Mailbag #182

What are your thoughts on the Nameseek app? Specifically the point brought up in this Cafe thread that says people are going to be suspicious of going to a website to look up a name, especially the younger generation because these days they look up everything via AI/ChatGPT. Fair point in your opinion?—AT

Generally, I don’t have any thoughts on the app as I haven’t used it. The creator once asked me to take a look at it, but when I said I didn’t do reviews on the site, that was kind of the end of the discussion. So I can’t tell you one way or another if the site itself looks or feels legitimate.

As for the contention that the younger generation doesn’t look things up on website, I get where that’s coming from, but it honestly sounds like an old person’s understanding of young people, gleaned entirely from Facebook posts shared by other old people: “They don’t look stuff up on websites. They can’t type; they can only communicate in emojis. All their dances come from Fortnite. And apparently, they’ve stopped having sex altogether!”

If anything, the younger people I know are a little more suspicious of AI than the older generation—at least the portion of the older generation that have adopted the technology and treat it like an infallible electronic grandson.

Regardless, this particular concern is easy enough to sidestep. First, don’t say, “Let’s look up the meaning of a name. Go to this website.”

Say, “Go to this website.” Then say, “We’re going to look up the meaning of a name.”

This gets them onto the site before the subject of name meanings comes up. Now it doesn’t make sense for them to leave the site and go to ChatGPT or something similar.

This is related to another piece of advice I’ve given in the past. Don’t say, “I’m going to read your mind. Write down a word.” That’s introducing those ideas in the wrong order. It gives them a chance to say, “Wait, why do I have to write down the word if you’re going to read my mind?” Instead, say, “Write down any word.” Only then do you introduce the idea that they’re going to “just think” of the word they wrote down and you’ll read their mind.

In both cases, you get them to take the action first and explain its significance second. That way, the explanation doesn’t give them an opportunity to question the action.

The other thing you have on your side—regarding why a website makes sense for this information—is that we’re not looking up the capitals of the world, where there’s one objectively correct answer and it shouldn’t matter where you get it. Name meanings are interpretive. Different sources give different answers. So choosing one particular source to work from makes perfect sense.

If you feel you need to justify it, you can say, “Do you know what your name means? The problem is, a lot of names have multiple meanings, and if you look them up online, you’ll get half a dozen different answers depending on where you go. I like this site. It seems to have the most thorough and up-to-date information, so this is the one I use.”

So if you were genuinely testing your ability to pick up on this information, you probably wouldn’t use ChatGPT. You’d want everyone working from the same supposedly reliable source. The website isn’t something you need to apologize for—it’s probably the more logical procedure.

---

Okay, I love your magic style, the thinking behind it, and the tricks in your books. But I think the part that actually amazes me the most is your work ethic. Ten years of posts, newsletters, and books always on schedule? That’s rare anywhere and basically unheard of in magic. How do you actually stick to that schedule? Do you have any tips for someone hoping to start a similar project?—NG

Eleven years, actually. Don’t undersell my accomplishment.

Have you ever heard the song Bug Like an Angel by Mitski? Incredible song and an even better music video.

It’s a song about alcoholism, and it has this lyric:

Did you go and make promises you can't keep?
Well, when ya break them, they break you right back

While I don’t have any addiction issues, I can still relate to this sentiment.

We tend to think that breaking a promise means screwing someone else over so we can do what we really want. But if the promise was sincere, the greater damage is to yourself. Every time you break one, you make it a little harder to believe yourself the next time you say, “I’m going to do this.”

I didn’t promise you I would write 180 posts a year, a couple hundred pages of newsletters, and a book every 18 months. I promised myself that. I adore myself. I’m delightful. I don’t want to fuck that guy over, so I keep the promise.

You can’t rely on the universe or other people to always give you what you need. But you should be able to rely on yourself or else you’re just adrift.

The catch is that you can’t promise yourself everything. They need to be rationed. If everything is a promise, then nothing is. You’ve just turned “I promise” into a more dramatic way of saying, “I’d like to.”

So you choose a few areas—your work, your relationships, your health—and make a small number of serious promises about them. Keeping those promises becomes a priority. In other areas, you make an honest effort but without demanding perfection.

I know you want a better answer. Some clever scheduling or productivity advice or something. But none of that really matters. You get the things done that you don’t give yourself an option not to get done. If the project you’re considering is important enough to you, you’ll make that commitment. If not, you’ll probably start it with the best of intentions and then eventually abandon it because staying on top of something like that is hard as hell.

Dustings #151

The testing in yesterday’s post raises the question: How much does using a phone in the set-up for a trick affect the reactions? In other words, both versions have the same physical reveal, but one has a physical set-up and the other has a digital one. For example, in one version, they select an item at their chosen number from a note on your phone. In the other, they select an item at their chosen number from a page in a physical notebook. Will the digital version have the same detrimental effect?

Probably not to the same degree. I think that when you have a physical reveal versus a digital reveal, you’re comparing something that feels innocent with something that seems questionable. But with a physical set-up versus a digital set-up, both are going to seem somewhat suspicious. That’s just the nature of setting up a magic trick. So I think using a phone will have less of an impact there. Although perhaps we’ll test it officially at some point in the future.

---

Here’s a little magic-advertising trick for you.

Let’s say you’re releasing a packet trick. It’s just a simple packet trick. What could you possibly say about a simple packet trick?

That’s right: “Much more than a simple packet trick.”

This works with anything.

It’s just another version of the 21 Card Trick?

“Much more than just another version of the 21 Card Trick.”

It’s just a Hot Rod?

“Much more than a Hot Rod.”

The beauty of this phrase is that it acknowledges the exact objection people will have when they see the trick, then gives the impression of answering it without actually saying anything at all.

Congratulations. You’re now a magic copywriter.

---

Customer support at Vanishing Inc. responding with a Zen koan designed to free magicians from their attachment to owning things.

”This download can’t be downloaded.” Please meditate upon this until you no longer desire the instructions.

The Physical Advantage

In Toronto this past spring, we did some focus-group testing on the impact of a digital prediction versus a physical prediction. We’ve looked at this before, but we wanted to do some additional testing because this feels like one of those areas where the data will always be in flux. Our relationship with technology—and the technology itself—is constantly evolving. Here’s what we looked at:

1. Creepy Child Variation of Directed Verdict with a paper reveal.

2. Creepy Child Variation of Directed Verdict with an Instagram reveal.

3. Lottery Prediction with a physical paper reveal.

4. Lottery Prediction with a photo reveal on the magician’s camera roll.

5. Lottery Prediction with a photo emailed to the participant before the effect.

Twenty-five individuals saw each of the first two effects. Twenty individuals saw each of the final three. There was no overlap between any of these groups. In total, 110 individuals saw one of these effects, among other things we tested.

For the Creepy Child effect, the participants shuffled and cut the deck. The card they cut to matched either a folded paper prediction or an old Instagram post featuring a drawing—both supposedly created by some creepy kid the magician knows.

For the Lottery effect, the participants freely named numbers. Those numbers were either printed on a physical lottery ticket in the magician’s wallet, shown in a picture of a lottery ticket on the magician’s camera roll, or shown in a picture of a lottery ticket sent to the participant as an email attachment before the trick began.

We asked the participants to rate each effect on only one metric: how impossible it felt.

Results

For the Creepy Child effect

The digital version scored a 7.0.
The physical version scored an 8.4.

For the Lottery effect

The camera-roll version scored a 7.3.
The email-attachment version scored an 8.2.
The physical-ticket version scored a 9.1.

Takeaways

It may seem like there’s only a point or two separating these variations, but that point or two actually represents a very significant difference.

Having done this testing for over a decade now, we know that unless we spend a significant amount of time explaining how to rank the effects on a scale from 0 to 10, any effect that fools someone will receive at least a 5.5 or 6—even the most basic one-in-four multiple-out effect. At the high end, people are hesitant to call anything a perfect trick—at least when they feel like they’re trying to provide an accurate rating for data purposes—so most people won’t go beyond a 9.5, even when they genuinely have no idea how something was done. In practice, then, the scale runs from about 5.8 to 9.5.

That gives us a meaningful range of only 3.7 points. Viewed within that range, the 1.4-point advantage of the physical Creepy Child prediction over the digital version represents nearly 40% of the entire usable scale. With the Lottery Prediction, the 1.8-point gap between the camera-roll reveal and the physical ticket represents nearly half of that scale. Put another way, on the portion of the ten-point scale that people actually use, the physical reveals had more than twice the impact of the basic digital reveals. What looks like a difference of only a point or two is actually a massive gulf.

All else being equal, you should always be trying to pull your reveal into the real world because the potential gains are enormous.

Of course, all else is never really equal. Here, we were able to perform identical versions of the tricks regardless of whether the reveal was digital or physical. But that usually won’t be possible. We could only accomplish such a clean physical Lottery Prediction because we had a hidden printer and a secret assistant. That’s hardly an apples-to-apples comparison with a photo prediction you can perform impromptu with nothing but your phone.

There are Lottery Predictions you can perform more casually that still have a physical reveal, but the number-selection procedure isn’t as free. So then the question becomes, “Is it worth sacrificing the freedom of the number selection for a physical reveal?”

Before this testing, I think I would have said, “Go with the free selections and the digital reveal.” But after seeing how much greater the impact was with the physical ticket, the choice no longer seems nearly as obvious. I’m now far more willing to sacrifice some freedom—or accept a more convoluted selection procedure—to get that prediction into their hands.

Follow the Leader Final Beat

Here’s an ending for Follow the Leader-style effects. I’ve always been intrigued by the effect and have enjoyed practicing it, but in performance, I think the reactions are rather weak.

Here’s Juan Tamariz performing it.

I would describe the trick as repetitive, confusing, anticlimactic, and entirely magician-centric.

It’s repetitive because it’s essentially the same thing happening every time. It’s confusing because the spectator has to keep track of which face-down packet is where. It’s anticlimactic because there’s no real ending. And it’s magician-centric because the audience has no influence over anything that happens.

Here’s a simple ending you can add to the routine that helps address these issues.

You do the routine as you normally would, and the audience will react to it as they normally would—i.e., “This is kind of interesting, but surely the magician is secretly moving those packets around somehow. I don’t know…I’m not following it 100%.”

That’s fine. In this version, those early phases aren’t being asked to carry the entire trick. They’re building toward a climax—one that’s narratively consistent with the traditional Follow the Leader effect but places the outcome in the spectator’s hands.

To accomplish this, you’ll need a card like this:

You’ll also use Paul Harris’s Pointer Anomaly principle (explained here).

After the main body of the trick is completed, you explain, “Now, you’re going to say, ‘Oh, I don’t know exactly what he was doing, but I’m sure he was secretly moving the cards around somehow.’ I swear, that’s not how it’s done. In fact, I won’t even touch the cards during this final round.”

You introduce the red-and-black arrow card. You allow them to spin it between the piles and turn it over in either direction. No matter what, it will end up pointing to the correct piles.

To make the arrow card, put the image on a blank-faced card. From the back of the card, you’ll need to mark the opposite end from the point of the red arrow. That’s the end you want facing the red pile when the card is face down.

Allow the person to rotate the face-down arrow card between the piles, stopping whenever they want, with the card lying horizontally between them.

If your mark is near the red pile: Give them the choice of turning the card over from right to left or left to right. Imply that this is the final, most important choice.

If your mark is near the black pile: Say, “And now, so it’s a totally blind choice on your part, I’ll swap the piles a few times so you don’t know which is which.” Put the piles under the table or behind your back and exchange them once, although you pretend you’re doing it multiple times. Bring them back out, making sure to set the red pile on the appropriate side of the arrow card. “Now for the choice that will make all the difference: Turn that card over to the left or to the right.”

Ultimately, of course, it’s just a 50/50 hit. But the important thing is that it no longer feels like they can write the whole thing off as sleight of hand. The spectator positions the arrow, chooses how to turn it over, and apparently determines everything that happens. Yet the red arrow still points to the red pile, and the black arrow still points to the black pile. The trick finally ends with one clear image that was seemingly out of your control

My favorite version of Follow the Leader—although I’m no expert—is Peer Pressure, which can be found in this Liam Montier book. It’s very simple and straightforward for both the performer and the spectator to follow.

The Alright Force

Today I have a Toxic Force alternative to share with you that’s significantly better than the Toxic Force in almost every respect. (The only drawback is that you need to be using an iPhone.)

It has three significant advantages:

1. All the math is real. Spectators can follow along on their own phones or calculators.

2. The initial numbers are freely chosen, and the final number is apparently generated at random by the spectator.

3. Everything happens out in the open. The phone never needs to be turned over or hidden from view.

This comes from supporter Dave Mason, who built the web app and gave it to me to share with you today.

Here’s what it looks like: Your spectator performs a series of calculations on your phone. For the final one, they apparently enter a random number by drumming their fingers across the screen. When they press equals, the result is whatever force number you’ve set.

For example, you ask for a random number. The first spectator offers the number 69. You enter it in the calculator. You ask another spectator for a number to multiply it by. They also choose 69. Such clever spectators you’ve assembled. “Now let’s add a completely random number.” You have a spectator drum their fingers across the screen, apparently entering digits at random. When they press equals, the result is your force number—in this case, 123456789.

If you want, you can have one or more people follow along on their phones to confirm the total.

I call it the Alright Force because toxic things are bad, while alright things are, at minimum, alright. More importantly, Dave Mason of Traffic wrote “Feelin’ Alright”, and you don’t get many opportunities for that sort of naming synergy.

Instructions

[Note: This is designed for iPhone and has been successfully tested on an iPhone 14 Pro and an iPhone 15 Pro Max. It think itshould work on other modern iPhones as well, although it has not been tested. The code will be provided at the end for anyone who may need to make a modified version.]

Add CalcDM to Your iPhone

CalcDM doesn’t need to be downloaded from the App Store. You can add it directly to your Home Screen:

1. Open Safari on your iPhone.

2. Go to calcdm.com.

3. Tap the Share button—the square with an upward-pointing arrow.

4. Scroll down and tap Add to Home Screen.

5. Tap Add in the upper-right corner. (Keep Open as Web App on)

An icon will now appear on your Home Screen. It looks identical to the normal calculator icon. In the future, open the calculator by tapping that icon rather than returning to Safari.

Because the two icons look identical, put CalcDM somewhere you’ll remember—or move Apple’s Calculator app somewhere else—so you don’t open the wrong one during a performance.

Set the Force Number

Before performing:

1. Open CalcDM from your Home Screen.

2. Enter the number you want to force.

3. Press and hold the clock icon in the upper-left corner for about one second.

4. The calculator display will clear.

5. The clock hand will swing to the 9 o’clock position. This is your confirmation that the force number has been set.

The calculator is now ready to use normally.

If you need to confirm the force number later, press and hold the calculator icon in the upper-right corner. The stored number will be revealed so you can peek at it.

Perform the Force

1. Begin with ordinary calculations

Let the spectator enter numbers and perform calculations using addition, subtraction, multiplication, or division.

You’ll generally want to be careful with division here, as it’s likely to produce some decimal monstrosity you don’t really want to deal with. It will still work; it may just make things messier than necessary.

Continue until you’re ready for the final calculation that will produce your force number.

2. Choose the final operation

You can force the number using any of the four operators: addition, subtraction, multiplication, or division.

For the simplest handling:

• If the current total is below your force number, use addition.

• If the current total is above your force number, use subtraction.

2.5. Check the required number

This isn’t a necessary step—hence the .5—but it allows you to check whether a particular operation will produce a convenient number before you commit to it.

Press and hold the operator you’re considering. CalcDM will show you the exact number that would need to be entered with that operation to reach your force number.

For example, holding × will show the number by which the current total would need to be multiplied. This allows you to check whether the required number is a whole number, a decimal, or longer than you’d like.

Release the operator after checking it. Nothing has been armed yet.

3. Arm the force

When you’ve chosen your final operation, tap its operator twice in succession.

For example, tap + twice to arm an addition force.

The app automatically calculates the number required to bring the current total to your force number. You do not need to calculate anything yourself.

4. Have the spectator drum their fingers

Ask the spectator to drum their fingers across the face-up screen. Each tap secretly enters another digit of the required number.

Watch the clock icon as they tap. When all the necessary digits have been entered, the clock hand will move to the 6 o’clock position. This is your cue that the number is complete. At that point, have them stop tapping.

You don’t need to keep an eagle eye on the clock icon. It will likely be obvious when the entire number has been entered. The icon is just there so you can be sure.

If you’re worried they’ll notice that the digits appearing on the screen aren’t necessarily the ones beneath their fingers, I don’t think you have much to worry about. But to eliminate even that possibility, ask them to look into your eyes—or simply away from the calculator—as they drum.

5. Complete the calculation

Have the spectator press =.

The result will be your previously selected force number.

Because every previous calculation was legitimate, they can enter the same numbers into another calculator and confirm the final result for themselves.

Quick Performance Summary

1. Enter the force number.

2. Hold the clock icon to save it.

3. Perform any ordinary calculations.

4. Hold an operator if you want to preview the required final number.

5. Double-tap the operator you intend to use.

6. Have the spectator drum their fingers on the screen.

7. Wait for the clock hand to reach 6 o’clock.

8. Press = to reveal the force number.

CalcDM was created by Dave Mason and is available at calcdm.com. The source code is available on GitHub.

Mailbag #181

Just wanted to add some additional context to the part about Paul Romhany in the latest Dustings.

He's been obsessed with AI for awhile now. I actually sent you an email about him awhile back. He has a second Facebook "Paul R" (and an Instagram) where he posts a combination of AI generated performances and real performances of tricks. The real performances are a mix of Paul doing magic, other magicians doing magic at conventions, or snippets from trailers.

According to Jeff Prace, Paul doesn't ask permission before posting these performances at conventions and does not credit the original performer in the videos.

A new thing he has been posting lately is clips of magic trailers with their voices dubbed over with AI. Like here’s one where he just took a clip of Craig Petty from the new Sucker Punch trailer.

View this post on Instagram

A post shared by Paul R (@paulrmagic)

There was also a big controversy where he used AI [for an image in CAM Magazine] and it turned Rachel Wax into a completely different person. She called him out for it and then he said he had a family emergency and didn't have time to edit the [image] and that he was being bullied and harrassed for it.

That's about it. Just wanted to let you know how deep the Paul Romhany AI rabbit hole is.—AO

Oooh…I see. So this isn’t a one-off. He’s like a weird goon, then? That’s good to know. I was feeling kind of bad for him because he had to use AI to generate discussion about his magazine. If he hadn’t tried to pass the AI hosts off as real people, I would have just quietly pitied him.

But I guess he’s into all sorts of weird AI stuff. He’s definitely got a “grandpa just discovered ChatGPT” energy about him. A “Grok, take this picture of my neighbor and put her in a bikini” vibe.

Most confusing to me is, of all the things about Craig Petty you could use AI to replace…you chose his voice? That’s the part of Craig you take issue with?? If you asked Craig what he’d change about himself, I’m not sure his voice would make the top 50. If you’re going to steal his content and unnecessarily alter it, at least go all-in. Give us Craig Petty post-LASIK, post-hair transplant, and six months into a vigorous course of GLP-1s.

---

I need your advice. I have a venture in the magic world that I’ve been trying to get off the ground for a while now and I’m having a hard time getting traction with it. I’ll spare you the details so you don’t think I’m just trying to get you to promote it, but it’s similar to what you’re doing: putting out magic content and having readers support it. I’ve been trying my best but haven’t found an audience yet. My question isn’t how to get an audience but at what point do you bail on a project if you’re not finding support?—AH

Ideally, you decide that before you start. That's when you're least likely to confuse "I still believe in this" with "I've already put too much into this to stop."

Let’s say I'm opening a restaurant. (It’s a grilled cheese and chocolate milk restaurant. But that’s not important.) Before I sign a lease, I decide what would make me want to keep running it two years from now. Maybe that's two metrics:

1. $10,000 a month take-home.

2. More customers walking through the door than a year earlier.

If I hit both marks, great. If I miss both—I'm nowhere near the income I need, and the customer base isn’t growing even slowly—the decision is painful, but it isn't mysterious. If I hit one, I have a judgment call to make. But at least I know I’m partially on pace.

You can do the same thing now, even though you didn't do it at the start. Pick a date, maybe six months from today, and decide what would make the project worth continuing for you. It could be a number of paying readers, a monthly income, or steady growth that suggests you're on your way. Include what the work costs you, too. A project that earns little but gives you enormous pleasure is different from one that earns little and makes you dread sitting down to write.

Then give it an honest shot until that date. When it arrives, judge the project by the standards you set today—not by how much you've already poured into it, or by the one encouraging comment you got on a day you were thinking of quitting.

Dustings #150

I’ve never read a page of Vanish Magazine, and this—quite possibly the dumbest thing I’ve ever seen in magic—won’t be what changes that.

He had AI create a podcast "discussing" his magazine. And he's trying to pass it off like these are real people. They're not. They're fake hosts, fake enthusiasm, fake opinions—and he's promoting it as "NON-MAGICIANS talking about the latest issue of Vanish."

They're not non-magicians. They're non-humans.

That tells me two things.

One: he thinks so little of the material he's publishing that an AI-generated summary of it counts as worthwhile content to him.

Two: he thinks so little of his audience that he expects them to mistake two digital sock puppets for actual people having an actual conversation.

The kicker is—"non-magicians talk about a magic magazine" would already be the dumbest fucking thing you've ever heard if it were real. The fact that he looked at that premise and thought this needs to be faked takes a special kind of imbecile.

Of course, now I’m left wondering how much of the magazine itself is written by AI.

---

Here is the same magician selling his snowstorm effect. And what could be more aligned with the delicate beauty of producing a magical blizzard of snow than wincing and grimacing like you just took a sledgehammer to your scrotum before you activate the gimmick? He does this every time he activates it in the demo.

If only Houdini had braced himself this thoroughly before taking that punch to the abdomen. He’d still be dead, of course, but maybe he would have stuck around long enough to give us a sequel to The Grim Game.

---

I like to have fun, but on a day like today, I’m grateful to the Magic Café for setting frivolity aside and reminding us of the solemn importance of the occasion with its annual, breathtakingly beautiful tribute to those we lost.