Telling Players What They Need to Hit

During a battle in last night’s White Sandbox game that involved seven PCs, ten werebear allies, and another ten men-at-arms, I often made announcements like: “OK, everybody who’s firing missiles, go ahead and roll your attacks. A 12 will hit the harpy, or a 10 for a hero.” In doing so, I fly in the face of several people I respect greatly:

Don’t ask me what you need to hit. Just roll the die and I will let you know!

– Dave Arneson

and howandwhy99, in response to a thread at the OD&D boards called “Do you tell players what they need to hit?“:

Never. Or for any roll. It defeats the entire design of the game, if a referee does this. IMO it is the purpose of the game for the players to figure out what works and how through play. At least it is in mine. I’ve pointed out before I view D&D and some other early RPGs as interactive pattern finding games. This may be scoffed at, but it’s what I see a lot of folks doing, if even only part way. This doesn’t mean all groups must play according to this design and/or in this manner, but in my experience it delivers the most enjoyment for exploration of the unknown – at least the unknown outside one’s own desires.

I think this idea of “RPGs as interactive pattern finding games” is valuable and interesting, and have been working on understanding it through conversations with its proponent both online & at GaryCon. Some analogies he’s suggested that I find useful:

  • In the game Mastermind, one player sets up a hidden pattern of colored pegs. Gameplay consists of the other player trying to guess the arrangement of pegs and receiving feedback from its creator; the game ends when the pegs placed by the guesser match the ones hidden behind the screen. (Paul Jaquays chose to write about a similar game, Black Box, as his entry in The 100 Best Family Games).
  • In a classroom, one person sets up a syllabus of knowledge they possess that others do not. Teaching consists of the other people trying to demonstrate their understanding of the material and receiving feedback from the teacher; at the end of the course, the students’ success is gauged by how well their understanding of what the syllabus means matches its creator’s.
  • In Dungeons & Dragons, one person sets up a map with a key showing the locations of monsters and treasure. Gameplay consists of the other players guessing which areas shown on the map to explore and getting feedback about where rewards can be found and dangers avoided. At the end of an adventure, the player’s success can be gauged by how well their map of the area matches the one behind the screen.

Although as with any analogy, there are points of difference between each of these situations, I am swayed by the argument that figuring out what the referee knows that you don’t is a key aspect of D&D play. So why don’t I make players roll to hit?

To answer that let’s look at another kind of pattern-guessing game, which I was introduced to on a high school camping trip. All the Outdoor Club newbies are told that there’s a story that old members all know. As a rite of initiation, you’re going to learn what this story is, just by asking yes or no questions. So you go: “Is the story about Outdoor Club?” and everyone replies “Yes.” And then you go “Does the story take place here?” and everyone goes “No.”

The remarkable thing about playing the game is that everyone’s responses are unanimous; whatever question you come up with, it takes hardly any time at all for the group to decide whether the answer is “yes” or “no”. This goes a long way towards making you believe that there really is a time-hallowed story whose contours are well known to everyone who’s been on an Outdoors Club story before. But soon discrepancies creep in. If Thea asks “Does the story include Tavis, Sunyoung, and I?” the answer is “no”, but if I ask “Does the story include me, Thea, and Sunyoung?” the answer is “yes”. Puzzling out these discrepancies eventually lets you figure out the pattern whose logic creates the story, which is that (spoiler alert – select the following text with your mouse to see the answer) if the last letter of the question is a vowel, the answer is no; if the question ends in a consonant, the answer is yes.

So if D&D is a pattern finding game, would gameplay be enhanced if I added this layer of pattern to be discovered? If we take this rhetorical question seriously, the answer varies from “obviously not” – it’d break the logic of the rest of the game if “I rolled a 12 against the harpy, does that hit?” and “does a 12 hit the harpy?” got different answers-  to “maybe as a one-time thing.”

For example, solving this kind of puzzle might make a good trick room in a dungeon (with excellent old-school credentials; when I played with Frank Mentzer at the first Tower of Gygax, we encountered some NPCs at a tavern who needed help figuring out how to split their rations and gold pieces to make it come out even, and also an insectoid monster who couldn’t be hit by normal weapons; gameplay in both situations involved making guesses and taking damage if they were wrong.)

The reason I don’t usually stock my dungeons with doors that require this kind of puzzle to get past, and that I don’t usually make players figure out what they need to hit, is that doing so uses up time that I’d rather spend on other kinds of pattern recognition. Last night I called out what you needed to hit the monsters because saving a few seconds on giving feedback about that would speed up our mutual exploration of the patterns we care more about, like “will the Grey Company defeat the Beast Lord?” and “what new aspects of the dungeon’s hidden pattern of forces will emerge as a result of his death?”

I don’t think that giving away to-hit numbers greatly weakens the tactical decision-making aspect of combat. I usually don’t tell what you need to hit on the first round, so there’s still uncertainty about initial target selection, and I’ve always given lots of narrative feedback to reduce that uncertainty: first “This slime is covered in rocks, this one is clinging to bones in plate-mail armor” and later “One arrow finds a chink between the rocks, the second one pings off the armor”. I’ve got nothing against the mini-game of figuring out what you need to hit by tracking which numbers do or don’t, and mechanics (like Power Attack in 3.x D&D) that reward players who solve this puzzle by letting them optimize their tactical choices. It’s just not where I want the pattern-seeking emphasis to fall in this particular campaign.

I can see the argument that asking for target numbers breaks immersion and reminds players that it’s just a game. But carrying this logic further, I could increase immersion by rolling all the dice for PCs and NPCs alike and just narrating the results – or, to keep the bit of pattern-finding data about whether or not a 12 hits, by having the player roll the dice where I can see it, but not allowing anyone to announce the number out loud.

The thing is, I think rolling the dice and seeing what your fate will be is pure awesome. And part of why rolling a natural 20 is so much fun is that the feedback is immediate; your fate is sure to rock. When you roll a 12 and have to ask me whether that hits, the gratification is delayed, but if I’ve told you ahead of time that a 12 will succeed, there’s no lag between rolling the dice and getting immediate confirmation of your success.

All the above is a lengthy explanation of things of which I’m already convinced. Or, at least, areas where I already know where I want to draw the line. For example, I’m OK with another delayed-gratification step in combat, in which the player rolls for damage and has to wait for me to tell them whether that kills their target. I could remove that by telling everyone the HP of each creature in the battle, and letting players tally their decline so that they’d know right away when their blow drops a foe. In an OD&D campaign I don’t see the time that would save as being worth the decrease in immersion and uncertainty, but in a 4E convention game I might decide differently.

Here are some things I haven’t decided about:

In last night’s game, Arnold used Zolobachai’s Impertinent Invitation and a good reaction roll to gain an audience with the Beast Mistress (or at least her successor, the previous holder of the title having been killed in the previous session by “drunken adventurers” with no relation to the current party). Once contact had been established, he attempted to poison her with a waterskin full of the Water of Life, whose generative properties he knew had caused all manner of foulness to spawn from the attempt to heal the dead leather of its lining. I rolled her save vs. poison, which was a success. Then I decided to roll a new reaction roll to see how she’d respond.

This was an exciting moment for me; separated from the rest of the party, surrounded by minotrice in their lair, Arnold’s fate hung on the roll of the dice. They came up 9, “favorable”, so she assumed he was an idiot who didn’t know that the Water of Life can only be transported in a living vessel. But since I kept that roll to myself, the players didn’t share in the tension, and had no way of knowing how close Arnold had come to being killed and eaten.

When the Beast Lord was in combat with Merselon the Magnificent, before I rolled for the blow that killed that brave adventurer, I announced “The Beast Lord needs an 8 or better to hit.” I did this because it was a key moment, and to enhance the feeling of hanging on the roll of the dice, I wanted everyone to be able to know right away whether Merselon had or hadn’t dodged that lethal axe.

Should I have similarly announced my thought process beforehand with the wineskin: “The Beast Mistress will correctly intuit that Arnold was trying to kill her unless I roll a 9 or better on 2d6?”  This goes against my previous thinking about reaction rolls, which kept some of the outcome secret from players, but I feel like that scene might have been more satisfying for the players if, knowing what was at stake, they’d had a glimpse of the odds and a ringside seat when the dice delivered their judgement.

Another way to not keep patterns secret from the players that I’ve been considering recently comes from a post at Delta’s D&D Hotspot:

Showing wilderness encounter charts. In a very concise format, players get a concentrated, playable dose of campaign setting information. They can use this information to strategize about the exact advantages and disadvantages to different travel routes and adventuring locations, connecting gameplay to campaign knowledge in a deep way. It simulates well the lifelong experience and rumor-mongering that their PCs would have. Anticipation is raised to an intense degree when encounter checks are rolled.

I want there to be a place in my game for the ideas that inform howandwhy99‘s decisions not to tell players what they need to hit – that information is an essential resource in old-school play, and there can be great satisfaction in wresting knowledge about how things work from the DM’s cold, dead hands. Although in many ways I admire the way that 4E makes the rules of the game transparent and equally curated by everyone at the table, it struck a chord with me that, when George Strayton was talking about his 4E Temple of the Frog game at the Arneson Gameday, he said that it took a dose of the Master’s “don’t ask me what the rules say, just imagine the situation and let me tell you how it responds to your actions” to get his players into the mode of exploring an unknown and often deadly pattern instead of simply stepping through a series of comfortably level-appropriate and rules-bound encounters.

At the same time, I also want there to be a place for just handing players concentrated, playable doses of information in order to raise anticipation to an intense degree. These motives pull in different directions, but I don’t believe they’re ultimately incompatible because giving away some parts of the puzzle of how things work can free up time to devote to solving other parts. Speeding up combat gets us quicker to the question of what Tavis and Paul Jaquays know, but the players don’t, concerning the consequences of touching the solid gold throne of Zeus…

UPDATE: Just saw this post by Cyclopeatron in which he lays out arguments for and against telling players what to hit based on that OD&D board thread I cited above, which he started!

11 Responses to “Telling Players What They Need to Hit”

  1. May 23, 2010 at 6:43 pm

    This, and several of your other essays, should be required reading for people about to embark on an old-school DMing exercize.

    Incidentally, back in the day we played a variation of your Outdoors Club game: “My grandmother loves coffee but she hates tea.” The pattern, of course was that she hated anything with the letter “T” in it.

  2. May 23, 2010 at 6:55 pm

    Good argument for judgment over policy.

    For me it would depend on the players. If any of them were serious power gamer min-maxer types I’d assume they would abuse knowledge of the to hit roll to overanalyze the emergent properties of the game. (I’m going to give a few examples a couple of posts from now in my own blog but let’s just say, if you give players options to adjust their to hit and defense there are going to be some wonky ways in which you can squeeze out a 5% advantage here and there.) Otherwise, it is fun to know the roll you have to make. Making the hard rolls then becomes that much more awesome.

  3. May 23, 2010 at 7:05 pm

    “Speeding up combat gets us quicker to the question of what Tavis and Paul Jaquays know, but the players don’t, concerning the consequences of touching the solid gold throne of Zeus…”

    Oh, I think we can make an educated guess. Though if John Fighter really is the Lost King of Thracia, maybe he’s allowed to sit there and reclaim his kingdom if he can undo whatever sins caused the Thracians to abandon and seal off the caverns…

  4. May 23, 2010 at 9:16 pm

    Tavis has said as much, but it bears repeating. Even though I’m with you on the discovery part of D&D, decoding to-hit numbers is not as interesting as the tension that comes from knowing what’s riding on any given roll. It’s the same reason why craps is such an addictive, social game–everyone at the table knows what’s at stake, and instantly knows whether a given roll of the dice was a win or a loss.

  5. 5 ragnorakk
    May 23, 2010 at 9:51 pm

    Man… lovely post!

  6. 6 Invincible Overlord
    May 24, 2010 at 12:09 am

    Excellent analysis as usual, Tavis. FWIW, I’m absolutely on the side of as much mechanical transparency as possible. Arneson did and said so many smart things, but I wish people would let that one go — it was not his best day and he just sounds like a cranky overworked DM (like we all do from time to time). Shared information processing makes play better for two chief reasons:
    a) it’s faster.
    b) it’s more engaging, because less passive for the player.

  7. May 24, 2010 at 12:45 pm

    IO, I just posted a comment in reply to A Paladin in Citadel’s [blog post] on this issue in which I try to lay out the reasons Arneson’s stance made sense in his situation. I think that the arguments against transparency that apply to us are:

    – wanting our actions to be dictated primarily by what we think makes sense in the situation, and avoiding the tendency you see in 3.x/4E D&D to do things because you know it’ll be rewarded by the rules whether it makes sense in the situation (e.g. tripping an ooze)

    – wanting to spare the players some dealing-with-complexity; in this regard I’m thinking that I might track the hit point boxes for some or all of the PCs, since that regularly proves befuddling for new folks

  8. 8 1d30
    May 24, 2010 at 8:12 pm

    I generally ask the players to tell me their hit numbers until it sounds like they figured out what they need to hit it. After that there’s no surprise!

    But I often have modifiers for the players’ rolls that they don’t know about – how big the modifier is or even that there is one! For example, one PC may have picked up a magic sword, and another may be secretly cursed. I have one PC who lost a level due to energy drain, but I described it as a “permanent chill in the wound” that he couldn’t heal (to describe the lost HP maximum) and I manually deduct the attack roll penalty for his lost level.

  9. May 25, 2010 at 3:28 am

    Thanks for another beautiful and articulate post!

    After some experimentation I’ve pretty much converged on the same play style you have. The main reasons I reveal the to-hit roll needed (most of the time at least) is: (1) it’s faster and easier, (2) the psychological tension is ALWAYS more fun, (3) it makes sense that after a round of melee a PC should know how hard it will be to hit and damage an opponent, even if they made a crap roll on the first try.

  10. May 25, 2010 at 7:55 pm

    Terrific Post!

    One argument for keeping the reaction roll of the Beast Mistress secret from the players is that it allows for ongoing dramatic tension:

    The Beast Mistress takes a sip from the proffered waterskin, and her eyes widen in surprise and revulsion! Her head whips to one side as she reflexively spews out the befouled liquid. After a short fit of coughing and spitting, she wipes her mouth with a napkin and turns her attention towards her guest. For several seconds she gazes at Arnold, her blank face as inscrutable as stone. “Thank you for the drink,” her words break the silence, but her expression remains unchanged, “but I think the water may be contaminated. Perhaps you might find a different container?”

    [at this point, Arnold might try to profusely apologize for her discomfort]

    “Yes, yes,” the Mistress broke in curtly, “we all make mistakes. It is water under the bridge. Or in this case, under the table. Please continue what you were saying before”

    From this exchange, the PCs are left wondering whether the Beast Mistress didn’t catch on to the poison attempt, or whether she did know and elected not to exact her revenge at that time. From then on, all dealings with this character will be shot through with an undercurrent of uncertainty, which wouldn’t exist if Arnold knew “he” rolled well enough for his assassination attempt to escape notice.

    While I do think that secret reaction rolls can be fun, the DM should give the players context clues to get at least an idea of how they rolled. For example, if you had rolled a 12, then maybe the Beast Mistress thinks that Arnold is person who is genuinely friendly towards her, but has been used as a pawn in an attempt on her life. She might then take him into her confidence in an effort to help her root out the “true” conspirators. This would give him a position of trust from which he could make further attempts, or exert influence in some other way. He wouldn’t technically know what the roll had been, but he’d be able to deduce that it was pretty good.

    If you go this route, you might want to announce that you are going to make a roll, do so, and then proceed with your narration. Don’t say what number you rolled, and you don’t even have to say that you’re rolling for reaction. Just inform your players that something is up, and let them interpret the scene as they will.

  11. 11 howandwhy99
    June 4, 2010 at 6:54 am

    Hey, I just saw your post. Thanks for the nice comments. It’s an awesome post and I can’t help by share. Your comments are quoted for specific response.

    “But carrying this logic further, I could increase immersion by rolling all the dice for PCs and NPCs alike and just narrating the results”

    –The results of the dice let the players know how good of an attempt they made. The determination that rolling higher in combat will result in a hit is one usually made very early on. Knowing you rolled a 17 and didn’t hit means the player has likely deduced this is a very difficult opponent to harm through this method. If other PCs are in sensory ability proximity to this attempt, they also learn the results of the die roll said aloud. (This is in contrast to taking players aside when out of range to cut down on metagaming.)
    –If this was the first attempt, then the players can make a series of deductions. It’s time to run. This PC is weak in this area anyways. Maybe we should try a different attempt to achieve similar results?
    –Rolling all the dice behind the screen is a common method. The drawback I see is the inability to express the numerical relationship the die results. 10’s are “average hits”, 11’s are “slightly better than average”, etc. And this example holds only for attacks. Imagine needing terminology for every result for every type of roll. It’s too much of a bother.

    “This was an exciting moment for me; separated from the rest of the party, surrounded by minotrice in their lair, Arnold’s fate hung on the roll of the dice. They came up 9, “favorable”, so she assumed he was an idiot who didn’t know that the Water of Life can only be transported in a living vessel. But since I kept that roll to myself, the players didn’t share in the tension, and had no way of knowing how close Arnold had come to being killed and eaten.”

    –Reaction rolls are better when rolled in front of players IMO (or just those with PCs in proximity if the player chooses). This open rolling let’s the players know something they said was an action covered by the rules. One where the results are randomly determined.
    I generally use the term Attempts when referring to distributive patterns, but every “I do this” question receives an affirmative or negative response based upon the hidden code or irrelevancy thereof. So they are all Attempts in truth.

    “These motives pull in different directions, but I don’t believe they’re ultimately incompatible because giving away some parts of the puzzle of how things work can free up time to devote to solving other parts. Speeding up combat gets us quicker to the question of what Tavis and Paul Jaquays know, but the players don’t, concerning the consequences of touching the solid gold throne of Zeus…

    –This is up to you when you generate the code before the campaign begins. If you have a lot of rules for combat and few for metals, shapes, and deities, then more time will be spent on combat. 4E has a lot of combat rules which make it next to impossible to use as part of a pattern finding game. If I were running such, I would need to follow up on each player’s attempted action, having them further parse it until their specifics fell under the rules or did not.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Past Adventures of the Mule

May 2010
« Apr   Jun »

RPG Bloggers Network

RPG Bloggers Network

Enter your email address to subscribe to this blog & get email notification of updates.

Join 1,045 other followers

%d bloggers like this: