The History of Arrow Puzzles: From Sliding Blocks to Clear Shot

Pick-up sticks: take one, disturb nothing

The oldest relative is a game played with a heap of thin sticks. The Victoria and Albert Museum in London has a set of carved spillikins, probably made around 1900 from ivory or bone. One player drops the pieces in a pile. Each player in turn then lifts out one piece with a small hook, and the turn ends as soon as any other piece moves.

The game has many names. The Horniman Museum lists Pick-Up Sticks in Britain, Mikado across Europe, Jackstraws in the United States and Jonchets in France. Both museums record it as thought to have come from China, and neither gives a date, so that is as much as can be said about its beginnings.

The link to an arrow puzzle is loose but real. In both, you look at a tangle and ask which piece can come away right now. A stick lying on top is the heap’s version of a free arrow. The difference is that a heap of sticks is ruled by balance and a steady hand, and an arrow board is ruled by straight lines.

Sliding blocks in a tray

On December 14, 1907, Lewis W. Hardy of Chicago filed a patent for a puzzle of ten flat blocks of different sizes in a shallow box. One block is taken out, and the rest are shifted by sliding, without lifting or turning them. The patent was granted on February 20, 1912. A cardboard version called the Pennant Puzzle carries Hardy’s name and the year 1909, and a 1926 version was sold as Dad’s Puzzler.

The goal in Dad’s Puzzler is to bring the large square block from one corner of the tray to another. Every other block is only in the way. A close relative with the same idea, known as Red Donkey and also as Klotski, was patented in England by J. H. Fleming in 1932. Edward Hordern’s 1986 book Sliding Piece Puzzles, the standard reference, ranks these two behind only the Fifteen Puzzle in sales.

These puzzles introduced the question that every later “clear the way” puzzle asks: what is blocking this piece, and what is blocking that. In a sliding puzzle the answer keeps changing, because every move puts a block somewhere new.

Rush Hour: one car out of the jam

The Japanese puzzle designer Nob Yoshigahara first imagined a traffic-jam puzzle in the late 1970s, according to The Strong National Museum of Play, which holds a copy. It went on sale in the United States in 1996 under the name Rush Hour, from a company then called Binary Arts and now known as ThinkFun. The museum’s record calls it the company’s most popular seller.

The pieces are cars and trucks, two or three squares long, on a small grid. Each one can slide only along its own length, forward or back. You shuffle them until one particular car has a clear run to the exit. A piece that moves in one direction only, and needs an empty line ahead of it, is the nearest thing in the older puzzles to an arrow with a line of flight.

Rush Hour is also much harder than it looks. In 2002 Gary Flake and Eric Baum proved that the generalized puzzle, on boards of any size, is PSPACE-complete, which puts it among the problems computer scientists consider hardest to solve in general. Robert Hearn and Erik Demaine later gave a simpler proof and showed the same for ordinary sliding-block puzzles.

Tap-away blocks on phones

Phones produced a simpler kind of unblocking puzzle. A game known as Tap Away 3D, from the German studio Popcore, has been tracked on the App Store since May 21, 2021. Its store page describes the idea in two sentences: tap the blocks to make them fly away and clear the screen, and each block will fly in one direction only.

That is a real change from Rush Hour. A sliding piece stays on the board and gets in the way somewhere else. A piece that flies away is gone for good. Each tap makes the puzzle smaller and never makes it worse, so the challenge shifts from planning a long sequence to seeing which pieces are free.

Many games of this kind followed from many publishers, and their records are store listings and download charts, not histories. Who first had the idea of a piece that leaves when tapped is not something the available sources settle, so this chapter doesn’t name an inventor.

Arrows on a grid

The flat version, with arrows drawn on a grid in place of blocks, took off on phones in 2025. The best-documented example is a game known as Arrows – Puzzle Escape, from the German studio Lessmore, which trackers date to the summer of 2025. Its store page states the goal as extracting each arrow without causing a collision.

It spread quickly. Sensor Tower listed that game among the five most downloaded puzzle games on Android in the United States for the last quarter of 2025. An analysis of Sensor Tower data for April 2026 reported that two arrow games together accounted for a quarter of all downloads in their puzzle category that month. Many similar titles now sit beside them in the app stores.

The games differ in what a mistake costs. The Lessmore listing describes play with no timer, where collisions cost hearts. Others add a clock or a move limit. The shared core is the one rule: an arrow leaves if the line ahead of its head is empty.

The mathematics: why the board always clears

Draw a dot for every arrow, and draw a link from one dot to another whenever the first arrow can’t leave until the second has gone. Mathematicians call the result a directed graph, and each link is a dependency. A cycle is a ring of links that leads back to where it started. The simplest cycle is two arrows aimed at each other along one line, and neither could ever leave.

A standard result says that the dots can be put in an order that respects every link exactly when the graph has no cycle. Such an order is called a topological order. For an arrow board, that means the board can be cleared if and only if its blocking graph has no cycle. Removing an arrow only deletes links, so a clearable board stays clearable whatever you shoot.

In 1962 A. B. Kahn described a method for finding such an order. Collect every item that depends on nothing, take one out, and see which items that sets free. Repeat until nothing is left. It is exactly what you do in this puzzle: find a free arrow, shoot it, and look at what just opened. The same method runs wherever jobs must wait on other jobs, such as course prerequisites or the order in which to build the parts of a large program.

How LarkSpark plays it

Clear Shot is LarkSpark’s own daily take on the idea. The board is a 20×20 grid packed with arrows, usually between 80 and 100 of them, and their shafts turn corners. Everyone who plays on the same date gets the same board, so times can be compared fairly. The full rules cover the details.

There are no lives and no move limit. The only measure is the clock. Tapping a blocked arrow is a bump, which adds five seconds and is counted on your result. A hint lights up one free arrow at no cost in time, and the number you use is kept with your record.

Every board is built so that it can be cleared, and at each moment at least one arrow is free. That is the mathematics above put to work: there is no wrong order, so the puzzle is about how quickly you can see. The solving tips start with the border sweep and go on from there.

The history of Clear Shot