Monthly Update #45 (September 2026)

🐸 Pond-life precognition and two Halloween treats from the bottom of the bag!

Well, this September monthly update is very late, even by my standards. But I do have a good excuse—this time, at least!

Back in Monthly Update #41, I told you that my role at work was at risk of redundancy because it didn’t appear in the proposed new structure. At the time, I had no idea what my day job would look like by the end of the year, or even whether I’d still be working in higher education. Well, I’m pleased to report that I still have a job at the university! 😅

I applied for and accepted a new role with a slightly different focus: rather than working solely on learning technology and supporting educators, I’m now managing a small team responsible for delivering all staff development across the university. Thank you again to everyone who sent kind messages when I first shared the news. They meant a great deal.

While I’m extremely grateful to still be in employment—and to still be working with many familiar faces—this has been a very difficult period in my professional life. Many friends and colleagues I’ve worked alongside for years have left, some through voluntary severance and others through compulsory redundancy. My thoughts are very much with everyone affected; I know how hard this has been for them, too.

My new role has also kept me busier than ever, as several of the areas I now oversee are completely new to me. It’s a bit like learning a new memorised deck: slow going at first, but I’m sure I’ll have it down by Christmas! 😉 As for my plans to become a part-time pro, they’re still a possibility—they’ll just have to wait until the New Year.

Right, that’s enough about my day job. On to the magic!

Toadal Recall 🐸

The other day, I stumbled across the “Cage Principle” on YouTube, explained by magician Greg Hughes of Trapdoor Magic in his video “The CAGE Mind Reading Method”.

Briefly, it works like this: you fold a sheet of paper into eighths, unfold and flatten it, and then write the letters A to H in the eight spaces, with A, B, C and D along the top row, and H, G, F and E along the bottom:

A B C D
H G F E

Following the creases, a participant folds the paper up however she likes, folding the short or long edges in any order she wishes. You trim the edges of the folded packet with a pair of scissors, which separates it into eight slips, and mix them so that some land face up and some face down. The slips are sorted into two piles by orientation, one pile is chosen, and your volunteer is asked to make a word from its letters. She always makes the word CAGE, hence the name of the trick. (Greg also adapts it to numbers chosen by the audience, which is a lovely use of the same idea.)

The underlying principle is an intriguing piece of origami-inspired mathematics. To see why it works, colour the eight spaces like a chessboard: each fold flips the paper, so all the spaces of one colour end up facing one way, and all those of the other colour face the other.

Thanks to this clever application of the parity principle, one pile always collects the first and third letters of the top row plus the second and fourth letters of the bottom row; the other pile gets the leftovers. As a result, whichever way the participant folds, the slips always sort themselves into two predetermined piles (in this case, CAGE and BDFH), even though, from your audience’s perspective, the process feels random and even a little chaotic.

Is CAGE a special case?

My first question was whether CAGE is the only feasible four-letter word when using consecutive letters of the alphabet, or whether there’s a viable alternative. It turns out CAGE is the only workable option—and emphatically so.

I asked Claude to check every consecutive run of the alphabet, at every usable length, against Collins Scrabble Words—that’s 279,077 entries! Only two runs produce a word anyone would recognise:

• A to F (six letters) gives A, C, E—ACE

• A to H (eight letters) gives A, C, E, G—CAGE

There are a handful of other technical hits, such as GIE, MOKI, SUQ and WUS. But these are tournament words nobody would build unless they play a great deal of Scrabble.

So there is no five-letter version of this trick, and no rival four-letter version, because the alphabet does not offer one up. That is the whole reason the effect has survived in exactly this one form and has become known as the “Cage Trick”. I’m not sure who to credit for its invention, though—let me know if you do.

CAGE has a second virtue that is easy to miss: its four letters make only one word. Give somebody A, C, E and G, and CAGE is the only four-letter word they can form. Plenty of other letter sets would leave them with a choice, and a force that offers a choice is, well, not a force at all! (One nice subtlety is to ask her to form a three-letter word first; she’ll make ACE or AGE. Then ask her to try and make a four-letter word. This helps disguise the fact that CAGE is her only option.)

Forcing a different word

The alphabet is only one thing you can write in those eight spaces. Write two ordinary four-letter words instead—one along the top row and the other along the bottom—and the same folding and cutting process forces whatever those two words conspire to spell. The two rows already encode the pattern. Therefore, with careful selection of words, the required letters will alternate automatically to form your target word. Because each word remains in normal left-to-right order, there’s also no need to reverse direction and use a snaking path when writing the letters on the paper.

Or you can skip the paper entirely. Lay the letters out as cards—a Lexicon deck, an alphabet deck, Scrabble cards or eight cards from any letter game. Four across the top for the first word, four underneath for the second. Your participant then folds the layout in on itself, row over row, column over column, turning each stack over as it goes, until the whole thing is one pile. The mathematics is identical, and it removes the need for scissors. One note of caution: when using cards, each “fold” must treat a whole row or column as a single unit, i.e. every card in the column or row must be folded inward.

I asked Claude to find the words with the longest list of usable pairs. I then had a lengthy conversation to work out which was the most suitable word from a practical standpoint. The winner was TOAD, with forty-four distinct pairs. Three things make it the best option:

1. Forty-four pairs is almost a full deck of options. You can choose pairs that look innocent enough and never repeat.

2. Every discarded pile spells nothing. Not one of the forty-four leaves a second word on the table. That is by design, not luck—pairs whose other pile made a readable word were thrown out.

3. TOAD is effectively a unique reading. Its only anagrams in Collins Scrabble Words are DATO (a Philippine chieftain or ruler) and DOAT (an archaic variant spelling of the verb dote). Neither is a word that your average person would produce, especially while under a little time pressure.

For comparison, SEAL has a far longer list, but S, E, A and L also make SALE, ALES, LEAS and LASE. While most spectators would spell SALE or SEAL, there’s a good chance someone might pick ALES, especially if you’re performing for a sharp-minded Scrabble enthusiast who’s actively trying to catch you out. (This could be addressed, of course, by using some kind of multiple out.)

The Forty-Four Pairs

Write the first word along the top row of the grid and the second along the bottom. After folding and trimming, the two piles hold the letters shown in the list below. One is always TOAD; the other never spells anything meaningful.

• THAN/WORD (TOAD & HNWR)

• WANT/DROP (WNRP & TOAD)

• PART/DOOR (PROR & TOAD)

• COST/DEAD (CSED & TOAD)

• DEAR/LOST (TOAD & ERLS)

• GOOD/TEAM (GOEM & TOAD)

• HAND/TOOL (HNOL & TOAD)

• DEAL/SORT (TOAD & ELSR)

• COLD/TRAP (CLRP & TOAD)

• IDEA/TOOK (IEOK & TOAD)

• FOOT/AIDE (FOIE & TOAD)

• GOLD/ALTO (GLLO & TOAD)

• FOOD/AUTO (FOUO & TOAD)

• HOLD/ANTE (HLNE & TOAD)

• ROAD/ARTS (RARS & TOAD)

• SOLD/ANTS (SLNS & TOAD)

• FACT/DOOM (FCOM & TOAD)

• MOST/DEAN (MSEN & TOAD)

• DATA/HALO (TOAD & AAHL)

• SOFT/DRAB (SFRB & TOAD)

• BOAT/DEAF (BAEF & TOAD)

• BOUT/DIAL (BUIL & TOAD)

• BOLT/DRAG (BLRG & TOAD)

• BOOT/DUAL (BOUL & TOAD)

• FAST/ODDS (FSDS & TOAD)

• COLT/DRAM (CLRM & TOAD)

• HARD/OATH (HRAH & TOAD)

• LAND/OATS (LNAS & TOAD)

• SAND/ONTO (SNNO & TOAD)

• DRAW/FONT (TOAD & RWFN)

• CARD/TROD (CRRD & TOAD)

• LOUD/TEAL (LUEL & TOAD)

• LAID/TWOS (LIWS & TOAD)

• SAID/TOOT (SIOT & TOAD)

• TOLD/TEAR (TLER & TOAD)

• WOOD/TRAY (WORY & TOAD)

• YARD/TROT (YRRT & TOAD)

• BOLD/TEAK (BLEK & TOAD)

• BOND/THAW (BNHW & TOAD)

• TIDE/YOGA (TOAD & IEYG)

• DUTY/SODA (TOAD & UYSD)

• LORD/TEAS (LRES & TOAD)

• SOFA/TIDY (SFIY & TOAD)

• THAT/CORD (TOAD & HTCR)

Which of the two piles ends up face up depends on how the packet finishes, not on which word you wrote first. If you want to swap the piles, simply turn the packet over so the face-up and face-down groups trade places.

Or you can let them freely select either pile and then, if they appear to be struggling to spell a word, have them use the other pile (this functions as a very nice convincer that the process is more random than it really is).

Alternatively, present this as a game between the two of you. Use Deddy Corbuzier’s Free Will Principle to make the prediction read true whichever pile the participant ends up with. (If you’re unfamiliar with this principle, here’s a good post from Don Bursell, of Don’s Magic & Books, that provides a list of learning resources on the topic.)

Forcing the pair

The neatest (and easiest) method I have found uses a deck of double-blank playing cards. Write one word of a pair on each side, so that whichever face a participant sees or selects, the reverse carries its partner. The choice is entirely free, but a force pair is guaranteed every time.

To deepen the deception, make up a fifty-two- or fifty-four-card deck by adding eight or ten decoy cards to throw people off the scent. These cards should force nothing or a different target word.

This two-word approach opens up many different possibilities, but it sacrifices something the original method provided free of charge: a reason for the letters to be on the paper (or the cards). A to H is self-explanatory. Two random words are not.

Dressing it as a memory game

The props still need a reason to exist, and my favoured approach is to frame them as cards from a paired-word memory game or test (hence the name of the trick). Nobody will question unrelated word pairs if memorising them is the whole point of the game. You can also have them play one honest round first, using only the decoy cards, then leave them out “accidentally on purpose” while pocketing the rest of the cards. If anyone is curious enough to check the letters, they’ll find the pairs make nothing at all (or different words from the target word).

With Claude’s help, I have also built two additional versions of this fake memory/recall game.

The printable PDF below works in the same way my mum used to run spelling tests when she was a primary school teacher: a masking slip you slide down the page, covering and uncovering one line at a time. It holds a genuine practice set that forces nothing, five sets of eight forcing pairs, and five full-sized recording sheets ruled into eighths and ready to fold.

The web app does the same job digitally and is hosted on itch.io (a free hosting site for indie games). It plays as an ordinary memory challenge until it is secretly switched into “Toad mode” (triple-click the word “training” in the footer to switch it on/off)! I’ve written up some basic instructions for the web app and published them to a secret page on my blog.

Paired Associate Recall Task

16.6KB ∙ PDF file

Download

A printable PDF that you can use to perform Toadal Recall.

Download

Two notes before you try it

Instruct your participant to create a three-letter word first. T, O, A and D also make DOT, TAD, OAT, TOD and ADO. Then ask her if she can manage to make a four-letter word. This avoids the situation where she might insist on forming a three-letter word, such as DOT, instead of TOAD. (If she makes TAD, you can reflect that the choice was very apt, as a tadpole does, in fact, turn into a toad!)

And if she arranges the right letters in the wrong order, don’t panic! As you can only make one everyday four-letter word, simply slide the required slips into place to “help her out”.

Over to you!

I don’t think this is a finished routine yet. The method is solid, and the prototype props are built, but the presentation is still a memory test with an odd reveal at the end. It deserves a better climax—the production of a realistic rubber toad, say, which you can buy online as a practical joke. 🐸 Or, if you decide to use cards, those decoys could be put to good use. Perform a lead-in effect to the prediction with them—something based on a progressive anagram, perhaps.

I am sharing it at this stage precisely because of its unfinished state. If something here ignites a better routine or idea, I would be delighted to hear from you and see where you take it. It’s half-baked by design, and I would be glad for it to become better in other people’s hands.

I will return to the routine myself and write up full instructions for it, but I’m unsure when that will happen.

---

From the Bottom of the Bag: Jazz, Murder & Magic! 🎷🪓🪄

In this new section, which will appear in my monthly updates every now and again, I’ll rummage through Marty’s Bag of Tricks and pull out an older blog post that deserves another outing.

With Halloween just around the corner, “Jazz, Murder & Magic!” seems like an appropriate choice. 🎃 The article includes links to two semi-automatic card tricks inspired by the terrifying real-life story of the Axeman of New Orleans, perfect for a spooky autumnal performance.

As these two tricks are largely self-working, you should be able to learn one of them well enough to perform this Halloween.

---

Three Videos Worth Watching 👀

Here are three more videos that I think are worth your time.

Card To Box by Lloyd Barnes

Duration: 13 minutes

A lovely Marlo-inspired tutorial from Lloyd Barnes. I don’t always agree with Lloyd’s opinions and general approach to magic—especially his views on the Elmsley Count (see my post “A Terrible Twist”)—but he’s a very generous creator who shares some excellent material on his YouTube channel. Funnily enough, he does use the Elmsley Count in this particular routine.

All Alike Display by Denis Behr

Duration: 26 seconds!

Here’s a very old video of German card expert Denis Behr performing a silky-smooth all-alike false display. If you’re wondering, the count he uses at the end is Eddy Taytelbaum’s Super Flushtration Count. I’m not sure why I haven’t used it before, as it’s very deceptive (thanks to the pacing and rhythm).

Why Magicians Hate Packet Tricks

Duration: 8 minutes

Staying on the same topic, here’s a great opinion piece by David (Magic Orthodoxy) that resonated with me. Obviously, I think packet tricks are great (see my regular Ruseletter column Packet Trick Paradise), and I’ve never understood why so many magicians profess a deep hatred of them. As far as I’m concerned, performers who snub them on principle are really missing out on some of the strongest and most practical card magic available.

---

Thank you for patiently waiting for this update; I’ll try my very best to get the next one published on time.

I’m also making good progress on the next edition of Corrupting the Classics, which will now feature no fewer than nine new card tricks (and some links to previously published material). I’m hoping to get this published in the first half of November.

Yours magically,

Marty