Gypsy Update

General Discussion
Dreadfox Games
Posts: 5
Joined: Fri Oct 14, 2011 6:16 pm

Gypsy Update

Postby Dreadfox Games » Fri Jul 06, 2012 3:06 pm

  • Use Magic Device added to the gypsy's class skill list.
  • Charlatan’s Gift (Ex): At 3rd level, the gypsy gains a +4 bonus to Use Magic Device skill checks. This bonus increases to +8 when applied toward spells from a single spell list, and the gypsy does not need to emulate an ability score when making checks related to that spell list. This spell list must be chosen when charlatan’s gift is gained and may never be changed.
Months of playing, testing and listening have revealed one area of the gypsy that could be objectively improved.

Dead Deck Play.

A great deal of the gypsy's power is drawn from the size of her deck. She can discard one card from her deck to draw an additional card if she doesn't like her first draw. She can discard roughly half of her deck to draw one specific card when she needs it. And she can discard a number of cards equal to a spell's level to apply metamagic to that spell with a total disregard for her maximum spell level. These abilities help keep the gypsy's gameplay very active and add a card game experience at the table that supplements the deck-building experience at home. The inherent cost of exercising more control over a powerful but random deck is losing spells per day - a constant balancing act that separates the great players from the good.

When initially designing the gypsy, we gave her Use Magic Device as a class skill for flavor reasons. Once playtesters showed us how powerful the gypsy could be in the hands of a great player, we took it away for balance reasons. UMD is arguably the most powerful skill in the game, and including it in an already large skill list seemed too powerful at the time. The number of skills in a class's skill list is indicative of that class's utility and, in general, indirectly proportional to the class's power at early levels. The bard has 28, the rogue has 21, the inquisitor has 20, the wizard has 16, the cleric has 13 and the fighter has 10. Yes, the wizard is more powerful than the fighter which seems like it should upset this correlation, but that's not the case at lower levels when the value of class skill bonuses are at their peak. With a sorcerer's allotment of spells per day, extra cards from auguries and access to every spell list, the gypsy is probably stronger than the wizard at early levels when it doesn't really matter whether you draw color spray or sleep. So we wanted to keep the gypsy equal to or inferior to the wizard in terms of class skills (16).

What we learned, however, was that this created a "trait tax" and a "feat tax." Virtually every player that we saw play the gypsy well took a trait, a feat or both to enable the reliable use of Use Magic Device. Taxes aren't fun - if any singular trait or feat is so much stronger than the other options as to make it the only sensible option, the class, trait or feat involved needs adjusting. Players should be encouraged to build what they WANT; players should not be required to build what they NEED.

The root of the problem was that the mitigating factors we designed don't do anything for a gypsy that has already depleted her deck. One unlucky manifest destiny and you're done for the day. Cards like Strength and The Star won't help you at that point, because they're already gone. Spells like call lightning and magic jar can't add any longevity to your deck at that point, either, because they're already gone. While we want the use of manifest destiny to be rare for fear of this situation, both for balance reasons and flavor reasons, we don't want gypsies to be effectively stunned until the next day if this should happen. We want them to have a little something they can do to bide their turns until they can reshuffle their deck. Nothing as powerful as drawing cards, but nothing as useless as level 12, 10 Strength, low-BAB scimitar attacks.

This leaves 2 options.

1. Give the gypsy an ability that lets her create a makeshift deck once her deck is empty.

We explored this option at length. Utter reliance on class features and total independence from items makes it very appealing. But this solution creates new problems. Giving the gypsy an inherent ability or augury that lets her refill her deck on a regular basis at low levels makes her far too powerful. She already towers over other casters in terms of absolute power. That power is tempered by chance. (Absolute Power)*(Chance) = (Actual Power). If we call the gypsy 40% more powerful than a wizard, and call her chance of drawing a good spell at any given time 70%, then (1.4)(.7) = .98, i.e. she is 98% as powerful as a wizard (there are a variety of other factors to be considered, the previous numbers are made up and the true values vary greatly based on player skill, party composition and campaign tendencies). The point is, bumping that .7 up to something in the .85 range drastically upsets the power curve, which is what this ability would do in conjunction with draw-card auguries. Its implementation would require a major overhaul to the class, and is something we are very exciting about developing as an archetype for release in October, 2012.

2. Give the gypsy Use Magic Device without taxing her.

UMD is a very good solution. Casters using magic devices pale in comparison to casters using their spell lists. Spells cast from magic devices are relatively weak. Spells cast from magic devices are slow. Spells cast from magic devices are expensive. Imposing these penalties on a gypsy that has run out of spells puts the class right where we want it - it's debilitating enough that you should play your best to avoid running out of spells, but it's not so debilitating as to make you helpless for the remainder of a session once you have. The dynamic is elegant; we only needed to lift the tax.

A free +3 bonus and another free +4/+8 bonus do just that. You can now use magic devices reliably without investing any resources greater than skill points. You can still choose to spend resources to enhance that reliability, but an effective +11 bonus makes this an option, not a necessity. We are particularly excited that this update makes 12 - 14 Charisma more appealing for gypsies with generous point-buy systems, as encouraging charismatic gypsies can only be good.

We know that a +8 bonus to a skill at level 3 seems imbalanced, but the numbers are very misleading. 1 level in any spellcasting class = automatic success, or +19, to UMD checks related to spells on that class's spell list. 3 levels in gypsy + 3 skill points = +10/+14 to UMD checks. The gypsy gets roughly 66% of the benefit* for roughly 300% of the cost^. The class has its own advantages and disadvantages, but in light of the way that we want the class to play, a +8 bonus to Use Magic Device is on the conservative side. We want automatic success with wands to occur around 6-8th level with regards to the spell list a gypsy chooses to favor. Assuming 10 Charisma and 7 skill points^, automatic success occurs at 7th level, which falls comfortably within that range (1 is always a failure when rolling UMD, so it's actually a 95% chance of success, but it's as close as a character can get*).

We had initially feared that adding this extremely powerful skill to the gypsy's class skill list would make her more powerful than we could accept around levels 1-2. But extensive number crunching has given us confidence that UMD is a poor in-combat option at levels this low, even if the gypsy can manage to acquire a wand that early. In most level 1-2 cases, cantrips, spell-like abilities (such as force missile) or even low-BAB attacks are preferable to UMD attempts (assuming you are fighting level-appropriate creatures and only have access to level-appropriate magic devices). And starting around level 3, the deliberate drawbacks of the gypsy begin to surface, warranting their slight skill advantage just as the advantage starts to become applicable.

This update has gotten us even more excited about developing new archetypes, auguries, feats and traits for release in October, 2012. These developments require much more testing and we hope that you are looking forward to them as much as we are.

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest