Proficiency Dice for 0e?

When running any OSR system, especially one which is based on (or is) the original edition, it comes to head fairly quickly - how do you resolve situations that can't be role-played? Climbing a wall, for example: it's impractical to ask your players to demonstrate on your backyard fence or cliff overhang wall climbing in order to justify the character action. If one of you out there does run a table like that - I would like to participate at least once just to see - but that's beside the point. For Thief specialties, it's easy: there is a chart provided in Greyhawk or usually included by default in your system of choice. For other actions, precedent exists - the rules for kicking a door open, for example are X in 6, as are the rules for Surprise. Logically, it can follow that other things can be done X in 6. 

That's is the approach I had taken in Weapons, Wits, & Wizardry to start. I changed the six-sider to an eight-sider to allow for different growth rates and to facilitate modifying those chances by ability scores - a more agile character is more likely to walk a tightrope, etc. - but having written a post about dice pool resolution, and later having been edified that there's actually TSR precedent in The Forgotten Temple of Tharizdun for ability-based dice pool resolution, I started thinking - could dice pools rolling under abilities serve in the place of skills?

Here is the system that I have in mind and have bounced off my unfortunately sporadic home game:

Proficiency Dice

When advancing a character, the player should have an option to invest experience points into improving their capabilities in a given skill or task, defined as an action which is not role-playable nor directly derivative of another class or character feature, in the form of Proficiency Dice: at increasing cost for each successive as follows:

Die No.	XP Cost
1st	600
2nd	1,800
3rd	5,500
4th	16,500
5th	50,000

Note, these are per die, per proficiency. So a character investing in two dice total - a first in, say, Hiding in Shadows and another first in, say, Climbing Sheer Surfaces, would pay 1,200 XP total - or two times 600: not 1,800.

When investing in dice, the current XP pool of the character is reduced by the amount invested - thus, a character with 5,800 XP which takes a second die in, for example, disabling traps would reduce this XP total from 5,800 to 4,000 and mark the Proficiency Die on its character sheet accordingly.

At the discretion of the referee, a character may not de-level in this manner - that is, a level three character may not invest experience sufficient to fall to level two - in order to advance a particularly expensive proficiency.

Task Resolution

Confluent with dice pool resolution, when executing a task in which the character has proficiency, the normal dice as prescribed by the referee are rolled. In addition, a number of D6 equal to the characters proficiency dice are likewise rolled. Of the net pool, only the lowest are retained. 

Thus, a character attempting to perform a task rated at 3d6 which has a 2 proficiency dice relevant to the task would roll 5d6 total, summing the lowest 3 in the pool, and comparing that result to the relevant Ability score:

• If the result is equal or under the relevant Ability score, the task succeeds. 
• If the result exceeds the relevant Ability score, the task fails.
  

In Practice

In practice, it works. Or, at my table, no one has complained.

However, this assigns more value to the attributes than they typically have in classic editions - understandably, some will be turned off by the idea. Regardless, for the sake of having tried it in play and liked how it felt, ability scores on 3d6 being fairly tightly in the middle, it felt worthwhile to explore the mathematics behind it and see how it balances against the traditional Thief. 

Why buy dice?

Buying Proficiency Dice allows a degree of customization on the part of the character without compromising two big benefits of a nominally class based system. This could be a blog post all its own (maybe a podcast episode?) but such will have to be another entry. 

First, one of the key elements of OSR gaming - defining "OSR gaming" as a romanticized emulation of the experience of the first 10 years of D&D's existence - is the dispensability of characters. You play multiple in a stable - sometimes multiple at a time, if your player group is too small - you generate them quickly and swap them out according to the adventure (or when one of them dies), and you can port them table to table: presuming that the campaigns are running compatible systems. This is why Thief skills in OSR games typically are not modified by anything apart from level. Level needs to be the important deciding factor - among the reasons for which is the interoperability of characters in a campaign. 

Sandboxes are essential to the movement.

Open tables, west marches: these styles of play have taken a forefront in public games.  

With skill dice purchased with XP - you still have the same character on the same track with the same basic abilities - you just have an added bonus in very specific circumstances. Very easy to adjudicate.

Second, having an increasing cost associated with purchasing Proficiency Dice means that you continue to be bound within the level and fighting capability aspect. If I spend 600 XP on a die, it means I can do that thing better, but I sacrifice the rapidity with which I advance in my primary role. Fighting Men who are specialized in dungeoneering activities will fight less effectively than their brethren because they advance less quickly. The difference isn't necessarily that pronounced with the smaller dice, however to become a true specialist, the costs add up and the character noticeably slows down in terms of their contributions to the combats.

Similarly, this prevents the acquisition of too much skill for your level range. I physically can't afford a third Proficiency Die if I have not sufficiently leveled up that the bandwidth between now and next level is sufficient to cover it. In the same way that a level 1 Fighting Man is not going to be able to stand their ground against an Ogre with six hit dice, you know that a level one Thief character equivalent is not going to be picking the pockets of a passing name level Arch Wizard.

But how do I know my chances?

This is a fair criticism. Unless you are an avid statistician, you probably won't know precisely. However, for the sake of compatibility and intellectual honesty, I did a little math to figure out the answer. As follows are the chances of success at mid-range tasks at varying levels of skill as defined by proficiency dice pool:

		Proficiency Dice Added
		0	1	2	3	4	5
Pool: 3	7:	16%	36%	53%	66%	76%	83%
	10:	50%	73%	86%	93%	96%	98%
	13:	84%	94%	98%	99%	99%	99%
Pool: 4	7:	3%	9%	17%	26%	35%	45%
	10:	16%	34%	51%	65%	75%	83%
	13:	44%	67%	81%	89%	94%	96%
Pool: 5	7:	1%	1%	3%	6%	9%	49%
	10:	3%	10%	19%	29%	40%	81%
	13:	5%	32%	48%	62%	73%	96%

Truthfully, I did not do a lot of math - anydice.com did a lot of math - but that is beside the point. Please note, also, that the numbers are approximate and decimals are rounded according to how I was feeling at the moment.

For comparison, here are the Thief percentages as presented in Greyhawk - the initial implementation, compatibility with which I would be most concerned about when running an original edition game:

Thief Level	Open Locks	Remove Traps	Pick Pocket	Move Silently	Hide in Shadows	Hear Noise*
1	15%	10%	20%	20%	10%	33%
2	20%	15%	25%	25%	15%	33%
3	25%	20%	30%	30%	20%	50%
4	35%	30%	35%	35%	25%	50%
5	40%	35%	45%	45%	35%	50%
6	45%	40%	55%	55%	45%	50%
7	55%	50%	60%	60%	50%	66%
8	65%	60%	65%	65%	55%	66%
9	75%	70%	75%	75%	65%	66%
10	85%	80%	85%	85%	75%	66%
11	95%	90%	95%	95%	85%	83%
12	100%	95%	100%	100%	90%	83%
13	100%	100%	100%	100%	95%	100%

* Hear Noise is a chance in six - but a percentile is provided to approximate: e.g. 33% in lieu of 2-in-6.

At first glance, comparing the table, it immediately becomes evident that the dice pool strategy results in much more rapid advancement - in terms of success rates. With a pool of four dice, it looks parallel to start - assuming an ability score of 10, the 16% success rate mirrors the Apprentice (level 1) - however a first proficiency die brings the character in line with a Burglar (level 4)! Additionally, to think on it, a zero proficiency character does not imply Apprentice. Zero proficiency implies that you have not been trained, you have no experience in it, that you are going on instinct. Thus, comparing one Proficiency Die to the Apprentice rank would be more apt. Comparing thusly, 4 dice is entirely too generous, breaking adventures written with TSR numbers in mind by making those challenges far too likely to bypass.

Moving up to a pool of 5 dice, the prospects are bleak for a character with a ability score of 10. While the first Proficiency Die delivers a probability that is under the Apprentice - one third the likelihood for traps and locks, half for picking pockets and moving silently - the second Proficiency Die brings you into Footpad territory, a third into Robber, and the 4th into Cutpurse - giving Burglar a miss. That's not too bad: against a target of 10, five dice delivers - accounting for Proficiency Dice being added to the pool over time, percentage success chances increase proportionally to what would be expected by an 0e adventure. 

But - how many Thieves have a Dexterity of 10?

The classic Thief archetype implies agility. This is addressed in OSR games as a prime requisite bonus- the character will level faster as a Thief if they have a high dexterity. Thus, we should bump ourselves to the success chances at the Ability target of 13: more in line with what we're likely to see at the table. The compatibility here is a bit bleaker - the first Proficiency Die  grants a whopping 32% - in line with a Burglar (again, fourth level Thief). The second Proficiency Die jumps to about a Sharper, the third to a Master Pilfer, on average. So, a character with a 13 Dexterity is going to advance more quickly with Proficiency Dice than a Thief would with a 13 Dexterity.

Come to think of it - though - who said that they have to be one to one?

The point of rescue for this system becomes in the experience balancing. How much experience does a Thief have to have in the original edition to hit those percentages? In all of them - the Thief advances more rapidly than other classes. Where a Fighting Man might require 2,500 XP to hit Level 2, at the same 2500 xp, a Thief has hit Level 3. This continues through the course of their advancement track. Thieves will have low chances to succeed, will have the smallest hit die, but will also hit the higher chances quickly and gain hit dice more rapidly than other party members - allowing them to "catch up" in terms of effectiveness. Hence comes the pricing model prescribed to the dice.

Each Proficiency Die, in invested experience, is adjusted assuming a target 13 Ability score to roll under with 5 dice. Yes, it starts off as a Burglar - but isn't a common complaint for the Thief class that they start off useless? Bumping them to one-in-three is hardly a game breaker and will make the character feel more useful, mechanically - and, as the player will be investing in multiple dice across the board in order to keep up with all of the functions (assuming the player is building a Thief to a T) the XP invested, sacrificed away from fighting capability and/or hit dice, will be proportional to a 4th level Thief: a net zero in terms of total difference.

After the first, each die purchased advances the skill Chance by about two levels - so, a character with a 13 Dexterity that invests in Proficiency Dice going forward will - to achieve the same percentile success rate - again, be parallel in terms of how much XP is needed to the classic Thief. 

Balance preserved

But what is to stop a character from maxing out? From investing entirely in one skill/ proficiency and thereby breaking that part of the adventure?

As mentioned above, the referee is encouraged to disallow de-leveling - thereby preventing too much investment in one place without first advancing in level to appropriately higher position. Will it eliminate all disparity? Will characters intentionally try to be better and one thing versus another? It's possible. But if the goal is to maintain compatibility - allow you to run an 0e, Thief-less game in an adventure that assumes a Thief's existence, this helps. It enables that end while also allowing a modicum of customization: which, for Weapons, Wits, & Wizardry, was a design objective all along.

To Conclude

I presently like where these numbers sit. 

I intend to present this to my players and allow them to try it - hammering out the kinks as we go. The numbers may change, I may or may not update this post, but you can bet that the final version will be in the Weapons, Wits, & Wizardry booklets - assuming it scales as well as it plays at low level.

There are disparities for higher abilities - I understand that - looking at the 16 target range, one Proficiency Die on a pool 5 task has a 90% chance to succeed - but I'm going to wait to see if that's actually a problem before I try to solve it. Though I get that other people who play the original edition enjoy that the Abilities offer very little to the character. That makes sense to me, too. If you desire, try this system out while retaining a static Target of 13 with 5 dice - call it the Lucky 13 rule - that way everyone will remember it. And the character having progressed, the equivalent of leveling up, as a Thief, will be preserved.

So - like it? Hate it? Would love to hear your thoughts- readers and players alike. 

In any case, delve on! 

 

Public domain artwork retrieved from OldBookIllustrations.com. Attributions in alt text.

Weapons, Wits, & Wizardry: Skirmish & Duel Combat

Updated: 10/9/2022

Man to Man

...with Multiples

The primary appeal of Chainmail resolution to me is twofold - first, its simplicity and its expeditiousness; and second, its utilization of concurrent hits to casualty larger than life figures. Where the traditional approach of hit points can be used to abstract both grit and skill - the lesser fighter will take a more grievous injury than the superior fighter who knows how to roll with the punches - it doesn't fit well in the context of the game: giving a character multiple attacks, but requiring multiple hits to "connect", better presents the target as defending itself and scaling that defense according to martial prowess: determined by level of experience.

So, with troop combat - there is speed of dice pool successes; and with man to man - there is the granularity of weapon versus armor: how to combine the two? 

To take a hint from Fantasy Combat resolution (and reverse an idea from Majestic Fantasy for Swords & Wizardry to make it fit!)

Man Equivalence and Attack Pool

Each character should consider its own offensive capability in terms of a pool: this pool is a number of attacks equal to the character’s Man Equivalence and is refreshed each round. During the appropriate phase - Missile or Melee - the attack sequence below is then executed.

Attack Sequence

1. Assign: The character assigns a number of attacks from their pool to a given target, up to their remaining attack pool.
   
   
2. Resolve: Dice are rolled – 2d6 – and the value, after modifiers, is compared to the target number of the weapon versus the armor (or armor equivalence) of the target.
   
   If the target number is met or exceeded, the target suffers a number of Hits equal to the number of attacks allocated in Step 1.
   
   
3. Remove & Repeat: The number of attacks assigned in Step 1 are then – regardless of success or failure in Step 2 – deducted from the attack pool. If the character has additional attacks remaining, the controlling player may then repeat this process on the same or a new target.
   

Once the pool reaches 0, the character may make no further attacks until the following combat round.

Injuries

When a defender has accrued a number of Hits in one round (cumulative between all sources) equal to or greater than its own Man Equivalence, an Injury is suffered (or casualty inflicted, hit points lost, etc. - to taste).

Critical Blows

For an attack which succeeds and on which doubles are rolled on the 2d6 attack roll, a Critical Blow is struck. When a Critical Blow occurs, the hits inflicted on the defender are equal to double the attacks assigned to the roll.

Additionally, hits against vulnerable targets - targets unaware they are under attack, targets which are prone on the field of battle, and so on - are likewise doubled.

This is designed to prevent characters of only slightly higher level than an adversary from being invincible to that adversary in man to man while still preserving the possibility of great warriors or powerful monsters only a hero could hope to tackle alone.

The Math: Hits and Crits

A hit on the Man to Man table can be achieved - depending on weapon vs armor - on between a 5 and 12 on 2d6: producing a wide variety of probabilities. To prove I've thought about it - for the purposes of illustration, we'll focus on target numbers of 6, 8, and 10 - an easy, a normal, and a challenging result.

Provided, a matrix of all possible combinations on 2d6 - with their respective result:

2d6 1 2 3 4 5 6 6 7 8 9 10 11 12 5 6 7 8 9 10 11 4 5 6 7 8 9 10 3 4 5 6 7 8 9 2 3 4 5 6 7 8 1 2 3 4 5 6 7

In red, a result which will miss all target numbers. In yellow, a result which would hit only the easiest target number, 6. In blue, a result which would hit both the easiest 6 and the moderate target number of 8. In green, a roll which would hit all target numbers. In deeper highlight and italicized, a double: representing on hit a result which would indicate a critical result.

Observing the provided, there are 36 possible outcomes - with six possibilities for a double. Thus:

Target Number	Chance for a Hit	Chance for a Double
6	26-in-36, or around 72%.	4-in-36, or around 11%.
8	15-in-36, or around 42%.	3-in-36, or around 8%.
10	6-in-36, or just shy of 17%.	2-in-36, or just shy of 6%.

A character well armed - having a weapon appropriate to the enemy's defenses - may reasonably expect to hit frequently and hit above their pay-grade on a d20 equivalent of 19: whereas a character armed poorly relative to the enemy's defenses will have a difficult time hitting - and will hit above their pay-grade on a d20 equivalent to a natural 20.

Inspiration

The Hero can stand against a squad of normal men, holding his own; the Wizard stares down a fell beast which has laid waste to an army before it! Chainmail accomplishes this via the Fantasy Combat table - a solution to a problem where some units take many hits before going down: making them prime candidates to strategically lock down other units. Concurrent hits - as I describe above - are actually very uncommon: only Lycanthropes, Heroes, Super Heroes, and Wizards (implied) use them: however, this is an even more extreme example - as two Heroes who run into each other on the battlefield, using troop combat - have a statistically insignificant chance (almost 1 in 1,300) to end their conflict in anything other than a draw. Thus, the Fantasy Combat matrix is provided to pit Fantastic creatures and figures against one another - catering to the fiction from which they are drawn: Dragons cannot be killed in normal combat, but can be slain by a Hero with a bow and magic arrows (*cough* *Esgaroth* *cough*) with only a moderate degree of luck; Trolls are fearsome opponents, but are vulnerable to the magical weapons Elves carry (*cough* *The Broken Sword* *cough*). The key take-away here, however, is less the fiction: but from the gamist perspective, the rapidity of resolution. 

Chainmail is a wargame. The intent of Chainmail is not to imitate epic show-downs, but instead to allow armies to clash and to vie against one another for land, treasure, or honor. For that reason, it has to be quick - to bog the game down would result in the inability to utilize the rule in a tournament setting: and might inspire players to ban the Fantasy Supplement altogether - favoring the tight rules for more real-world military engagements. And it accomplishes this smoothness by pairing Fantastic adversaries against one another on a chart - requiring one roll only to decide the fray.

But how to accomplish the same result without ever and infinitely expanding a literature-based table?

An Equivalence Solution

I referenced Majestic Fantasy for Swords & Wizardry above. In reality, I should have referenced a blog post - Multiple Fighter attacks revisited for Swords & Wizardry: written by Rob Conley, proprietor of Bat in the Attic Games the author of the aforementioned Majestic Fantasy and a perennial figure in the OSR ecosystem. You're shilling him awfully hard  this post, CWR - what's up with that? I want to highlight Rob and his product line because - while, admittedly, I like some of it more than others - by and large he's a sharp cookie who has good insights into the experience of an OSR sandbox game. If you play Swords & Wizardry, or if you play OSE or any other TSR-friendly OSR system - I would encourage you to consider looking into his blog, checking out some of his work, and seeing if it's right for you. 

But - for the time being, we'll focus on what I saw that changed how I wanted to approach combat in Chainmail and Weapons, Wits, & Wizardry: namely, multiple attacks. In AD&D 1e, a Fighter gets multiple attacks - one per level - against enemies of 1 HD or less. This is an import from the Chainmail roots of the game, promoting the Name-Level-Fighter-As-A-Wrecking-Ball of mass combat. Encountering 40-400 Kobolds? No problem. You're taking out 10 at a time - plus, by that point, you're assumed to have retainers behind you: but that's another story. This didn't sit well with me - as I commented on Rob's post when I first read it - because a Fighter is a whirlwind of destruction against 1 HD enemies - skeletons, normal men, goblins - but then becomes no better than his own normal men troops immediately facing anything else - zombies, Elves, a freaking badger, etc. How Rob handles this - I recommend his blog post for a better explanation - but he counts down: the Fighter "spends" attacks according to the HD of the enemy - attacking, dealing damage, and so on as normal with each. It's a considerable boost - but one which is much needed for the character archetype that, though allegedly Gary's favorite, falls behind fellow classes in effectiveness as the levels accrue. Approaching it from a Chainmail perspective - I had the opposite problem: too many attacks and how to pare them down!

In the Spirit of Fantasy Combat

The resolution strategy of Fantasy Combat is solid - the problem comes with extensibility. In Chainmail, Fantasy Combat is a chart - depending on which figure engages which other figure, cross-referenced on the matrix. Knowing this, each new Fantastic monster would need to have a new entry on the list - and values determined for each potential opponent. To stick to literary inspiration - this would also pose a problem as some new monsters have no precedent to appear next to one another in the fiction: naturally providing a narrative limitation on the addition in addition to the margin limitation: that is, how big can the table get before it no longer fits in a book?

The Man to Man approach with a multiple-hits caveat solves this problem.

By treating the attack and defense of a Fantastic creature according to an equivalence on the Man to Man table - which can be extended indefinitely in rows, if the need arises - it provides a natural answer to "how would this monster affect this other monster?" and "how would a human heroic character overcome this monster?" Thus, say we create a Grue. The Grue - according to our own table - has 4 hit dice: a fighting man of normal bearing has no chance to hurt it. However - a Hero - armed accordingly - has a chance. A single success will strike the creature down - as would a Fantasy Combat entry - and also it allows the Hero to outfit himself for the occasion. We know that Grue attack by eating their foes: perhaps bulkier armor (Plate), preventing easy swallowing, will affect its ability to inflict that attack. Weapon and armor choice continues to matter - giving the Fighting Man more tools in the toolbox. But that's another story.

What about Monsters?

Ostensibly, monsters should receive multiple attacks. A cave troll crashing into the packed ranks of pikemen, rending left and right, throwing men asunder before coming face to face with a champion in its way - this is accomplished in Chainmail by troop combats followed by Fantasy Combat when the pair square off. In the same sense - combining the two under the Man to Man approach described above, the troll would - having a high Man Equivalence - be entitled to multiple attacks (the aforementioned left and right) and then few - perhaps one or two - against the hero, who likewise has an elevated Man Equivalence.

If it gets to the point where it's tedious? Say, rolling 2d6 three or four times in a row?

In that case - I may recommend defaulting to troop combat. Roll dice, on 6s - count a hit: appropriate to the light/heavy/armored spectrum - just in the name of speed. However: for the time being, the rule stands - and unless play-test proves a problem? We'll see how it goes!

Thanks for reading!

Public domain artwork retrieved from OldBookIllustrations.com and the National Gallery of Art and adapted for use. Attribution in alt text.

Bumping Thief Skills by Level

Another "fix" for Thieves!

Moreover, referencing my recent post regarding Thief Skills for Non-Thieves, it dawned on me - in the same call-in episode to the Red Dice Diaries that I had called into, so also had my friend Rob C. of the Down in a Heap podcast called in with his own house-rules for "fixing" the Thief skill progression. He seemed to have a similar line of thinking to other refs with whom I have played before - advancing the skill chances to succeed according to level, retaining the table but making it a little less punishing - and he did so in a simple form: so, if you're still looking to make Thieves more effective early on, here is what Rob C. does (or, at least, my take on it)!

Accelerated Thief Skill Success Chances

A Thief character, when tabulating their chances of success at a given Thief skill (optionally excluding Hear Noise, at the behest of the referee's sense of verisimilitude), is treated as a level of experience greater than themselves equal to the ability modifier of their prime requisite ability, Dexterity. Thus, for a 2nd level Thief - for example - attempting to Move Silently:

• At a Dexterity of 12, or +0, have a 25% chance of success.
• At a Dexterity of 17, or +2, have a 35% chance of success.
• At a Dexterity of 8, or -1, have a 20% chance of success.
  

A character cannot go below 1st level when determining the effectiveness of their skills.

The Specialist Thief (Optional)

In addition to the above, a Thief may choose to specialize in one of the six percentile-based Thief skills. In so doing, the Thief gains an additional increase of 2 effective levels: further increasing their chances of success in their chosen specialty. So, if the same 2nd level Thief from before were to specialize in Move Silently, their chances would change accordingly:

• At a Dexterity of 12, +0 from prime requisite and +2 from specialization, have a 35% chance of success.
• At a Dexterity of 17, +2 from prime requisite and +2 from specialization, have a 45% chance of success.
• At a dexterity of 8, -1 from prime requisite and +2 from specialization, have a 30% chance of success.
  

In fairness, I think Rob's rule is a bit less generous - in that he grants a +1 only at a 16 or above - but I'm a magnanimous DM, aren't you?

But this has me thinking...

While jotting this down, it made me think - how much does this differ, in terms of Thief effectiveness, from adding their Dexterity directly to the modifier? Specialization excluded - as that was not a part of the original houserule, a quick break-down using Hide in Shadows looks as follows; with "Bump by Modifier" referring to boosting level according to Dexterity modifier (this houserule) and "Bump by Score" referring to boosting the percentile directly (the previous houserule):

		Bump by:
With Dex Of:	R.A.W.	Modifier		Score		Delta
For a Thief of 1st Level
8 (-1)	10%	10	1	18%	3	-2
12 (+0)	10%	10	1	22%	3	-2
16 (+2)	10%	20	3	26%	4	-1
For a Thief of 3rd Level
8 (-1)	20%	15	2	28%	5	-3
12 (+0)	20%	20	3	32%	5	-2
16 (+2)	20%	30	5	36%	6^	-1
For a Thief of 6th Level
8 (-1)	35%*	30	5	43%	7	-2
12 (+0)	35%*	35%*	6	47%	7	-1
16 (+2)	35%*	55%	8	51%	8	+0
For a Thief of 9th Level
8 (-1)	65%	55%	8	73%	10	-2
12 (+0)	65%	65%	9	77%	10	-1
16 (+2)	65%	85%	11	81%	11	+0
* Technically, B/X has a 6th level Hide in Shadows at 36%,     but as far as I know, this is widely accepted as a typo. ^ On the nose, as far as typos are concerned!

Comparing the delta - bumping the Thief's effective level for the purposes of skill percentages, excluding the Specialization skill, is less generous than adding Dexterity to the percentage, as a whole. We knew this - or, could have inferred it easily - regarding the lower Dexterity scores: the former method applies a penalty for negative Dexterity, while the latter is always beneficial to the Thief. However, an interesting trend - the higher a character is in level and the higher the Dexterity score of the Thief is, the more closely the two align. 

At the lower bound, there is as much as a three-level difference between a dexterous Thief and a sinistrous one; however, looking at the higher level thief - this shrinks to 2 levels for the clumsy Thief, but is down to even for a Thief of respectable Dexterity - going one higher, if the Thief were to have rolled an 18 - the increase by bonus rule would actually outpace the flat bonus by score: offering a 9th level Thief the equivalence of 12 where the 82% chance granted by Dexterity only would still truthfully only be approximate to an 11th level character. Further - to compare percentiles alone, the bonus by modifier actually does outpace the bonus by score in terms of raw success chance: with both our 3rd level and 9th level characters benefiting more from a +2 Dexterity modifier than a 16 Dexterity.

So to conclude: this version of the Down in a Heap suggestion benefits higher level characters equivalently or better than does the CWR suggestion of two weeks ago - however, the CWR suggestion benefits lower level characters much more generously. 

What kind of experience do you want to create with the Thief? 

Once you've decided, pick your poison!

So Kudos!

Kudos, Rob - over at Down in a Heap: thank you for the rule and thank you for getting my brain running on the subject again! Kudos, John - of the Red Dice Diaries - well played in your Thief episode, getting a bunch of folks thinking on the subject with you! And kudos, readers, for bearing with me on yet another Thief fix post. Surely, with the sheer volume of houserules regarding Thieves, we definitely haven't run out of appetite for mechanical comparisons and new ideas yet.

Delve on, readers!

Public domain artwork retrieved from OldBookIllustrations.com and adapted for thematic use. Attribution in alt text.

Thieving for Non-Thieves

What do you do - as the ref - when a player not playing a Thief wants to do something that, typically, only a Thief is allowed to do?

This is a problem with skill systems - and an expose on how, the more you add to a game in terms of mechanical character complexity, the more you take away from the agency and imagination of the players. On a recent episode (or... well, an episode, at least) of John and Hannah Large's Red Dice Diaries podcast, John speculated on the subject - and I called in with my own solution. 

Or, at least, the solution I use: as always, I refuse to claim that I thought of it first!

But either way - before I forget about it, I wanted to write it down - to post it here - in case you, dear readers, were interested in that solution and in the impact it has, numerically, on the game as played. 

Maybe it'll work for you, too!

Attempting a Thief Skill as a Non Thief

If a character of class other than Thief wishes to attempt an action which should fall under the purview of a percentile-based Thief skill, they may do so - rolling 1d% under their Dexterity ability score. Thus, a Fighter with a Dexterity of 15 would have a 15% chance to open a lock - assuming said Fighter is carrying lock-picking tools or can otherwise improvise under the approval of the referee.

A non-Thief character attempting to Hear Noise has a static 1-in-6 chance of success.

Modified Thief Profile

A Thief character attempting to use a percentile-based skill other than Climb Sheer Surfaces, to determine their own chance of success, sums the value appropriate to level from the Thieves' Abilities table with the character's Dexterity ability score. Thus, a Thief with a Dexterity of 15 would have a 30% chance to open the lock referenced above - or, finding the lock to be trapped, a 25% chance to remove the trap.

If a particular skill - other than Pick Pockets - would be driven above 99% by summing the table value with the character's Dexterity, treat it as 99%: such that there is always at least a 1% chance of failure.

When attempting to Climb Sheer Surfaces, a referee may opt to apply the Dexterity modifier to the chance of success - but the chance of Climb Sheer Surfaces being so favorable to begin with, even for low-level thieves, this modification may be considered optional. 

Hear Noise (Optional)

When attempting to use the Hear Noise skill, a referee may opt to have the character apply their Wisdom modifier to the roll. So - for example - a Thief of 5th level with a Wisdom of 13 would have a 4-in-6 chance to succeed rather than 3-in-6; a Fighter of any level with the same Wisdom of 13 would have a 2-in-6 chance; or a Thief of 8th level with a Wisdom of 4 would also have only a 2-in-6 chance.

If a positive Wisdom modifier would take the chance of success above 5-in-6, instead allow a re-roll on a failed attempt, succeeding on the margin of success: ergo, a name-level Thief with a Wisdom of 18 would have a 5-in-6 chance to Hear Noise as a base, but on rolling a 6, would be allowed to roll again: still succeeding on 3-in-6.

If a negative Wisdom modifier would take the chance of success below 1-in-6, instead force a second roll on success with an inverted chance of failure: that inverted chance being 6 less the negative modifier. For example, a Fighter with a Wisdom of 5 would have a 1-in-6 chance to Hear Noise as a base, but on rolling that 1, would need to "confirm" that success on 4-in-6 due to the -2 Wisdom penalty.

But what effect does this have on the game?

Surprisingly little, actually. 

First, to speak to non-Thief characters the average Dexterity score is going to fall between 9 and 12 - so, comparing that level to a first level proper Thief, the Thief's niche is still protected.

So, assuming a Dex of 10:

	Open Locks	Remove Traps	Pick Pockets	Move Silently	Climb Sheer Surfaces	Hide In Shadows
Non-Thief Character	10%	10%	10%	10%	10%	10%
Thief, 1st Level (RAW)	15%	10%	20%	20%	87%	10%
Delta	+5%	+/-0%	+10%	+10%	+77%	+/-0%

Even RAW (that is, when the houserule is applied to non-Thieves, but the Thief is not boosted with their own Dexterity) - this gives a non-Thief a chance to accomplish something: but the Thief class is not superannuated. The Thief is still a good investment to have - or, even, multiple Thieves: in case you come across something that disallows retries (e.g. Open Locks).

But how about Thieves, themselves?

A Thief character is likely to have a higher Dexterity than the average character. This isn't because we're doing point buy or anything FOE like that - but if you're choosing to play a Thief, it's frequently because you rolled a high Dexterity anyway, which compliments your XP gain: plus, with B/X, at least, you can reduce your non-prime requisite abilities to boost your prime requisite - contributing to the difference.

Thus, assuming a Dexterity of, say, 13 (the minimum to achieve a +1 modifier and the minimum required to benefit from a Prime Requisite XP bonus): we find ourselves with the following, examining a cross-section at various levels:

	1st Level			3rd Level			9th Level
Thief Skill	RAW	Ruled	Lvl Eqv	RAW	Ruled	Lvl Eqv	RAW	Ruled	Lvl Eqv
Open Locks	15%	28%	4th	25%	38%	5th	75%	88%	10th
Remove Traps	10%	23%	4th	20%	33%	5th	70%	83%	10th
Pick Pockets	20%	33%	4th	30%	43%	6th	75%	88%	10th
Move Silently	20%	33%	4th	30%	43%	6th	75%	88%	10th
Climb Sheer Surfaces	87%	88%	2nd	89%	90%	4th	95%	96%	10th
Hide In Shadows	10%	23%	4th	20%	33%	6th	65%	78%	10th

Above, we have the RAW version of the Thief skill, the new success chance based on the house-rule, and the equivalent Thief level, RAW, which has a comparable chance. Note - the levels aren't perfect - for example, with Open Lock - a 4th level Thief, RAW, has 30%: not 28%, which is indicated as equivalent in the chart for 1st level - but it's the closest in the ballpark: and a fair approximation. Regardless - the implications are interesting. 

Unsurprisingly, Climb Sheer Surfaces represents, on average, a 1 experience level improvement. However, for each of the other skills, the benefit appears to diminish as the Thief gains experience. That is, early on, we see an improvement of success rate of 2 - if a particularly good Dexterity is rolled - potentially even 3 experience levels: then, in the mid-level range, the gap shrinks, reducing from a 2-3 level bonus to a 1-2 level bonus at best. Finally, as the character approaches name level - the improvement firmly sits in the 1 category: and, at a glance, by 11th level, the impact is effectively moot: a name level Thief with a 3 Dexterity or a 18 Dexterity won't have a noticeable difference in their chances to succeed at most (if not all) of their percentile skills.

Having not delved into the Hear Noise equivalencies - really, it's x-in-6: it should be an easy compare for anyone willing to eyeball it - I like this rule a lot. Less so as a mechanism to cope without a Thief and more so as a way to make the Thief more attractive. The rule helps out lower level characters, who need help the most, but then evens out at higher levels: slowly but surely making sure that the character's experience level - not the randomly generated number they were assigned by fate at char-gen - is paramount to the determination of their success: something that Gygax was very concerned about, regarding the relevance of character abilities compared to class abilities in their effectiveness in game.

So - looking for a way to boost low-level Thieves, but not wanting to break your game or change the style of play or impact the thematic experience? This rule might be for you!

Delve on, readers - and let me know how it rolls!

Public domain artwork retrieved from OldBookIllustrations.com and adapted for thematic use. Attribution in alt text.