I remember the first trick I saw at the first magic convention I ever attended. I remember the last trick of the gala show that Saturday night. Everything in between is a blur.
Not because the middle was bad. I am sure it was excellent. But if you asked me right now to describe what happened between the opener and the closer, I would give you vague impressions — a table, some cards, something with rings? — and nothing concrete.
I used to think this was a personal failing. That maybe I was not paying close enough attention, or that the middle acts were somehow weaker. Then I discovered that this is not a personal failing at all. It is a universal feature of human memory, and it has a name.
Two names, actually.
The primacy effect: we remember the first items in a sequence better than those in the middle.
The recency effect: we remember the last items in a sequence better than those in the middle.
Together, they are called the serial position effect, and they have been one of the most replicated findings in memory research for over a century. Every psychology student learns about them in their first year.
But here is the part that most performers do not know, and it is the part that changed how I think about show structure. When Gustav Kuhn and his colleagues at the MAGIC-lab at Goldsmiths University tested the serial position effect specifically in the context of magic performances, they found something surprising.
The advantage fades. In about ten days, it is gone.
The Research That Surprised Me
The researchers showed participants a magic show — multiple effects in sequence — and then tested their memory at different time intervals. Immediately after the show, the serial position effect was clear and strong. People remembered the first and last effects significantly better than those in the middle, exactly as the psychology textbook would predict.
But when they tested again after a longer delay — roughly ten days — the advantage of the first and last positions had largely disappeared. The primacy and recency effects, which seemed so robust in the immediate aftermath of the show, eroded as the overall memory of the show degraded.
This finding stopped me in my tracks. Because the conventional wisdom in magic — the advice given in virtually every book on show construction I have read — is that your opener and closer matter most. Put your strongest material at the beginning and end. Bookend your show with your best effects. The middle is where you can take risks or experiment.
And that advice is correct. But it is only correct for the first ten days.
After that, the entire show flattens. The primacy advantage disappears. The recency advantage disappears. What the audience remembers — if they remember anything at all — is determined by factors other than position.
What Survives the Ten-Day Threshold
This raised an obvious question for me: if position-based memory advantages fade, what determines what survives beyond ten days?
I spent a long night in a hotel room in Salzburg thinking through this, mapping it against my own experiences as both a performer and an audience member. Here is what I came up with.
Emotional intensity. The effects that provoke the strongest emotional response — the biggest gasps, the loudest laughter, the most intense moments of astonishment — survive longest in memory. Emotion is a memory fixative. It acts like a chemical preservative, protecting certain memories from the decay that claims the rest.
Personal involvement. If the spectator participated in the effect — if they held the card, made the choice, felt the impossible thing happen in their own hands — that memory is encoded more deeply and resists decay better than memories of effects they merely watched.
Narrative coherence. Effects that tell a clear story, that can be recapped in a single sentence (“he knew what I was thinking before I said it”), survive because they are easy to store and easy to retrieve. Complex, multi-phase effects with elaborate procedures are harder to compress into a memorable narrative and are therefore harder to recall.
Distinctiveness. If one effect in the show was dramatically different from the others — a mentalism piece in a show full of card tricks, a dangerous-looking stunt in an evening of comedy — the distinctiveness of that moment protects it from blending into the undifferentiated middle.
These four factors — emotional intensity, personal involvement, narrative coherence, and distinctiveness — are position-independent. They work regardless of whether the effect is first, last, or buried in the middle of the show.
The Two Time Horizons
What I realized, standing in front of the bathroom mirror in that Salzburg hotel room at one in the morning, brushing my teeth and still thinking about this, is that there are two fundamentally different audiences for every show you perform.
The first audience exists in the zero-to-ten-day window. They are the people you will see at the after-party, the networking session, the coffee break the next morning. Their memory is still fresh enough to be shaped by serial position. For this audience, the traditional advice holds: your opener and closer will dominate their memory.
The second audience exists beyond the ten-day window. They are the people who will tell the story of your show to friends weeks later, who will decide whether to book you for next year’s event, who will remember you — or not — when your name comes up in conversation six months from now. For this audience, serial position is irrelevant. Only the four survival factors matter.
Most performers design exclusively for the first audience. They optimize for the immediate reaction. And that is important — the immediate reaction determines the applause, the standing ovation, the feeling in the room. But the second audience is the one that determines your career.
How This Changed My Show Structure
When Adam Wilber and I were developing material for Vulpine Creations, we spent a lot of time talking about what makes an effect memorable. Not impressive in the moment — memorable over time. The distinction matters more than I realized before I understood the serial position research.
Here is how I restructured my thinking.
For the opener, I still choose something strong. The primacy effect is real for the first ten days, and those first ten days include the event feedback forms, the post-event conversations, and the decision-maker’s immediate impression. You do not waste your opener.
For the closer, I choose something even stronger. The recency effect ensures it will dominate the audience’s immediate memory, and if I choose correctly — something emotionally intense, personally involving, and narratively clean — it will also survive the ten-day threshold.
But here is the change: I no longer treat the middle as a place to hide weaker material. Because the middle is where the effects need to earn their own survival. Without the crutch of serial position, middle effects must be intrinsically memorable. They need emotional intensity, audience involvement, narrative coherence, or distinctiveness — ideally more than one of these.
This means the middle of my show actually needs to be designed more carefully than the bookends, not less. The opener and closer get a positional bonus for free. The middle gets nothing. Every effect in the middle has to justify its own existence in the long-term memory of the audience.
The Keynote Parallel
I see the same principle in my keynote speaking work, which is where most of my performance happens these days. Keynote audiences remember the opening story and the closing call to action — for about a week. After that, what survives is the single idea that resonated most deeply, the personal story that connected most emotionally, and the moment that felt most relevant to their own lives.
I had an experience in Klagenfurt last year that made this vivid. I ran into a woman who had been in the audience at a keynote I gave nine months earlier. She could not remember how the talk started or ended. But she remembered, in precise detail, a story from the middle of the presentation about a time I failed at something and what I learned from it. She quoted it back to me almost verbatim.
That story had survived because it hit three of the four survival factors: emotional intensity (I was honest about a genuine failure), narrative coherence (it was a clean, simple story with a clear lesson), and distinctiveness (it was the only moment in the presentation where I showed vulnerability).
Position? She had no idea where in the talk it appeared. And it did not matter.
The Implication for Effect Design
If you accept that the serial position effect fades in roughly ten days, the implication for effect design is straightforward but uncomfortable: every effect in your show needs to be designed for long-term memorability, not just immediate impact.
This means asking harder questions during the design phase.
Will they remember this in a month? Not “will they react to this in the moment” — that is a lower bar. The question is whether the effect has the structural qualities that make it resistant to memory decay.
Can they tell this story in one sentence? If the effect requires a three-minute explanation to convey why it was impressive, it will not survive ten days. The effects that last are the ones with a one-sentence headline: “He predicted what I was going to say.” “The card appeared in my wallet.” “He read my mind.”
Did they feel something? Intellectual appreciation decays fast. Emotional impact decays slowly. If the audience thought “that was clever,” it will fade. If the audience felt “that was impossible,” it will stick.
Were they involved? Effects they watched will be forgotten. Effects they participated in will be remembered. The spectator who held the card, made the choice, opened the envelope — they have a personal investment in the memory that protects it from decay.
The Uncomfortable Truth
Here is the uncomfortable truth I arrived at after months of thinking about this.
Most magic shows are designed for the immediate aftermath. For the applause. For the standing ovation. For the “how did you do that?” at the bar afterward. And that is fine, as far as it goes.
But the effects that actually matter — the ones that build your reputation, generate word-of-mouth, and make people remember you years later — are not necessarily the ones that got the biggest reaction in the room. They are the ones that survived the ten-day threshold. The ones that made it through the steepest part of the forgetting curve. The ones that had enough emotional weight, enough personal relevance, enough narrative clarity, and enough distinctiveness to persist after the serial position advantage had evaporated.
I think about the effects I remember from shows I saw years ago. Not the openers or closers — I cannot remember which was which. I remember the moments that hit me. The moments that involved me. The moments that told a story I could retell.
Those are the moments that survived. And designing for those moments, deliberately and systematically, is a different discipline than designing for the immediate reaction.
Both matter. But only one of them lasts.
What I Do Differently Now
Every effect in my show now gets evaluated on two separate axes. The first axis is immediate impact: Will they react? Will they gasp, laugh, applaud? Will the room come alive?
The second axis is ten-day survival: Will they remember this after the serial position advantage fades? Does it have emotional intensity? Does it involve them personally? Can they recap it in one sentence? Is it distinct from everything else in the show?
If an effect scores high on immediate impact but low on ten-day survival, it can go in the opener or closer position, where serial position gives it a temporary boost.
If an effect scores high on ten-day survival, it can go anywhere — including the middle, where it will earn its own persistence without any positional help.
If an effect scores low on both, it does not belong in the show.
The serial position effect is real. It is powerful. And it expires. Knowing all three of those things simultaneously is what changed my approach to building a show that matters not just tonight, but next month, and next year, and in the stories people tell about what they saw long after the lights come back on.