Connect any three sides, any two corners, or make a loop.
Generated at 07/09/2020, 05:20 from 5189 logged games.
Representative game (in the sense of being of mean length). Wherever you see the 'representative game' referred to in later sections, this is it!
Each turn, place a stone of your colour in an empty space.
You win if you complete one of the following patterns:
General comments:
Play: Combinatorial
Family: Connection,Spiel des Jahres
Mechanism(s): Connection,Pattern,Strict Placement
Components: Board
| BGG Entry | Havannah |
|---|---|
| BGG Rating | 7.26223 |
| #Voters | 184 |
| SD | 1.52152 |
| BGG Weight | 3.2778 |
| #Voters | 18 |
| Year | 1981 |
| User | Rating | Comment |
|---|---|---|
| ChiDrago | 10 | |
| mrraow | 8 | Feels like a cross between go and hex. |
| Mjoellnir | 6 | Used in Essen 2007? |
| arian | 8 | |
| longshots | 7 | I play online... it's fun, you think you have the upper hand, only to find out you are fooling yourself... |
| Arctic Jack | 6 | A very good abstract but it can't compete for my attention with Chess and Go. This will get played and enjoyed occasionally |
| bayspiel | N/A | 06.02.04.01 |
| zublord | 8 | |
| okada | 7.5 | |
| fogus | 7.05 | [2015.06.01] This game no longer subsumes Hex and Y for me. I've grown to appreciate their elegance while at the same time souring (just a little tiny bit) on the racing-feel of Havannah. It's still a great game though. [2014.10.10] - 8.75 Subsumes Hex, Y, and Six for me. It's a brain-burner, but in that exquisite way. I look forward to playing 100 more games ASAP! |
| cannoneer | 8 | I've never been a fan of connection games. After playing Havannah, I think I understand what about it makes it better that other similar games. It's been mentioned but bears repeating - one player's winning condition does not stop the other player from winning quicker. So you aren't always forced into a tit-for-tat game of response - sometimes you can let the other side build a bit, if you think you can make a quicker-winning setup with your pieces. I really like this aspect of the game; it seems to make it slightly more strategic in nature, something I feel other connection games really lack, being entirely tactical. I play on IG Game Center. |
| ervandarnell | 7 | Improvement on Nash (Hex). The loop rule and fork rule combine to force players to both cut and connect. |
| dolzandavid | 5.5 | Very similar to Hex, with a couple of added twists. It somehow failed to grab me, as did Hex for that matter. OK-ish, but nothing special. |
| MFinch | N/A | To Play. |
| freechinanow | 7 | Havannah is a very nice take on the whole Hex "connection game" idea. Why have one objective when you can have 3? I really enjoy being able to work on multiple goals or at least fake out my opponent into thinking that I'm working on one thing while completing another. So, obviously not as pure as Hex but super intriguing and a solid variation! |
| Stephen Glenn | 10 | |
| games | N/A | f |
| emike | 9 | Connection game more complex than Hex and more fun than Twixt. |
| AntGamer | 8 | |
| crystalpunk | N/A | By Christiaan Freeling of http://mindsports.nl/index-mindsports.html |
| rri1 | 9 | Very connection game with 3 different objects which are well balanced. |
| demuur | 7 | |
| Tolkana | 9 | |
| hell on wheels | 8 | |
| boardmathias | 8 | It's interesting but I need to play it more to provide some judgement. I am slow in understanding connection games. |
| osyed | 10 | This rating is based on about 50 postal games played since 2003 on www.mindsports.nl. This game should be ranked up there with Go because Havannah is the equivalent of Go for connection games. Simple, scalable, fun to play, computer resistant and its very deep. I congratulate Christian Freeling on creating such a classic game. |
| gunnaronthesaw | 7 | Really challenging abstract - I think it is deeper than Twixt and definitely more interesting than Hex |
| mobo63 | 7 | Print & Play |
| 2d20 | N/A | PnP |
| MrEdCollins | 9 | |
| figge | 7 | |
| nycavri | 5 | I'm just terrible at connection games . . . |
| Valenox | 7 | См. коммент к Unlur. |
| Schachtelhalm | 8 | Someone described Hex as a knife fight in a phone booth. If you have claustrophobia you should try this Hex variant. I would like to buy a wooden game set. |
| simpledeep | 8 | |
| schwarzspecht | 8 | |
| mhowe | 10 | |
| luribe | 8 | A connection game that's both relaxing and challenging. Like Hex but on an hexagonal playing board. |
| Michael Becker | 7 | I enjoy abstracts and this is a good one. Three ways to win the game and a very good strategy guide are included in the rules. I would definitely study this game more if I didn't own so many games already... The box cover art is yucky though - ugly, ugly, ugly... |
| Dingus | N/A | Use Ingenious, Wabanti, or Celtis board |
| Windu47 | N/A | 1 |
| bcnevan | 7.5 | A strong twist on connection games that substitutes the players' march of semi-forced call and responses that is found in other more traditional connection games with victory condition races. The tensions change in interesting ways. |
| The Player of Games | 7 | Print and play board. Stones from [thing=11997] [/thing]. Great connection game. However, I like [thing=4112] [/thing] more due to its simplicity of rules. Anyhow, Havannah has its own flavor and is therefore a nice supplement. |
| _mrq_ | N/A | printed board |
| seneca29 | 8 | I played a couple games and I liked it a lot! I NEED TO PLAY THIS A LOT MORE! I barely scratched the surface of this game... -homemade |
| ankeszu | 9 | |
| unic | 8 | Very different connection game - as one player winning does not block the other, defence is no longer equal to offence, and it sooner or later turns into a race game - who can finish their connection first? |
| LosSchabossDragon | 10 | |
| menes1964 | 7 | Good strategy game. Easy and fun. |
| Mingy Jongo | 8 | |
| wizhyun | 7 | |
| byturn | 6 | DIY. played using magnetic baduk pieces on a Gemblo board. 2019.12 - purchased 80 pieces of 10mm wooden discs of black and white from piececraft shop. |
| CorinnaS | N/A | used, Essen |
| ceenan | 8 | |
| Magny2 | 7 | |
| lulu35 | N/A | 2j abstract (similaire kamon) |
| struggler1 | 10 | |
| whac3 | N/A | need print board & rules |
| Zapawa | 7 | This was Christian's first game, and it shows -- while it's a formidable sports weapon, it lacks his usual deliberation and carefulness. Compared to Hex, it's hard for me to say if the gain of depth is big enough to compensate the loss of elegance. |
| dancingdanslc | 2 | 2 Player Only: Android Play No AI Similiar to hex with a hexagon board. Connect 3 sides to win. So many connection games....why bother with this one? Out of my collection. |
| MobyNostromo | 8 | Great connection game. Nice twist on Hex. |
| Jedrique | 9 | |
| jgravitt | 7 | Well, I play it on the hex board that came with Ingenious. We like the multiple victory conditions, it's easy to teach, and quick. |
| dave doma | 8 | |
| Talisinbear | 6 | printed off a board, will decopage it onto a piece of board and give it a whirl with some glass beads as pieces. A few plays, way better than Hex, but still not exactly my thing. |
| Playoria | 8 | Deeper and better version of Hex. |
| Thorphas | N/A | 2 30min - abstrakt - Reihe |
| ungeschrieben | 8 | |
| chenke | N/A | - Empfehlungsliste Spiel des Jahres 1981 |
| Jugular | 8 | |
| marcoms | 8 | |
| Takvorian | 8 | |
| ChristianDK | 7.5 | Great connection game. |
| halladba | 9 | |
| xulu | 7 | |
| andy61 | 6 | Quick, enjoyable abstract, easy to play with non-gamers, but enough challenge to be interesting. |
| steadym | 7 | |
| kokatron | N/A | zv 10 |
| RDReilly | 9 | I'm surprised more fans of abstract games don't play this. I made my own wooden board. Fascinating game! |
| morphles | 10 | Though more complicated rules than hex. End result justifies it, by having even richer and more intricate gameplay. Probably my favorite game ever. |
| kentawt | 7 | |
| dw5k | N/A | DIY/Computer |
| twixter | 8 | I have played this game by email. It's very deep and subtle, especially on the larger size 10 board. There is a "race" aspect to this game I am still trying to get my head around. It is often very difficult to see who is ahead in the race. |
| Tony van der Valk | 9 | Three times champion on littlegolem.net on base-10 |
| i_am_ignis | 10 | |
| latindog | N/A | Currently Unplayed :( |
| FauxSloMo | 10 | |
| germanggek | 9 | Brainburner. Something like GO. |
| megamau | 6.9 | Fun and heavy in strategy. However I love Hex mainly for its simplicity, which doesn't hinder the deepness. Havannah tries to improve on Hex and in my opinion achieves the opposite effect. For a better "children" of Hex look at *Star. |
| Pensator | 9 | |
| Butsudoka | 9 | I haven't played this game much yet, but I'm already deeply impressed. Basically, it's a blend of Hex and Trax; works like clockwork. Like all good abstract games, it has simple rules, but much depth. I assume that I eventually will give this game the highest grade. A criminally underrated and forgotten game, which deserves more attention and recognition. |
| molnar | 7.5 | Another interesting Hex variant. Not as compelling as Y. Make your own board, or play on paper. |
| Rijssiej | 5.5 | Fun: 4 Quality: 7 -much influence (+1). |
| Kpato | 7.8 | |
| rayzg | 9 | Favorite connection game. The best part is the multiple winning conditions. |
| ImogenisPuzzle | 9 | |
| Tventano | 6 | The board is just a little to small for the official rules of Havannah. For a casual game this is good enough. He made many more chess like games, which have not been commercially published. |
| aarondesk | 8.2 | A hex variant with enough variant to make it more interesting. #to-play |
| S.K. Lator | 8.5 | Feels like a cousin of Hex and Game of Y. If you don't mind the ugly colors of the Ravensburger version, here's an unexpensive, top-notch connection game, that gets played here more often than Twixt or Pünct. |
| orangeblood | 8 | Wonderful design from Christian Freeling. More fun to me than Hex. |
| King Lear | 7 | let's play Havannah at igGamesCenter. My username is ror1991 |
| clayhaus | 9 | After a couple of plays I can see why this is considered a classic by some: lots of variability and playability |
| Arkeo | 8 | |
| Reinhard Sabel | 9 | For me an all-time abstract classic. The tree different ways of winning are a special - and very good - quality of this game |
| pled | 7.5 | |
| Schmelzi | N/A | 8-2 |
| kees | 8 | |
| RichardIngram | 9 | |
| Colonnello Vincent | 9 | One of the best abstract games I've ever played. |
| metzgerism | 9 | |
| madjer | 10 | |
| TurboGeist | N/A | Category: Abstract Strategy Mechanic: Pattern Building Family: Connection Games $ - ? |
| gmoralesor | 8.6 | tablero de celtis |
| crimescene | 8 | A few months ago a friend had the luck that he could buy 2 big boxes with games from the 1970s to the end of the 1980s for little money at a flea market. He didn't know most of the games, so he asks me if I can help him. It's a good thing I've been living a little longer. I knew many games from my youth. We both had our fun playing the old games. EDITION: Havannah -Ravensburger He had the game as a kid. It's for 2. The game reminded me of GO. It's a strategic game. A draw is almost impossible. It was a lot of fun, when I find it, it comes into my collection |
| T0afer | 9 | |
| DanielBeaver | 8 | A neat connection game with intuitive rules. Blocking is more difficult than other connection games, so it is more of a race. I've found that new players like this better than Hex. |
| bthermans | 6.5 | Nice simple strategic game. |
| schmozart | 10 | Perfectly playable with Omega (nestorgames). A real hex battle. |
| dispatch134711 | 8.2 | A connection game with three different win conditions. One of Christian Freeling's best contributions (I feel), in that he put effort into promoting the game via his competition with programmed AI. Another reason I really need to get a hex-hex board. |
| aaronseeber | 8 | Probably my favorite connection game. Deserves to be better known than it is. |
| alekerickson | 10 | |
| CDRodeffer | 7 | Interesting connection / formation game in the same family as Hex, Mudcrack Y, Y, Poly-Y, Unlur, Star and *Star. |
| russ | 8 | A modern classic abstract connection game I'd heard about for a while but hadn't tried until the "mix and match" online abstract tournament. The 3 different ways to win add a nice race element so that it's not enough to secure an unbreakable connection (e.g. as in Hex or Twixt), but you must do it [i]faster[/i] than the opponent. The 3 conditions feel a bit arbitrary and there's wonky details about how the corners don't count as sides (as they normally would in a connection game) - but it's a case of "it works well in practice even if it's slightly inelegant". |
| PfrBraun | 6 | Havannah (Christian Freeling) – oder: Irgendeinen Namen braucht jedes Spiel… Vielleicht war Freeling ein Anhänger Fidel Castros. Der hat ja Havanna im Sturm genommen. Vielleicht war Freeling auch Kabbalist. ‚H‘ steht im Hebräischen für ‚5‘ und die ‚5‘ verweist auf den Pentateuch. Vielleicht verweist Freeling damit auf die Verbindungen zwischen Kibbuz und Kommunismus? Vielleicht hat er aber auch nur irgendwo aufgeschrieben, dass Havan nah ist. Am nächsten kommt wohl noch Theorie eins. Havannah ist nämlich ein abstraktes Rennspiel. Man muss durch Legen einzelner Steine entweder einen Kreis bauen, oder zwei oder drei Spielplanseiten miteinander verbinden. Freeling liefert hier also ein abstraktes Verbindungsspiel ab. 1981 war es für das SdJ nominiert. Gar nicht mal zu Unrecht. Die taktische Legespielerei weiß zu gefallen. Das psychedelische Farbdesign mit Knallorangefarbenen Steinen auf Khakibrett macht Laune. Danach legt man Blumen und Tiere. Nur das Gesicht von Che Guevara kann man nicht legen. Aber zurück zum Rennspiel: Es ist nichts weniger als nett. Aber was macht man mit nett? Kann man mal spielen? Sicher! Wird es vorkommen? Eher nein! Klingt das teilnahmslos? Eher ja! Warum? Noch ein bisschen Manifest: Mal spielen ist nicht mehr genug. Entweder: Ja, das reizen wir aus. Oder: Ja, zweite Chance. Sonst: Havanna uh la la. Half of my heart is in Havanna. Die andere Hälfte ist Havan nicht nah, sondern fern. |
| hyperpape | 8 | |
| moxtaveto | 8 | |
| mthomashow | 8 | near-classic connection game, worthy of analysis, and interesting to play |
| Cassiel | 8 | |
| zeder | 8 | |
| montsegur | 8 | Excellent abstract game |
| Spaten | 6 | One of my favorite abstracts. Besides blocking, it can take on the characteristics of a race toward competing goals. For me, just more fun than Hex. |
| davidjw | N/A | [33] |
| SpieleHolger | 7 | |
| strangeyeah | 10 | |
| mickwood | N/A | 2 players. Play on an Ingenious board. |
| Kaffedrake | 5 | Similar to Hex but achieves hexagonal symmetry, which appeals to me. However, the fact that the two players do not globally have mutually exclusive connection goals, but are racing to complete their chains, means Havannah seems slightly less elegant. |
| crazygames | 7.5 | Very good abstract game, deeper than the first plays might make you think. Simple rules, the plastic components and the box art are not too attractive. |
| boomtron | 8 | One of my favorite connection games. So much game comes out of simply taking turns putting pieces on the board, and interesting victory conditions. |
| GeorgiaHikerGirl | N/A | Use a HEX board. |
| escueladejuegos | N/A | Tablet android. |
| dbucak | 9 | |
| janus | 8 | |
| minismurf | 9 | |
| steveoliverc | N/A | Hex with mores sides and winning conditions. Print & play using Go or glass stones at http://www.mindsports.net/Arena/Havannah/Rules.html |
| Size (bytes) | 34350 |
|---|---|
| Reference Size | 10293 |
| Ratio | 3.34 |
Ai Ai calculates the size of the implementation, and compares it to the Ai Ai implementation of the simplest possible game (which just fills the board). Note that this estimate may include some graphics and heuristics code as well as the game logic. See the wikipedia entry for more details.
| Playouts per second | 67433.16 (14.83µs/playout) |
|---|---|
| Reference Size | 510907.88 (1.96µs/playout) |
| Ratio (low is good) | 7.58 |
Tavener complexity: the heat generated by playing every possible instance of a game with a perfectly efficient programme. Since this is not possible to calculate, Ai Ai calculates the number of random playouts per second and compares it to the fastest non-trivial Ai Ai game (Connect 4). This ratio gives a practical indication of how complex the game is. Combine this with the computational state space, and you can get an idea of how strong the default (MCTS-based) AI will be.
| Label | Its/s | SD | Nodes/s | SD | Game length | SD |
|---|---|---|---|---|---|---|
| Random playout | 85,091 | 4,017 | 10,251,574 | 484,588 | 120 | 19 |
| search.UCB | 52,038 | 4,081 | 40 | 11 | ||
| search.UCT | 51,079 | 3,765 | 43 | 11 |
Random: 10 second warmup for the hotspot compiler. 100 trials of 1000ms each.
Other: 100 playouts, means calculated over the first 5 moves only to avoid distortion due to speedup at end of game.
Rotation (Half turn) lost each game as expected.
Reflection (X axis) lost each game as expected.
Reflection (Y axis) lost each game as expected.
Copy last move lost each game as expected.
Mirroring strategies attempt to copy the previous move. On first move, they will attempt to play in the centre. If neither of these are possible, they will pick a random move. Each entry represents a different form of copying; direct copy, reflection in either the X or Y axis, half-turn rotation.
| 1: White win % | 51.28±1.36 | Includes draws = 50% |
|---|---|---|
| 2: Black win % | 48.72±1.36 | Includes draws = 50% |
| Draw % | 0.00 | Percentage of games where all players draw. |
| Decisive % | 100.00 | Percentage of games with a single winner. |
| Samples | 5189 | Quantity of logged games played |
Note: that win/loss statistics may vary depending on thinking time (horizon effect, etc.), bad heuristics, bugs, and other factors, so should be taken with a pinch of salt. (Given perfect play, any game of pure skill will always end in the same result.)
Note: Ai Ai differentiates between states where all players draw or win or lose; this is mostly to support cooperative games.
| AI | Strong Wins | Draws | Strong Losses | #Games | Strong Win% | p1 Win% | Game Length |
|---|---|---|---|---|---|---|---|
| Random | |||||||
| Grand Unified UCT(U1-T,rSel=s, secs=0.01) | 36 | 0 | 0 | 36 | 100.00 | 52.78 | 65.31 |
| Grand Unified UCT(U1-T,rSel=s, secs=0.03) | 36 | 0 | 11 | 47 | 76.60 | 42.55 | 47.79 |
| Grand Unified UCT(U1-T,rSel=s, secs=0.07) | 36 | 0 | 6 | 42 | 85.71 | 38.10 | 40.00 |
| Grand Unified UCT(U1-T,rSel=s, secs=0.20) | 36 | 0 | 6 | 42 | 85.71 | 40.48 | 45.31 |
| Grand Unified UCT(U1-T,rSel=s, secs=0.55) | 36 | 0 | 12 | 48 | 75.00 | 64.58 | 52.10 |
Level of Play: Strong beats Weak 60% of the time (lower bound with 90% confidence).
Draw%, p1 win% and game length may give some indication of trends as AI strength increases; but be aware that the AI can introduce bias due to horizon effects, poor heuristics, etc.
| Game length | 54.53 | |
|---|---|---|
| Branching factor | 142.47 | |
| Complexity | 10^116.77 | Based on game length and branching factor |
| Samples | 5189 | Quantity of logged games played |
Computational complexity (where present) is an estimate of the game tree reachable through actual play. For each game in turn, Ai Ai marks the positions reached in a hashtable, then counts the number of new moves added to the table. Once all moves are applied, it treats this sequence as a geometric progression and calculates the sum as n-> infinity.
| Distinct actions | 170 | Number of distinct moves (e.g. "e4") regardless of position in game tree |
|---|---|---|
| Good moves | 92 | A good move is selected by the AI more than the average |
| Bad moves | 78 | A bad move is selected by the AI less than the average |
| Response distance | 4.36 | Mean distance between move and response; a low value relative to the board size may mean a game is tactical rather than strategic. |
| Samples | 5189 | Quantity of logged games played |
A mean of 32.13% of board locations were used per game.
Colour and size show the frequency of visits.
Game length frequencies.
| Mean | 54.53 |
|---|---|
| Mode | [59, 61] |
| Median | 54.0 |
This chart is based on a single representative* playout, and gives a feel for the change in material over the course of a game. (* Representative in the sense that it is close to the mean length.)
Table: branching factor per turn, based on a single representative* game. (* Representative in the sense that it is close to the mean game length.)
This chart is based on a single representative* game, and gives a feel for the types of moves available throughout that game. (* Representative in the sense that it is close to the mean game length.)
Red: removal, Black: move, Blue: Add, Grey: pass, Purple: swap sides, Brown: other.
Colour shows the frequency of swaps on turn 2 if this move is played on turn 1; black < red < yellow < white.
Based on 100 trials/move at 0.1s thinking time each.
| Moves | Animation |
|---|---|
| h1,h15,g15,f15,e15,i1 |  |
| h1,h15,g15,f15,j1,d15 |  |
| h1,h15,j1,f15,g15,d15 |  |
| h1,h15,g15,f15,e15 |  |
| h1,h15,g15,f15,j1 |  |
| h1,h15,j1,f15,g15 |  |
| h1,h15,g15,f15 |  |
| h1,h15,i1,g15 |  |
| h1,k1,j1,i1 |  |
| i1,h15,h1,g15 |  |
| h1,h15,j1,f15 |  |
| k1,h1,g15,f15 |  |
Colour shows the success ratio of this play over the first 10moves; black < red < yellow < white.
Size shows the frequency this move is played.
Colour shows the frequency of swaps on turn 2 if this move is played on turn 1; black < red < yellow < white.
Size shows the frequency this move is played.
| 0 | 1 | 2 | 3 |
|---|---|---|---|
| 1 | 169 | 28730 | 2427854 |
Note: most games do not take board rotation and reflection into consideration.
Multi-part turns could be treated as the same or different depth depending on the implementation.
Counts to depth N include all moves reachable at lower depths.
Inaccuracies may also exist due to hash collisions, but Ai Ai uses 64-bit hashes so these will be a very small fraction of a percentage point.
No solutions found to depth 3.