Yes, I don't think they ever tried the triple u boat build with R4. I think it got by them.

I appreciate Alex's honesty (this was a great interview btw), but i'm once again baffled how they repeatedly miss such obvious builds in playtesting and balancing. The JM5K combo is just the latest example.

I mean deadeye+r4+ordnance was discovered literally the day the ships hit Vassal, it's such a straightforward synergy. There seems to be some sort of issue with the balancing process.

External playtesters are hired for their ability to not leak information, and not necessarily on their play skill.

FFG should probably try to hire a top player (multiple regional wins or better) on a full time position to assist them with it.

I suggest you read the playtest credits on the ships and see if any names ring a bell...

It's obviously not working. As I said, they need to get some really good people on it full time and not as unpaid volunteers.

Yes, I don't think they ever tried the triple u boat build with R4. I think it got by them.

I appreciate Alex's honesty (this was a great interview btw), but i'm once again baffled how they repeatedly miss such obvious builds in playtesting and balancing. The JM5K combo is just the latest example.

I mean deadeye+r4+ordnance was discovered literally the day the ships hit Vassal, it's such a straightforward synergy. There seems to be some sort of issue with the balancing process.

External playtesters are hired for their ability to not leak information, and not necessarily on their play skill.

FFG should probably try to hire a top player (multiple regional wins or better) on a full time position to assist them with it.

I suggest you read the playtest credits on the ships and see if any names ring a bell...

It's obviously not working. As I said, they need to get some really good people on it full time and not as unpaid volunteers.

This might sound counter intuitive, but working as someone who deals with QA on software projects, this is absolutely not what they need. If you've been involved with playtesting from the start, it becomes hard to recognise what's wrong vs what's been made so much more right.

I've heard it informally referred to as the 'dogfood problem': if all you've eaten is dogfood for six months, even a crappy steak is going to be bliss... and so too in playtesting, once you've seen too many iterations it becomes hard to recognize the one's available in this version.

The better solution is to, very carefully and selectively, assemble a 'final pass' QA group who see none of the process but the planned-to-send-to-printers version, and are given a week or two to fiddle around and suggest builds to the developers.

Note that this group gets much less influence per se than the originals - but hopefully they catch the missed combos, the overpriced hulls, and the like.

Trouble is, now you've got two different playtesting groups. And the latter is bordering on 'inside information' in terms of knowing what's coming. Who'd want to be in the first group? That's hard work with less refined product; it's not *fun*, per se.

Professional software vendors use early-adopter clients, who accept slightly shonky software in return for greater influence. Some wargames take the approach of releasing beta rules to communities for playtesting and feedback before the actual printed articles come out for legal play.

FFG, who sell to enthusiast markets while working hard to keep each new release a 'neat surprise' has no such luxuries on hand.

It's a difficult problem, no doubt about itdand - I'm not sure FFG can really solve it without sacrifice to the current model of business, somewhere.

I know a few playtesters, while bless their hearts, sometimes they even didnt know the exact official ruling on a card during playtesting, which is supremely dumb.

From what I know and hear, and from 4TLT and 3Scouts, it seems that FFG's playtesting process is particularly poor and the feedback received is not detailed enough to make fixes.

Also. I'm very much against their current idea of "make things from the screen seem super strong!!!" -> see Falcon, Ghost, soon Ups Shuttle. Large ships have too much synergy. Mararoo was a good pilot, but no range restriction made it harder to foresee.

It seems to me from the interview that they didn't test the most prevalent combo on the JMs (deadeye, OCR, and torps).

Sounded to me like they completely missed the combo. I am curious too what they think the Scout counter is as well.

So, quad TLT again? FFG really needs to reconsider their testing procedures.

I know a few playtesters, while bless their hearts, sometimes they even didnt know the exact official ruling on a card during playtesting, which is supremely dumb.

From what I know and hear, and from 4TLT and 3Scouts, it seems that FFG's playtesting process is particularly poor and the feedback received is not detailed enough to make fixes.

Also. I'm very much against their current idea of "make things from the screen seem super strong!!!" -> see Falcon, Ghost, soon Ups Shuttle. Large ships have too much synergy. Mararoo was a good pilot, but no range restriction made it harder to foresee.

Manaroo is fine. Dengaroo is mostly nuts thanks to a couple synergies that came in Dengar's expansion . I'm a little, shall we say, astonished Overclocked Astromech is a thing they decided to release.

It seems to me from the interview that they didn't test the most prevalent combo on the JMs (deadeye, OCR, and torps).

Sounded to me like they completely missed the combo. I am curious too what they think the Scout counter is as well.

So, quad TLT again? FFG really needs to reconsider their testing procedures.

I know a few playtesters, while bless their hearts, sometimes they even didnt know the exact official ruling on a card during playtesting, which is supremely dumb.

From what I know and hear, and from 4TLT and 3Scouts, it seems that FFG's playtesting process is particularly poor and the feedback received is not detailed enough to make fixes.

Also. I'm very much against their current idea of "make things from the screen seem super strong!!!" -> see Falcon, Ghost, soon Ups Shuttle. Large ships have too much synergy. Mararoo was a good pilot, but no range restriction made it harder to foresee.

Manaroo is fine. Dengaroo is mostly nuts thanks to a couple synergies that came in Dengar's expansion . I'm a little, shall we say, astonished Overclocked Astromech is a thing they decided to release.

the OCR mechanic is fun, big advantage (unlimited focus) vs big risk (unlimited stress). Dengaroo just kind of breaks the risk aspect and that is another thing playtesting could have picked up. At this point i hardly believe they had that combo during playtesting.

Also i still believe giving Scum all those 1pt super upgrades was a mistake - i get what they were trying to do (give scum a boost), but as of now Scum has some of the strongest upgrades in the game for cheap - Zuckuss, 4Lom,Boba,OCR. The downsides to these cards pale in comparison to their strengths and 1pt is an insult compared to some of the more expensive crew cards.

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

Complain about rexlar, not Han!

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

I think the problem with Dengar's ability is pairing it with the Jumpmaster's dial.

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

... Hans ability is pretty sweet. It's not "two attacks in a round" sweet, but it's a solid entropy mitigator, aka Firepower buff, and is arc and range agnostic.

Complain about rexlar, not Han!

It's Rexler, not rexlar. There, I complained about him

Rexler is fine. His ability's not amazing, but when it works, it's very, very good. Better than Defender!Maarek, anyroad.

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

Let's be honest though, Dengar actually costs 45 points. Nobody runs him without the title with the exception of the rare torpedo boat build.

Yeah, that's true on Scum getting boosted beyond reason. I'm most surprised/irritated by the undercosting of the Jumpmasters and Dengar's ridiculous buffing. A 9-skill pilot with one of the best abilities in the game in a great moving and sturdy ship for a mere 33 pts, and for a character with no established piloting skills.

It's probably the most egregious example of power creep in the game (compare to Han's practically useless ability or Corran Horn's limitation), breaking the biggest rule in the game for no downside.

Still, good discussion with Alex, I always enjoy listening to him speak.

Let's be honest though, Dengar actually costs 45 points. Nobody runs him without the title with the exception of the rare torpedo boat build.

And then some. I can get Dengar below 55.

He is great, but he is super fragile. 2green does not a soontir make.

Yes, I don't think they ever tried the triple u boat build with R4. I think it got by them.

I appreciate Alex's honesty (this was a great interview btw), but i'm once again baffled how they repeatedly miss such obvious builds in playtesting and balancing. The JM5K combo is just the latest example.

I mean deadeye+r4+ordnance was discovered literally the day the ships hit Vassal, it's such a straightforward synergy. There seems to be some sort of issue with the balancing process.

External playtesters are hired for their ability to not leak information, and not necessarily on their play skill.

FFG should probably try to hire a top player (multiple regional wins or better) on a full time position to assist them with it.

I suggest you read the playtest credits on the ships and see if any names ring a bell...

It's obviously not working. As I said, they need to get some really good people on it full time and not as unpaid volunteers.

This might sound counter intuitive, but working as someone who deals with QA on software projects, this is absolutely not what they need. If you've been involved with playtesting from the start, it becomes hard to recognise what's wrong vs what's been made so much more right.

I've heard it informally referred to as the 'dogfood problem': if all you've eaten is dogfood for six months, even a crappy steak is going to be bliss... and so too in playtesting, once you've seen too many iterations it becomes hard to recognize the one's available in this version.

The better solution is to, very carefully and selectively, assemble a 'final pass' QA group who see none of the process but the planned-to-send-to-printers version, and are given a week or two to fiddle around and suggest builds to the developers.

Note that this group gets much less influence per se than the originals - but hopefully they catch the missed combos, the overpriced hulls, and the like.

Trouble is, now you've got two different playtesting groups. And the latter is bordering on 'inside information' in terms of knowing what's coming. Who'd want to be in the first group? That's hard work with less refined product; it's not *fun*, per se.

Professional software vendors use early-adopter clients, who accept slightly shonky software in return for greater influence. Some wargames take the approach of releasing beta rules to communities for playtesting and feedback before the actual printed articles come out for legal play.

FFG, who sell to enthusiast markets while working hard to keep each new release a 'neat surprise' has no such luxuries on hand.

It's a difficult problem, no doubt about itdand - I'm not sure FFG can really solve it without sacrifice to the current model of business, somewhere.

Actually, neither of these approaches will work! The fundamental problem is that a game such as X-wing cannot be reasonably balanced by simply playtesting it among 100 high level players and hoping to get an accurate consensus opinion. It is much easier to play the game effectively than it is to balance the game effectively, and the latter requires a very specific background and set of technical skills.

You cannot accurately balance a game like X-wing by relying on playtesters' opinions.

The designers need to be able to do 2 things:

1. Have a solid understanding of the underlying mathematical mechanics of the system that they have constructed and use this to establish proper point costs for new pilots and upgrades.
2. Be able to perform deep dive technical analysis on analytical game data.

The playtesters' job should be to:

1. Uncover new combinations of cards and tactics that are not immediately obvious (i.e. Dengaroo).
2. Generate game data between high-level players for the designers to analyze.

FFG does not have any employees with the background to do Developer Job #1 (aka MathWing). And you can't do Developer Job #2 correctly without knowing how Developer Job #1 works.

The playtesters, in theory, should be able to perform Playtesting Job #1 right now. Unfortunately capturing data for Playtester Job #2 would require a tool more advanced than Vassal (which playtesters aren't allowed to use anyway, for obvious reasons), or recording all gaming sessions and then having an FFG employee go over every round of play and compile all the pertinent details (not feasible).

Some playtesters will attempt to perform Developer Job #1 at some level. This in and of itself is not a bad thing, and can actually help to perform Playtesting Job #2. The problem is when neither the playtesters OR the developers know how to perform Developer Job #1, then you have the blind leading the blind, and the final product will suffer as a result.

This is a problem with the entire tabletop gaming industry, not just FFG.

MJ, considering that I don't think the NFL has someone like this for when they propose rule changes, I sadly doubt FFG will ever have anything like this.

Not that it wouldn't be useful to make a better game.

MJ, considering that I don't think the NFL has someone like this for when they propose rule changes, I sadly doubt FFG will ever have anything like this.

Not that it wouldn't be useful to make a better game.

NFL rule changes are largely political not technical, it's a slightly different animal.

Blizzard is a great example of a company that does game balance right. It's more difficult to do it as a tabletop gaming company because you can't as easily "patch" fixes, but the same principle applies. Blizzard can actually get away with not being as solid on Developer Job #1, and just brute-forcing it with Developer Job #2, solid playtesting, and continuous patches.

For tabletop gaming you really need to do everything that Blizzard is doing, AND you need to understand the fundamentals even better, in order to get it "right" the first time.

I love the lines about what ordinance could have been, and wonder what X-Wing 2.0 will look like when it's released in 6 years...

You cannot accurately balance a game like X-wing by relying on playtesters' opinions.

The designers need to be able to do 2 things:

• Have a solid understanding of the underlying mathematical mechanics of the system that they have constructed and use this to establish proper point costs for new pilots and upgrades.
• Be able to perform deep dive technical analysis on analytical game data.

The playtesters' job should be to:

• Uncover new combinations of cards and tactics that are not immediately obvious (i.e. Dengaroo).
• Generate game data between high-level players for the designers to analyze.

Your summations are clear and definitely useful to constructively ponder the complex practice of X-wing game design.

I know some playtesters. They are good X-wing players. They would not do a good job if they were in charge of costing new pilots and upgrades.

As far as I can tell, X-wing game designers can't even do simple compare and contrast with existing pilots and upgrades. It's very obvious that little or no math is used because of all the inconsistency.

Chopper (crew) 0 point -- Do an action while stressed. Take a damage

Unkar Plott (crew) 1 point -- Do an action after bumping. Take a damage

Extremely similar abilities. Yet they are costed differently?

R3 astromech is weird in so many different ways.

Yet in the same wave, Black Market Slicer Tools is wonderful and costed appropriately.

Pretty interesting Podcast, thank you!

You cannot accurately balance a game like X-wing by relying on playtesters' opinions.

The designers need to be able to do 2 things:

• Have a solid understanding of the underlying mathematical mechanics of the system that they have constructed and use this to establish proper point costs for new pilots and upgrades.
• Be able to perform deep dive technical analysis on analytical game data.

The playtesters' job should be to:

• Uncover new combinations of cards and tactics that are not immediately obvious (i.e. Dengaroo).
• Generate game data between high-level players for the designers to analyze.

Your summations are clear and definitely useful to constructively ponder the complex practice of X-wing game design.

I know some playtesters. They are good X-wing players. They would not do a good job if they were in charge of costing new pilots and upgrades.

As far as I can tell, X-wing game designers can't even do simple compare and contrast with existing pilots and upgrades. It's very obvious that little or no math is used because of all the inconsistency.

Chopper (crew) 0 point -- Do an action while stressed. Take a damage

Unkar Plott (crew) 1 point -- Do an action after bumping. Take a damage

Extremely similar abilities. Yet they are costed differently?

R3 astromech is weird in so many different ways.

Yet in the same wave, Black Market Slicer Tools is wonderful and costed appropriately.

I guess I should have qualified that yes, something akin to MathWing would be even better. But if you don't have math, you really do want a two-stage process.

But the point about comparative abilities is a good one, and where you really want some critical thinking: a scyk is a TIE fighter with a couple more greens, a shield instead of a hull, and an extra point of PS. Should that have really been 2pts extra? Heck, it's only slightly 1pt better. MathWing might tell you this too, but even with intuition there's clearly a problem on hand.

No, what I'm talking about is 'how did Uboats with R4+GC+Protons+Deadeye get missed in playtesting': The answer is from iterations upon iterations of testing, whilst clear eyes spotted it literally within the first hour of the Contracted Scout being spoiled. We didn't even know about Overclocked yet!

A solid grounding in MathWing would be ideal. Two-tier QA would also be nice, and marginally more likely to happen - because it's still volunteer based.

Though I still don't understand how R3 made it through to release.

I'm willng to bet Unkar Plott will see more play than Chopper. You have more control over whether you are stressed, and therefore avoid Chopper's drawback, but Unkar Plott is Dauntless in crew form, rendering you immune to being blocked.

Well I think people are also missing a big point.... $$$.

Hey a little power creep in the Jumpmaster might (PS3 scum with an EPT???) assure a lot of these ships to be sold for a rather obscure ship. Better make it juicy! Besides they killed the 4TLT meta which was not exactly a bad thing.

Well I think people are also missing a big point.... $$$.

Hey a little power creep in the Jumpmaster might (PS3 scum with an EPT???) assure a lot of these ships to be sold for a rather obscure ship. Better make it juicy! Besides they killed the 4TLT meta which was not exactly a bad thing.

I honestly don't think this is the motivation - or if it is, it's a really bad implementation. FFG have so far demonstrated a good understanding of long term thinking in their game, they've tried to keep it as balanced as they could, because sure, you can make a tonne of money in the short term by releasing overpowered ships every wave, but before too long the tournament scene evaporates because it's no fun churning to the next OP thing every 3 months, and your game dies with it.

Keep releasing reasonably balanced stuff and the tourney scene is a lot healthier.

They made two significant missteps with this though IMO - Palp, and the Scout.

Though I still don't understand how R3 made it through to release.

I won't disagree that R3 Astro is a bit naff, but think of it from FFGs PoV, in the grand scheme of things (I think Alex has said this in an interview before):

It's much better to release a naff/over-costed upgrade than an OP one.

If it truly is OP, you might have to come back and fix it later, if it's a bit rubbish, you can always create synergies later on.

Also R3 is a defensive buff, and since C-3P0, Autothrusters and Palp, I think FFG have been very nervous around releasing more defensive upgrades.

Though I still don't understand how R3 made it through to release.

I won't disagree that R3 Astro is a bit naff, but think of it from FFGs PoV, in the grand scheme of things (I think Alex has said this in an interview before):

It's much better to release a naff/over-costed upgrade than an OP one.

If it truly is OP, you might have to come back and fix it later, if it's a bit rubbish, you can always create synergies later on.

Also R3 is a defensive buff, and since C-3P0, Autothrusters and Palp, I think FFG have been very nervous around releasing more defensive upgrades.

You're not wrong, but there's so many naff astromechs that we were kinda hoping they'd have learned their lesson by now. Further, and worse, if you want X-wings to be fixed, a nontrivial portion of that is giving them something useful other than R2 astromech for the rookies, and in that sphere, there's plenty of design space still untouched. Or if you must go for a 2pt defensive option, I adore Fickle's idea of letting you spend a focus to remove an opponents focus - whether this shuts down deadeye or not is a tricky question of timing and even then, largely beside the point - sometimes you're more worried about eliminitating 1-2 more hits coming your way than you are about boosting your own mediocre greens to usefulness.

Still, I guess there's next time. We've still got R4, R5, and R6 right?

Well I think people are also missing a big point.... $$$.

Hey a little power creep in the Jumpmaster might (PS3 scum with an EPT???) assure a lot of these ships to be sold for a rather obscure ship. Better make it juicy! Besides they killed the 4TLT meta which was not exactly a bad thing.

I honestly don't think this is the motivation - or if it is, it's a really bad implementation. FFG have so far demonstrated a good understanding of long term thinking in their game, they've tried to keep it as balanced as they could, because sure, you can make a tonne of money in the short term by releasing overpowered ships every wave, but before too long the tournament scene evaporates because it's no fun churning to the next OP thing every 3 months, and your game dies with it.

Keep releasing reasonably balanced stuff and the tourney scene is a lot healthier.

They made two significant missteps with this though IMO - Palp, and the Scout.

Sold a lot of Raiders and Jumpmasters with those 'missteps' though, didn't they?

You guys are reading too much of the wrong stuff into that interview. If your takeaway from what he said is that the play testers missed the combo then you should listen again. That's not what he said, it's not something he implied, and--particularly given how obvious the combo was--it's not at all a safe assumption.

"We were hoping the cost would keep it under control because it's not a cheap platform" around 36 minutes.

He then very clearly implies that the problem was the design team under rated the build.