Archivist AI data shows good-aligned heroes and support classes on the rise in tabletop RPGs

Archivist AI says analysis of thousands of real tabletop campaigns shows evil-aligned characters are rare, while Paladins and Bards are gaining ground in Dungeons & Dragons. The findings suggest players are leaning more toward cooperative, story-driven characters than the chaotic antihero stereotype.

Why it matters: - Archivist AI’s dataset offers a real-world look at what tabletop roleplaying groups are actually playing, not just what online stereotypes suggest. - The findings point to a shift in player behavior as the newest Dungeons & Dragons rules reshape class preferences and character archetypes. - For game makers and players, the data suggests tabletop culture may be moving toward more cooperative, good-aligned, story-first play.

What happened: - Archivist AI released new analysis based on real, in-progress tabletop campaigns drawn from actual player characters. - The company says the dataset shows evil-aligned characters are nearly extinct, while Paladins and Bards are climbing in popularity. - The release also includes new analysis of GM speaking patterns across 54.7 million words of session audio and transcripts. - Archivist AI is the campaign memory platform for tabletop roleplaying game groups. - The company launched the analysis on Aug. 20, 2026, in San Luis Obispo, California.

The details: - Across analyzed D&D 5e and D&D 2024 characters, about 2% have Evil alignment. - Nearly two-thirds of characters skew Good. - Under classic 5e rules, Paladins made up roughly 5.5% of characters. - Under D&D 2024 rules, Paladins rose to about 7%. - That rise pushed Paladins ahead of both Fighter and Wizard into the top three classes. - Bard also gained ground under the newer rules. - Archivist says the older “lone adventurer” archetype has lost ground. - Greg Wood, co-founder of Archivist, said the data suggests the old image of a small group of chaos-loving dungeon crawlers is outdated. - The company says it does not collect demographic data, so the cause of the shift remains open. - Archivist points to several possible explanations, including found-family storytelling, actual-play shows, a broader move toward less grimdark entertainment, and the 2024 rules making Paladins and Bards easier to build and play. - Archivist says its role is to surface patterns, not decide which explanation is correct. - GMs speak an average of 46.5% of a table’s words. - The standard deviation is 12 points, which means most tables cluster around that midpoint. - Eight in 10 tables fall between 31.3% and 61.6% GM share. - Only 1.6% of sessions have a GM speaking less than 20% of the table’s words. - Only 3.7% of sessions have a GM speaking more than 70% of the table’s words. - The pattern holds across systems, party sizes, and play styles. - Archivist says the average voice session includes about 21,000 words. - Additional findings on class, species, and cast composition are available on the Archivist AI website. - Archivist AI says it supports more than 250 RPG systems and has processed more than 20,000 sessions. - The company integrates with Discord, Foundry VTT, Obsidian, and other tools through its MCP platform. - Archivist AI says it has publicly committed to never building an AI Game Master. - The company was founded in 2025 by Greg Wood and Greg Harzelak. - More information is available at the company’s website. - Archivist AI also shared social links for LinkedIn, Instagram, Facebook, YouTube, TikTok, and Discord.

Between the lines: - The data undercuts the common pop-culture image of tabletop games as mostly disorderly or morally gray. - The class shift may reflect both narrative tastes and a ruleset that makes support-oriented characters more appealing. - The GM-speaking-time results suggest tabletop conversations are more balanced than many players might assume, but GMs still dominate a meaningful share of table discussion.

What's next: - Archivist AI says this is the first release in an ongoing series of trend reports from its campaign dataset. - The company plans to publish additional findings on class, species, and cast composition. - Groups recording their own games can compare their own speaking patterns against the GM-share curve within a single play session.

The bottom line: - Archivist AI’s first trend report suggests tabletop RPG players are more good-aligned, more support-minded, and less chaos-driven than the stereotype implies.