Connection and territory
Ponte del Diavolo is an underappreciated gem of a game. It was invented by Martin Ebel and first published by Hans im Glück in 2007. The rules are short and simple. However, the tactics and strategy that follow from those rules have great depth.
The board is a plain, 10-by-10, square grid. Play proceeds by each player placing two tiles or one bridge on their turn. It is a game of pure placement—no piece is ever moved or removed once placed.
To compensate for the first move advantage that is inherent in many two-player abstracts, Ponte del Diavolo uses the pie-rule. As in, "You cut, I choose." The first player places two tiles on any two spots on the board, then the second may either accept their opponent's move and play on or they may choose to switch colours, accepting the played tiles as their own first move. The natural effect is for the first player to play their first tiles in the centre rarely.
The few other rules are these:
1. On each turn a player must either:
3. Each lone island scores 1 point towards victory. If two islands are connected by a bridge, then that group would score 3 points, taken together. Likewise, a three-island group scores 6 points, a four-island group scores 10 points, and so on through the sequence of triangular numbers. (Which is, n(n+1)/2. Note that this formula applies to one-island "groups" as well!) At the game's end, the player with the greatest number of points wins. If the players' point totals are equal, the first tie-break is the total number of islands. If still tied, the total number of bridges.
4. The game ends when either player can no longer legally place two tiles and chooses to pass. (Because a player who can not place tiles may still opt to place a bridge.) If the passing player is black, the game ends immediately. If the passing player owns the white tiles, then the player of the black tiles gets a final turn.
Poisoned Tiles
The most important lesson that a new player of Ponte del Diavolo must learn is what I call poisoned tiles. There are certain arrangements of tiles that can never be promoted to an island. This is a non-obvious result of the relatively simple rules. The plainest example would be two orthogonally connected tiles that are diagonally adjacent to two other orthogonally connected tiles. If a player makes this error, not only will those four tiles be lost for scoring purposes, but—perhaps more critically—every square adjacent to these tiles is now also ruined for the purposes of placing a tile that may become part of a scoring island!
See the pair of light sandbars in Diagram 1. (Where sandbar is the term given to any orthogonally connected group of same-colored tiles that numbers less than four.) To connect the two sandbars with X or Y would create an illegal island of five tiles.
Also, to expand either sandbar to an island of its own would mean an island that is diagonally adjacent to same-coloured tiles, which is also illegal. See the pair of dark sandbars in Diagram 1. The player cannot play tiles at Z.
The board is a plain, 10-by-10, square grid. Play proceeds by each player placing two tiles or one bridge on their turn. It is a game of pure placement—no piece is ever moved or removed once placed.
To compensate for the first move advantage that is inherent in many two-player abstracts, Ponte del Diavolo uses the pie-rule. As in, "You cut, I choose." The first player places two tiles on any two spots on the board, then the second may either accept their opponent's move and play on or they may choose to switch colours, accepting the played tiles as their own first move. The natural effect is for the first player to play their first tiles in the centre rarely.
The few other rules are these:
1. On each turn a player must either:
- Place two tiles onto unoccupied, unblocked squares, or
- Place one bridge from any one of their tiles across unoccupied, unbridged squares to one of their other tiles.
- As well, you may not place a tile under any bridge. You may not place a bridge over any tile, or any other bridge.
- There must never be an island of five or more tiles.
- An island must also never be diagonally adjacent to a tile of its own colour. (This also means you may not form an island if it would be adjacent to any of your other tiles.)
3. Each lone island scores 1 point towards victory. If two islands are connected by a bridge, then that group would score 3 points, taken together. Likewise, a three-island group scores 6 points, a four-island group scores 10 points, and so on through the sequence of triangular numbers. (Which is, n(n+1)/2. Note that this formula applies to one-island "groups" as well!) At the game's end, the player with the greatest number of points wins. If the players' point totals are equal, the first tie-break is the total number of islands. If still tied, the total number of bridges.
4. The game ends when either player can no longer legally place two tiles and chooses to pass. (Because a player who can not place tiles may still opt to place a bridge.) If the passing player is black, the game ends immediately. If the passing player owns the white tiles, then the player of the black tiles gets a final turn.
Poisoned Tiles
The most important lesson that a new player of Ponte del Diavolo must learn is what I call poisoned tiles. There are certain arrangements of tiles that can never be promoted to an island. This is a non-obvious result of the relatively simple rules. The plainest example would be two orthogonally connected tiles that are diagonally adjacent to two other orthogonally connected tiles. If a player makes this error, not only will those four tiles be lost for scoring purposes, but—perhaps more critically—every square adjacent to these tiles is now also ruined for the purposes of placing a tile that may become part of a scoring island!
See the pair of light sandbars in Diagram 1. (Where sandbar is the term given to any orthogonally connected group of same-colored tiles that numbers less than four.) To connect the two sandbars with X or Y would create an illegal island of five tiles.
Also, to expand either sandbar to an island of its own would mean an island that is diagonally adjacent to same-coloured tiles, which is also illegal. See the pair of dark sandbars in Diagram 1. The player cannot play tiles at Z.
It is a harsh rule that an island may not be adjacent in any direction to any tile of the same colour. Still, the rule applies to one's opponent as well and yields a richness of counterplay and tactics that is very satisfying. Some other common poison-tile formations are depicted in Diagram 2.
The formations in the top left and bottom left are other simple examples of poisoned tiles that cannot become islands. The top right shows a formation for the light-coloured tiles that might have become an island, but is poisoned now that Dark has taken square j8. The bottom right shows a good island and a 1-tile sandbar that is trapped by that same-coloured island. The player does not have enough legal squares to build it into an island.
Edge Play
In the beginning of games, most placements tend to be around the centre. As the game progresses, play necessarily moves towards the edges of the board. Here lie some interesting problems that do not exist in the centre. In the cramped area of the edges—and most often, doubly cramped by the existing islands in the centre, in addition to the edge of the board—it can be easy to trap an opponent’s sandbar in such a way that it no longer has the space ever to become an island. See Diagram 3 for an example of this wonderfully treacherous business.
Edge Play
In the beginning of games, most placements tend to be around the centre. As the game progresses, play necessarily moves towards the edges of the board. Here lie some interesting problems that do not exist in the centre. In the cramped area of the edges—and most often, doubly cramped by the existing islands in the centre, in addition to the edge of the board—it can be easy to trap an opponent’s sandbar in such a way that it no longer has the space ever to become an island. See Diagram 3 for an example of this wonderfully treacherous business.
On the left of Diagram 3, Dark’s last move was c5 & d1, perhaps eyeing a future bridge across e2. On the right, Light's reply, c1 & c2, traps the sandbar on d1 such that it cannot now form any part of an island.
In assessing the viability of a sandbar, a very useful concept may be borrowed from the game Go. That is, the idea of liberties. Because your opponent may place two tiles on their turn, a sandbar generally needs three free squares to build out on in order not to be subject to becoming trapped. This simple concept can save a lot of time in practice.
The Cut
The scoring scheme of Ponte del Diavolo must, of course, be the primary driver of your strategy. The general rule is that the largest island group will win. Though, this is general, and can sometimes be overcome.
Let us take a simple example of a game that concludes with Player One having a three-island group and a two-island group, while Player Two has a four-island group and a one-island group. Player Two has won, of course: 6+3=9, which is less than 10+1=11. But, observe that both players formed five islands and connected them with, at a minimum, four bridges. (The "minimum" because an island group could include a connection that consists of two bridges across a sandbar.) If we take this lesson to heart, the crucial point to remember is that you will encounter situations where you must sacrifice your own next island or bridge in order to play a purely blocking move to stop your opponent from forming a winning group. This is easily said, but in the quest for more points it is frequently overlooked, even by experienced players.
This is truly the heart of this beautiful game. Best play requires that you are both building the larger group (or groups), while, at the same time, placing your tiles and bridges in such a way as to cut-off your opponent from creating the larger group (or groups). The depth of strategy that comes out of this dynamic tension is what I feel elevates this game to the realm of the classics.
Endgame trap
The race to place the right bridge first across key squares—that ideally will also cut-off one's opponent—is central. But, there is still another twist. This wrinkle may be unappreciated for many, many games. It usually reveals itself in the form of an unexpected loss in a game one was certain of winning.
The race to place tiles on key squares first must be contrasted with another important factor, the guaranteed squares. Formations often occur that mean that your opponent cannot place either a bridge or a tile on certain squares. See Diagram 4. When those squares are useful to your scoring, this is, obviously, a helpful thing. It means that you may wait on placing tiles on those guaranteed squares and continue to fight for more points—or to impede your opponent's score—on other contested squares.
In assessing the viability of a sandbar, a very useful concept may be borrowed from the game Go. That is, the idea of liberties. Because your opponent may place two tiles on their turn, a sandbar generally needs three free squares to build out on in order not to be subject to becoming trapped. This simple concept can save a lot of time in practice.
The Cut
The scoring scheme of Ponte del Diavolo must, of course, be the primary driver of your strategy. The general rule is that the largest island group will win. Though, this is general, and can sometimes be overcome.
Let us take a simple example of a game that concludes with Player One having a three-island group and a two-island group, while Player Two has a four-island group and a one-island group. Player Two has won, of course: 6+3=9, which is less than 10+1=11. But, observe that both players formed five islands and connected them with, at a minimum, four bridges. (The "minimum" because an island group could include a connection that consists of two bridges across a sandbar.) If we take this lesson to heart, the crucial point to remember is that you will encounter situations where you must sacrifice your own next island or bridge in order to play a purely blocking move to stop your opponent from forming a winning group. This is easily said, but in the quest for more points it is frequently overlooked, even by experienced players.
This is truly the heart of this beautiful game. Best play requires that you are both building the larger group (or groups), while, at the same time, placing your tiles and bridges in such a way as to cut-off your opponent from creating the larger group (or groups). The depth of strategy that comes out of this dynamic tension is what I feel elevates this game to the realm of the classics.
Endgame trap
The race to place the right bridge first across key squares—that ideally will also cut-off one's opponent—is central. But, there is still another twist. This wrinkle may be unappreciated for many, many games. It usually reveals itself in the form of an unexpected loss in a game one was certain of winning.
The race to place tiles on key squares first must be contrasted with another important factor, the guaranteed squares. Formations often occur that mean that your opponent cannot place either a bridge or a tile on certain squares. See Diagram 4. When those squares are useful to your scoring, this is, obviously, a helpful thing. It means that you may wait on placing tiles on those guaranteed squares and continue to fight for more points—or to impede your opponent's score—on other contested squares.
In the left of Diagram 4, it is Light’s move, but there is no urgent need to place a bridge over square h7. Dark cannot block it with a tile. Nor could Dark at any point create a way to put a bridge of their own over h7. To the right, therefore, Light can tend to the pressing matter of building the sandbar at j7-k7 into an island with, j8 & k8.
When a player is depending upon a large number of guaranteed, but, as yet, unrealized points, they can get an unpleasant surprise. I will remind you of Rule 4, above: "The game ends when either player can no longer legally place two tiles and chooses to pass." If the board is filling up, it's not uncommon for the end to come sooner than expected. Then those "guaranteed" points may not arrive in time to secure the victory! This has happened to me and, I expect, to most players who have played enough games of Ponte del Diavolo.
Interestingly, this rule has been the subject of some controversy. My opinion is that it is an excellent, balancing rule. The goal is to win; the margin is of no consequence. The serious player should know when the game will end and ensure they have their points scored before that time.
Final Thoughts
There are a couple of general attributes that attract me to an abstract. The first is a range of strategies. In Chess, for example, people often talk about a certain game as having a slow, positional character. In contrast, other Chess games are considered to be quite different and of a dynamic and tactical bent. In the same way, I believe Ponte del Diavolo gives the player a choice of strategic approaches. Is your style more to race towards a high score? Or, should you harass your opponent’s efforts and pursue a defensive, lower-scoring game where you nevertheless obtain the decisive few points to win? In a quick look at recent games on BoardGameArena, I have found completed games (not resignations) with a combined score of 61 (37 to 24) and another with a combined score of a mere 17 (11 to 6). Clearly, these were very different games!
When a player is depending upon a large number of guaranteed, but, as yet, unrealized points, they can get an unpleasant surprise. I will remind you of Rule 4, above: "The game ends when either player can no longer legally place two tiles and chooses to pass." If the board is filling up, it's not uncommon for the end to come sooner than expected. Then those "guaranteed" points may not arrive in time to secure the victory! This has happened to me and, I expect, to most players who have played enough games of Ponte del Diavolo.
Interestingly, this rule has been the subject of some controversy. My opinion is that it is an excellent, balancing rule. The goal is to win; the margin is of no consequence. The serious player should know when the game will end and ensure they have their points scored before that time.
Final Thoughts
There are a couple of general attributes that attract me to an abstract. The first is a range of strategies. In Chess, for example, people often talk about a certain game as having a slow, positional character. In contrast, other Chess games are considered to be quite different and of a dynamic and tactical bent. In the same way, I believe Ponte del Diavolo gives the player a choice of strategic approaches. Is your style more to race towards a high score? Or, should you harass your opponent’s efforts and pursue a defensive, lower-scoring game where you nevertheless obtain the decisive few points to win? In a quick look at recent games on BoardGameArena, I have found completed games (not resignations) with a combined score of 61 (37 to 24) and another with a combined score of a mere 17 (11 to 6). Clearly, these were very different games!
Diagram 5 shows a completed and razor-close game. Points are tied at 11. Tie-breaker one is tied at seven islands each. Tie-breaker two gives Dark the win with six bridges to Light's five.
Another attribute I like a lot in an abstract is a great, potential depth of player skill. I will compare Ponte Del Diavolo to Chess again (i.e., the western standard game very many people are familiar with). In Chess, it is well known that a player with even moderate experience will have a large advantage over a novice. Likewise, the intermediate player has little chance against an expert or master; and those players will rightly tremble if they are facing a grandmaster. The great classics, like Chess and Go, have this property that you can play and study for many years, always getting better—and, hopefully, always finding an opponent who can still challenge.
Now, I am certainly not in a position to claim that Ponte del Diavolo can equal the classics as regards the heights of player skill. In my view, this is a question that can only be answered after many more years of play, by many more players. The future will tell. Still, all the signs I have seen so far, leave me very optimistic about the future of this fascinating game. ◾️
Another attribute I like a lot in an abstract is a great, potential depth of player skill. I will compare Ponte Del Diavolo to Chess again (i.e., the western standard game very many people are familiar with). In Chess, it is well known that a player with even moderate experience will have a large advantage over a novice. Likewise, the intermediate player has little chance against an expert or master; and those players will rightly tremble if they are facing a grandmaster. The great classics, like Chess and Go, have this property that you can play and study for many years, always getting better—and, hopefully, always finding an opponent who can still challenge.
Now, I am certainly not in a position to claim that Ponte del Diavolo can equal the classics as regards the heights of player skill. In my view, this is a question that can only be answered after many more years of play, by many more players. The future will tell. Still, all the signs I have seen so far, leave me very optimistic about the future of this fascinating game. ◾️
Acknowledgement
The header image show the box cover of the original 2007 edition of Ponte del Diavolo from the BoardGameGeek page, taken by Carsten Wesel, April 2007.
The header image show the box cover of the original 2007 edition of Ponte del Diavolo from the BoardGameGeek page, taken by Carsten Wesel, April 2007.
K.C. is an IT architect and long-time Chess player. He enjoys modern abstracts Tzaar, Circle of Life, and Santorini, in addition to Ponte del Diavolo. His blog covers about games, and other subjects.
According to the BoardGameGeek entry for Ponte del Diavolo Martin Ebel dedicated his game to Alex Randolph (1922 - 2004), designer of Oh-Wah-Ree, Ricochet Robots, Inkognito, Code 777, and many other games, including of course Twixt. His game Knight Chase was included in Sid Sackson’s great A Gamut of Games. Alex Randolph’s most enduring game legacy will probably be Twixt, a game that is still widely played by abstract fans. The bridges of Ponte del Diavolo of course recall the Twixt connections. Alex Randolph can be seen standing on the bridge on the original box cover of Ponte del Diavolo shown in the header. ~ Ed.
According to the BoardGameGeek entry for Ponte del Diavolo Martin Ebel dedicated his game to Alex Randolph (1922 - 2004), designer of Oh-Wah-Ree, Ricochet Robots, Inkognito, Code 777, and many other games, including of course Twixt. His game Knight Chase was included in Sid Sackson’s great A Gamut of Games. Alex Randolph’s most enduring game legacy will probably be Twixt, a game that is still widely played by abstract fans. The bridges of Ponte del Diavolo of course recall the Twixt connections. Alex Randolph can be seen standing on the bridge on the original box cover of Ponte del Diavolo shown in the header. ~ Ed.