Thursday, December 20, 2018

My thoughts on the Hyperspace format

FFG introduced the Hyperspace format as the premier competitive format going into 2019. It features a limited list of legal ships and upgrades, currently limited to 2.0 releases. Overall, I think the Hyperspace format is the right way to go, but the format is risky and can potentially produce worse metas. If you're strongly against Hyperspace, I think there are some strong reasons to support it. If you're strongly in favor of Hyperspace, I think there are some reasons to be cautious.

The restricted Hyperspace format is needed to make the game accessible to new players. Asking new players to buy out-of-print ships at some ridiculous mark-up on Ebay or shell out hundreds of dollars at one time for an exiting player's collection is not a reasonable way to build a competitive game. It's disappointing to not be able to use my full collection more often (I think there will continue to be some Extended events), but that's a price I'm willing to pay to make sure the game continues to grow and attract new players.

Some people seem to be hyping Hyperspace up as the answer for all the meta woes. I want to caution against these expectations. It's certainly easier to balance a smaller list of pilots and upgrades. FFG seems to have severe headcount limitations, so focusing their limited resources on a smaller card pool could be what's needed to create a balanced and open meta. However, a restricted format doesn't automatically create more balanced metas. Anyone who's played Magic: the Gathering or any other game with rotating formats knows this from experience. In fact, without a points adjustment, Hyperspace would probably be dominated by Marauder-Han Boba, Luke and Wedge, maybe TIE swarms, and perhaps some Resistance or First Order options. FFG will have to put in the work to keep this format balanced, especially with new releases (which isn't confirmed but will likely be Hyperspace-legal).

A smaller card pool could actually make the competitive play experience much worse. For example, suppose 10% of pilots are overpowered. For a format with 300 pilots, that means 30 pilots are viable at the top levels of competition. You can probably find at least one list you enjoy playing that has reasonable matchups against the best tournament lists (this has been my experience for most of 1.0 and the extended 2.0 format). For a format with 100 pilots, that means only 10 pilots are tournament-viable, and these don't have to be evenly-distributed across factions or archetypes. Your favorite faction might not have enough strong pilots to fill out even one tournament list. Your favorite archetype might not have any strong pilots in it. Your favorite archetype might only have a couple strong pilots which you don't like. You're less likely to find a list you enjoy playing that has reasonable matchups against the best tournament lists.

The solution would be to balance the game so more pilots are tournament-viable. If the top 30% of pilots are strong in the 100-pilot format, then we're back to the same number of viable tournament lists as the 300-pilot format with fewer traps. Even better, the gap between a strong pilot and an average one could be smaller. But again, this won't happen naturally; FFG will have to do the work to broaden the field of competitively-viable ships.

Part of the risk is the long 6-month time frame between rotations and point cost adjustments. On one hand, stability is good for creativity. I'm not motivated to list-build when I know the point costs will all change in a month. On the other hand, if there's a bad meta, it could take a long time for it things to change. These risks could be reduced if FFG committed more frequent reviews (maybe bi-monthly) for emergency point adjustments and/or bans. These reviews should be extremely limited and changes should only be made to the most egregious offenders. Most of these reviews shouldn't produce any changes, but having that safety net would go a long way to reducing the concerns of having a lopsided meta.

Overall, I think Hyperspace is the way to go simply to keep the game accessible. I just hope FFG puts in the work to keep an open meta with a restricted card pool.

Tuesday, December 18, 2018

Some remaining long-term concerns with X-Wing 2.0

X-Wing 2.0 is lots of fun and much better than the original edition. With the first point cost adjustment, I'm hoping to see some of the stronger options nerfed and more viable tournament-level options. Still, while a point adjustment can fix a lot of problems, there are some remaining systemic issues with X-Wing 2.0. These issues limit the design space and create list-building traps, and fixing these will require something beyond simple point adjustments.


There isn't enough variation in attack values

A 2-dice primary deals about half as much damage as a 3-dice primary, ranging from two-thirds against 0 green dice to one-third against 3 green dice with focus. A 4-dice primary deals about 60% more damage than a 3-dice primary.

These massive gaps between the available attack options severely limit the available design space. There are three stats to adjust a ship's durability, and extra points of agility only create differences of 15%-20% in average damage taken. There's only one stat which affects offense, and with rare exceptions, it can only take on two values.

As a result, ships feel "samey" on offense. Ships fly and take damage very differently, but without further upgrades, it's very difficult to make ships hit harder without making them too strong. I could never get over how the intimidating Decimator hits just as hard as the cute Mist Hunter or tiny Attack Shuttle. The mechanics simply don't fit the fantasy.

Worse, the low damage output of 2-attack ships makes it very hard to build lists around them. They're almost always relegated to being support ships. A list with only 2-attack ships can even auto-lose to tanky ships like Luke or Boba Fett before the game starts. Expensive ships with 3 attack suffer from similar problems where they can't put out enough damage to justify their point cost.

This is in theory possible to fix, but there were several systemic options that could have been implemented for the new edition that are now difficult to implement. Attempts to fix this now will create mandatory upgrades, confusing new players and forcing all players to buy (possibly multiples of) specific expansions.


The upgrade model is wrong

The benefit of upgrades depends on the ship's strength and durability. For example, a tanky ship gets to use an offensive upgrade for more turns, and a defensive upgrade is more valuable on a ship with higher damage output. The cost of an upgrade is a flat number. This is a big problem.

And no, variable costing is unlikely to fix this problem. Shield Upgrade is much better on Soontir Fel than on an Academy Pilot and it's better on Luke Skywalker than on Wedge Antilles. For variable costing to solve this problem, the cost of upgrades probably needs to be a percent of the ship's base cost, but this is too cumbersome for list-building.

This mismatch creates two problems. First, ships of different strengths can take the same upgrade. As an example, consider Shield Upgrade on a Blue Squadron Escort X-Wing compared to on Wedge Antilles. There are two approaches FFG can take and both create list-building traps. Shield Upgrade could be balanced for Wedge Antilles but too expensive for the Blue Squadron Escort. Outside of special synergies, this effectively makes Shield Upgrade a trap for any ship with a damage:durability ratio worse than Wedge Antilles. Alternatively, Shield Upgrade could be balanced for the Blue Squadron Escort and ships that make better use of the upgrade would pay a cost for their modification slot. This makes Shield Upgrade mandatory on those ships, which again creates a list-building trap.

The second problem is this mismatch encourages ships to either be run with no upgrades (outside of special synergies) or loaded with upgrades. Outside of dice caps, upgrades not only scale with the base ship's effectiveness but also with other upgrades. If you just gave your ship an upgrade to make it hit harder, then a defensive upgrade protects more value. If you make your ship harder to kill, it now gets more use out of an offensive upgrade. If upgrades are balanced for fat ships, they'll be too expensive to throw as a single upgrade on a ship. If upgrades are balanced to be the only upgrade on a ship, then fat ships will get too much from them.

It's important to note this isn't just a problem where some upgrades are stronger on some ships than others. For example, Juke is especially strong on the TIE Phantom and Whisper in particular. I don't think it's ideal, but it's not a big deal to make Juke a mandatory upgrade for them. This problem is systemic and affects all upgrades across all ships.

There is a relatively easy fix. Ships and pilots could get a flat discount or premium for all upgrades (similar to the 1.0 Vaksai title). For example, the Academy Pilot could get a 1-point discount on all upgrades while Dash Rendar must pay a 2-point premium on all upgrades. This means naked Dash can be balanced without making him a monster with upgrades, and those same upgrades can also be balanced on weaker ships. Costing each individual upgrade differently for each ship is too complicated, but the added complexity of this system is low enough so list-building remains feasible.


Effects that have wildly different values across matchups are impossible to balance

Suppose the meta is 50% list A and 50% list B, and an upgrade provides 2 points of value against list A and 42 points of value against list B. How much should that upgrade cost?

22 points sounds like the obvious answer, but there's a big problem with this. Against list A, taking the upgrade gives you a 20-point disadvantage and you're probably going to lose. Against list B, taking the upgrade gives you a 20-point advantage and you're probably going to win. By costing the upgrade at its average value, the game has turned into a coin flip decided by the pairings.

The solution is to overcost the upgrade, but this has its own problems. Let's say the upgrade is costed at 32 points. Now it becomes a trap. Someone might take this upgrade and effectively put himself at a disadvantage against the field. Even worse, someone playing list B might run into that player and take a random loss from their 10-point disadvantage.

The best thing to do with these upgrades is to make them so expensive they're effectively removed them from the game. High-enough point costs can be enough to discourage even newbies from taking it, thus also preventing random auto-losses from the pairings.

Thankfully, most effects don't have quite such a high swing in effectiveness across different matchups so this problem is more limited, even when those effects can be taken in multiples. Low amounts of matchup-dependent power is fine as long as in-game decisions can still overcome it.

Unfortunately, FFG has shown no fear of creating effects that differ drastically in power based on matchups: see Trajectory Simulator, pre-maneuver repositioning, and both versions of Rebel Luke Skywalker. Ahhav is a more recent example of this, although his gap in effectiveness is only ~10 points. The problem of large initiative bids and first player deciding games is a result of this systemic issue.

FFG's attitude seems to be to throw out cool effects and sort them out later with point cost adjustments, but I hope they realize some things are extremely difficult to balance through points and to stay away from them in the future.