by Dave Kot

Charlie, Eric, and Shawn introduced me to Dungeons & Dragons in the mid-1980s. Our rural bus ride to school gave us about thirty minutes each way to play throughout the week. We used the 1983 Red Box set, and I still own my original copy.

I became the referee almost immediately. I built a sandbox campaign filled with ruined shrines, caves beneath old hills, wandering monsters, and whatever else my adolescent brain thought sounded dangerous. Our starting party had a Cleric, an Elf, and a Thief, along with a steady stream of replacement characters after poor decisions underground. We made memories that outlasted graduation.

We barely understood the rules. Nobody argued about “rules as written.” We rolled dice on a moving school bus while debating torchlight, the weight of treasure, and sentient screaming mushrooms. We learned the game through repetition, improvisation, bad calls, and occasional luck. No internet culture surrounded the hobby yet. Nobody uploaded DM tutorials to YouTube. No podcasts explained the “correct” way to referee. We learned by doing, and I miss that approach.

The modern tabletop hobby turned refereeing into a technical profession obsessed with optimization. Publishers market encounter systems, pacing guides, lore books, villain manuals, and cinematic storytelling advice as necessities instead of optional tools. Wizards of the Coast built entire editions of Dungeons & Dragons around heroic character identities and carefully managed play experiences because those systems appealed to players, who were already shaped by video game progression (map) loops and modern fantasy media. Plenty of groups enjoy that style of play. I understand why.

Still, too many referees now equate good play with deeper prep, denser lore, and tighter narrative control. As an autistic referee, I think some of this advice pushes tables in unhealthy directions. Modern adventures often hide social scripting behind the language of immersion. Villains arrive with prepared speeches. NPCs reveal clues and information in a particular order. Emotional scenes expect the “right” reaction from players, or freeze. Before long, players stop interacting naturally with the world and start searching for the referee’s intended answer. I have watched entire tables lock up because nobody wanted to say the wrong thing to the important NPC. Worse, they defaulted to dice rolls while desperately pointing to their character sheets.

I created that problem myself when I was younger. I spent absurd amounts of time writing lore documents, timelines, regional histories, heraldry, prop letters, and carefully planned dialogue that barely survived contact with players. I burned through ink cartridges and stacks of paper, preparing material nobody at the table would ever see.

Eventually, I realized I had stopped preparing a game. I was outlining scenes, planning reveals, and hoping players walked into the correct conversations at the correct time. My process resembled drafting fiction far more than running a sandbox campaign.

Players are not actors hired to deliver emotional payoff for the referee’s outline.

The older I get, the more I value uncertainty at the table. Early editions of D&D embraced uncertainty constantly. Reaction rolls, morale checks, random encounters, and open-ended exploration disrupted the referee’s assumptions every session. NPCs reacted unpredictably. Players committed to ridiculous plans that somehow succeeded. Entire campaigns changed direction because somebody rolled well at the right moment or disastrously at the wrong one.

The referee gets surprised too.

…and this surprise creates genuine interaction instead of guided performance. Players negotiate naturally because nobody already knows how the scene should end. Referees improvise because the world no longer follows a hidden script.

For many neurodivergent players, that difference makes their gaming world more manageable.

Many autistic people already spend enormous amounts of energy navigating unspoken social rules, rehearsed responses, and pressure to locate the “correct” conversational choice in everyday life. Some heavily scripted campaigns reproduce that same tension at the table. Players begin second-guessing themselves instead of experimenting, roleplaying, negotiating, or taking risks.

Older styles of play avoid some of this by default. The referee presents situations instead of predetermined emotional outcomes. Players test ideas. Conversations collapse. Monsters react badly. Negotiations succeed unexpectedly. The game moves forward because everyone adapts in real time. This interaction and adaptation feel healthier to me as a referee, an autistic adult, and an autism advocate.

Ironically, I think many referees would improve more by reading fiction than by buying another stack of supplements. Read Howard. Read Leiber. Read Moorcock. Read modern fantasy if you prefer. Study how characters threaten, bluff, retreat, negotiate, panic, and survive under pressure. Visit your local library instead of purchasing another “Complete Guide to Tactical Monster Personalities” PDF. Referees who read broadly improvise better at the table. Meanwhile, I have watched referees drown themselves in supplements while growing less comfortable improvising a simple conversation with players.

Their problem usually reveals itself the moment players do something unexpected.

I still love Dungeons & Dragons. I still love graph paper, wandering monsters, improvised rulings at midnight, and terrible character accents that somehow survive entire campaigns. I also understand why modern publishers moved toward more cinematic and heavily authored styles of play. Those systems onboard players easily and produce recognizable experiences for streaming audiences and modern gaming culture.

Still, I think the hobby loses something important when referees stop acting like adjudicators and start acting like directors managing performances.

The moments my players still talk about were never scripted. Those moments grew out of terrible plans, lucky rolls, bad misunderstandings, cautious diplomacy, or details I invented thirty seconds earlier because somebody unexpectedly fixated on them.

None of those memories came from perfect preparation.

Most of it started with a handful of dice rattling on a rural school bus in the 1980s.