Looking at Hasbro's 2024 results, I think this might be the first time that D&D tabletop RPG revenue has decreased with the release of a new edition. How'd that happen? It lines up with a lot of what I've seen in the marketplace: (1/5) Fans just aren't excited. The OGL took the air out of the community around the game, and other than shaky claims of backwards compatibility the game has almost gone out of its way to avoid answering a simple question: Why buy this new edition? (2/5)

OK, so let me breakdown the root problem of CR. It was terrible in 2014 and it remains terrible in 2025. It ignores how we actually play the game and is a millstone around the 5e system. It also explains why low level 5e works.

But why is that? Check the image below. That table shows the CR of a single monster that is a high difficulty encounter for a party of four characters of the listed level. So, a 7th level party can face one CR 10 creature. That creature has, on average, 165 hit points and AC 18.

So what's that final column?

I was literally dragged into a call with one of the directors way back in 2019. They asked if a druid wildshaping into an owlbear was OK, and I said yes. The rules were supposed to go into Tasha's along with original concept for variant class features, but that didn't happen.

https://twitter.com/newbiedm/status/1883974615946649969

Tasha's alternate classes features were supposed to be new ways to play each class. For instance, the barbarian might get an alternate rage that augmented speed, allowed the use of Dex weapons, and had some other benefits. It was supposed to support the next 5 years of the game.

I had a few people ask if I actually fit satisfying sessions of D&D into an hour. I 100% and have been doing so for almost five years now. The key is rebalancing monsters for faster but still dangerous combat, and streamlining the core rules for everything else.

A thread: Working from home, I found that I cannot spend more than an hour in a single Zoom call. My brain just fries. So I started running RPGs at lunch. We started with Boot Hill, a game ran to celebrate the life of Brian Blume. He passed away early on in the pandemic.

Used the troglodyte in my Wednesday game. This session nicely wraps up what I want from one hour of D&D: two significant battles, exploration/puzzle, roleplay scene, and a chase. We also spent about 10 minutes catching up, and ended 5 minutes early. First bit of action was a quick chase using rules inspired by the classic James Bond 007 RPG. Party ran down a bullywug scout. Captured him, interrogated him, learned a bit about this dungeon level. RP all ran in-character, including me blubbering like a frogman.

There's a weird bit of technical debt embedded in D&D - the time and distance scales don't quite sync up. It's been in the game since the beginning and shows the game's roots in Chainmail.

This also ties to why 5e characters feel like superheroes. TL;DR - The modern idea of the "encounter" as one keyed room in a 10 or 5 foot per square map is wrong. The encounter is that entire map. Exploration takes place between those maps. Dungeons should be built as keyed nodes connected by passages/stairs/etc.

5e characters feel like superheroes because unless the entire dungeon has the chance to engage them, it's hard to build single shot fights that can threaten them. The sweet spot between easy fight and TPK is perilously small. Instead, you want threats that ramp up as the fight continues. You beat 5e parties through attrition. They can always out alpha strike you.

Bonus actions are hot garbage that completely fail to fulfill their intended goal. It's OK for me to say this because I was the one that came up with them. I'm not slamming any other designer! At the time, we needed a mechanic to ensure that players could not combine options from multiple classes while multiclassing. We didn't want paladin/monks flurrying and then using smite evil.

Wait, terrible example, because smite inexplicably didn't use bonus actions.

It’s a big topic that I will now try to unpack in a series of Tweets. t.co/1TRQh19OTk 3.5 and 4 were very much driven by an anxiety about controlling the experience of the game, leaving as little as possible to chance. They aimed for consistency of play from campaign to campaign, and table to table.

The only question in game design that really matters is, “Why does a player care about this?” Yet, it’s something I rarely see designers bring up. Most RPGs never answer it. D&D 5 aims to make you care about your character as something that is your personal creation, and the ties between your creation and everyone else’s (DM or player).

My earlier tweet about language and spoken bandwidth is prompted by some thinking I have been doing about dungeon design. I'm using this image as an example. Let's assume the party is standing at point M with a torch. As a DM, you have to describe both the room and the hallways depicted. That includes the location of four doors and three corners. That's a fair amount of info to juggle. How much of that can the players actually process?

I'm developing a new style of DMing for myself, riffing off my earlier tweets about railroading. It boils down to this - in a game of D&D, the DM provides the foundation of energy and action for the session, but NOT the direction. I'll explain using combat as an example. I've been putting a lot of work into my combat management and presentation over the last few months, especially as streaming is something I do more of and, honestly, really love doing. It has made DMing into a true skill you can watch grow. I love that.

Railroading has gotten such a bad rap in RPG circles that we forget it is one end on a continuum. The opposite end is aimless drifting, with a DM who sits back and throws no hooks, injects little or no action. I’ve played in these games. They are THE WORST. Too much choice is an empty wasteland. No choice is just the DM talking while the players listen. My experience is that most DMs tend toward railroading when they start, then back away when/if they gain exposure to other styles.