagine trying to playtest and assign point value with even less data than that. The point of using Math wizardry is to at least avoid making the most heinous errors *cough* TIE Advanced *cough*.

Major, this is great stuff. Can you provide a link to, or post, the efficiency of the various ships?

It's a "work in progress", but it is still pretty good. The 2nd post has the efficiency of each ship, which I break down by:

  • stat value (just based on attack / defense / hull / shields), aka "jousting" value
  • overall efficiency (includes everything including dial, actions)

http://community.fantasyflightgames.com/index.php?/topic/100360-using-lanchesters-square-law-to-predict-ships-jousting-values-and-fair-point-values-work-in-progress/

You still have to be aware of the limits of the model: it is difficult or impossible to have confidence in the "overall efficiency" of a ship that has unique capabilities, like a K-turn, because it requires using coefficients that are unique to that ship, so you have no comparative baseline. But about half the ships all share the same functionality, so the model should be pretty accurate for those. Generally higher efficiency = more successful competitive use.

Edit:

Major, we are gong to have to get you on the show sometime to talk about the Regional breakdowns you have been running. If that's cool with you.

I would be glad to sometime. PM me or email me at [email protected] and we can set up a time. Let me finish setting up my office first, we just moved and things are chaotic. :)

Edited by MajorJuggler

Neither the Download link nor the streaming link are working for me still...

Using Chrome on Android phone.

I had my webmaster look at it. The RSS feed is working, the download link, and the streaming... maybe your phone doesn't like the RSS feed somehow. I'll keep digging...

I had my webmaster look at it. The RSS feed is working, the download link, and the streaming... maybe your phone doesn't like the RSS feed somehow. I'll keep digging...

I tried looking at it on my phone last night, and ended up just using my laptop instead since I couldn't figure it out. The direct playback link at the bottom of the page wasn't on the phone but was on my laptop. I was also using Chrome on Android. Chrome on windows works fine. :)

Great show! The guests were excellent and the discussion was very interesting.

I haven't had any difficulties getting the show on my android through the Podcast Addict app.

Great show! The guests were excellent and the discussion was very interesting.

I haven't had any difficulties getting the show on my android through the Podcast Addict app.

Happy to hear! We could not find anything wrong on our end.

Thanks for the help Tervlon! And very happy you enjoyed the show!