The Simple SECRET to Better RANDOM Encounter Tables

Random encounters that don't quite feel right are a common issue for GMs. To solve this issue, we must go to the very source of randomness: MATH! ▶️ More below! ⏬

Best ways to support

RPG resources (affiliate links)
*use discount code "BOB" !

Thank you for all your support, and keep building :D

00:00 doctors hate this one weird trick!!1
00:52 the main problem of most random tables
03:18 another GM prep tool! (sponsor)
03:54 how the math always wins
04:53 random encounters, balanced by nature
07:09 why two dice are better than one
08:21 the right way to make a random table!
10:06 the simple math GMs must know
13:01 weird math & the final step!

#dnd #ttrpg #dungeonsanddragons

[tomassunaert]

I love how its technically possible to have to fight every single encounter at once

[rdigeri]

An alternative way to do it is to roll one die, but assign a weight to each result, such as a result of 1 to 6 on the die always resulting in kobolds, but only resulting in dragon on a 20.

The strength of this variation is that you can build the pacing into it, such as this:
You don't want the dragon encounter to happen early into the dungeon? Have a d20 table but assign the rarer creatures to values above 20. Then, as each encounter passes, add an increasing bonus to your die roll, so these rare encounters become possible, or even probable as the game goes on.

[kaytch]

The game Cyberpunk Red uses 2d6 tables for this reason, and it works really well. A very dramatic example is the injury table (if critical damage is rolled), where it takes a roll of 2 or 12 to lose an arm or leg (versus a foreign object or torn muscle is in the middle). Naturally one of the first injuries my character took was a lost arm, and since it happened so early it became a big part of her character. But it only happened once to a PC in the whole campaign, so it always felt like a big deal that it had happened.

[Kosmonomicon]

Great vid! A favorite dice mechanic of mine was for a homebrew where rather than using a d20, we rolled checks using d8+d12. The bell curve reigned in extreme rolls and felt less swingy. As an extra feature, the d12 was used to generate which body part you hit with an attack, and the d8 was used for damage. That way, a high attack result was almost guaranteed to also be a headshot for big damage. Very nifty go get 3 different results with 1 roll!

[nikolaosalexandroskaloumen]

5:09 “ I’m sure you’re familiar with the idea of food chain:

Plant grows > deer eats plant > human eats deer > dragon eats human.

Simple. “

😂

[Bugseid]

I like using hexflowers for random encounters! Or random weather, terrain, treasure, dungeon rooms...
They're cool because the last result affects the next one. Also makes it feel like a bit of a board game, *and* makes it harder for the players to learn what's actually on the table.
You'll start in one hex, then roll dice to determine which direction and how many hexes you move. Tada, now you have the new result! On regular tables, rolling the same exact result will always get you the same thing, but on a hex flower it can end up being different.

[MinecraftLovesSteve]

I actually have used a special encounter table in the past that i call "weighted random" where you make a 1d100 table and put the results you want across various ranges. For example, kobolds would be a 45 number range where as dragon would be a 5 number range and so on. This led to a set of 100 where more likely results took up more real estate, thus making them more likely to appear

[nettlesandsnakes]

7:57 Considering a smaller table is often better while still preserving this Bell curve kind of thing, I would suggest 2D4.

[CaseyWilkesmusic]

Hahaha, “that’s not a secret it’s page one of the original dm guide” was my first thought. Props to you Bob!!❤

[davidjennings]

Depending on the game I sometimes cross out an option on my roll table and roll again if I hit that (could replace it with roll two so things escalate as players kill more) this means that your rarer encounters become more likely. I.e. your dragon is more likely to come out the more of its minions you kill.

[JKevinCarrier]

On a random note, I really enjoyed your use of artwork from the 1st edition Monster Manual. I always say that D&D lost a little of its personality when the art started being so slick and realistic. That mimic with the fist coming out of its side is iconic!

[pops]

5:29 dragon eats man… 😂 This video can also be called “How Does Gambling Help Build a Better Random Encounter Table”. Great job introducing statistics to the world building community! This is why Vegas makes 7 the Lucky Winning number on the Craps table… you know the movies where they throw the 2d6 but there’s NO miniatures on the table;) They are applying this bell curve to gambling. This is kind of what happens when you “Roll w/ Advantage”; even though you don’t get to add them up for a better average it creates the possibility of hitting the MEDIAN roll of 10! Love this video!

[maitrecorbeau_gm]

As you told, this is a great way to push the story you want. And it reminds me of old school RPG on video games when you had to look for/avoid random encounters as you never knew when the big bad creature was about to ambush you !

[christopherscholl]

You can also bake the bell curve into a classic d100 table, and you still get the joy of rolling 2 dice. You can fine-tune the odds along a bell curve.

[johnathanrhoades]

The other really cool idea for random tables is weight it with less dangerous stuff at low numbers and more dangerous stuff at higher numbers, and then roll 1d6 + X where X is the “danger value” of the encounter. I think Dungeon Masterpiece did a video on that.

Also, making sure you have good thematic tables (including non-combat encounters) and an “activity in progress when encountered” table keeps things interesting!

[remixtheidiot]

if you make the bad encounters at the higher numbers and not that bad at the lower numbers, you can do what d&d does during character creation.

roll an extra advantage die or two. If monsters are on high alert, roll 3d6 and keep the 2 highest results per random encounter, or keep the lowest if the players are being particularly sneaky and good at traversing or something.

just a fun thought.

[DarthCasus]

I liked the pokemon reference and the heavy metal. Keep it up Bob, your content is great!

I would also suggest turning every random encounter table into 1d100 and assign percentage values to all your options. You won't be trapped into using every face of the die for an option, where if you only had three things you wanted your players to encounter, 1-15 is your first encounter, 16-50 is your ssecond encounter, and 51-100 is your third. It takes one extra step in math to set up, but this way you aren't dealing with bell curves and all sorts of extra encounters you really weren't interested in running.

[ronwingrove]

The 2d4+1d10 roll looks like it could be useful for weather. If it's the middle of summer, the chances of a blizzard should be fairly low, while variations on "warm and sunny" are basically a given.

[Lioneldehetre]

Back when I was a kid, basic and 1e days, it encounters didn’t matter as much and I used random tables all the time. When I returned to the hobby I found that I used them less and less favoring to preparing certain events/effects that may occur in a given session. Thanks for the great video!

[DyrianLightbringer]

I've been using fewer random tables over the years and more planned encounters. Partially because I was disappointed that certain encounters didn't show up during the game because they weren't rolled, but mostly because random encounters just started slowing the game down. It started to feel bad when the end of the night seemed to always line up with a part of the adventure where the party was traveling and we were just rolling for random encounters for like the last hour or two of the game session.
When I was using random encounters, I used to use the 2d6 method, typically with one result of "no encounter" so I'd have the random chance of having an encounter baked in with the result of which encounter we have, and if we were doing overland travel, I split my table into a day and night column. However, I started experimenting with a 1d12 method. It worked like this: I rolled Xd12, where X is the number of possible encounters I wanted to happen during the given time period, based on how dangerous the area was. Each player rolled 1d12. If any player roll matched one of my rolls, that encounter happened. That way, a bigger party naturally led to more chances for an encounter. If multiple dice matched, I would either run the encounter twice or would double the encounter, like there were twice as many monsters. Yeah, it resulted in some wild encounters sometimes, but it didn't happen often, and these encounters were usually one per day, so the party was typically at full health and spells, so each encounter could be more dangerous than normal without a TPK. Also, I sprinkled a number of beneficial encounters into the list as well, so the party had a chance of encountering friendly NPCs, or the remains of a previous battle, and could get some loot.