The following is an in-depth account of a typical game of Dvonn played between experienced players. It was played online on www.littlegolem.net, which is not quite either real time like BSW nor PBEM like Richard's server ( www.gamerz.net/pbmserv). Players log in, navigate to their games, view a graphic representation of the board, make their moves, and then split, or else hang around and hit refresh in the hopes that one of their opponents is doing the same. You can also set things up so that the server e-mails you to let you know when it is your turn in one of your games.*

What is particularly nice for those of us with a competitive side is that Little Golem also features a rating system. A new user starts his career with a rating of 1500; all of that user's games are then tracked in a database and as that player wins or loses his or her rating will go up or down. Beating a player with an equal rating will increase your rating 16 points and decrease your opponent's rating by 16 points; naturally, beating a player with a lower rating will earn you fewer than 16 rating points while beating a player who has a higher rating earns you more.

This game in particular was played against a gentleman from the Netherlands. My rating at the time was 1548, while his was 1657, so I was the underdog. The server assigned me white and Mr. Netherlands black.

You can read the entire progress of the match here using game notation. Picture the board situated lengthwise; the rows will be numbered 1-5 from bottom to top and the "columns" are lettered a-k, the columns slanting upward and to the left. So the move "a1-b2" would mean the jumping of a piece from the lower left corner of the board upward and to the right. If you have a Dvonn board, you can see the letters and numbers at the edges if you look carefully. If you don't have a copy of Dvonn, this report is probably going to be difficult to follow, unfortunately. We first placed the three red Dvonn pieces, and then our black and white pieces. The jumping and capturing begins at move 49.

white black
1. i4 2. c3
3. f2 4. g3
5. f1 6. e1
7. i5 8. j5
9. h3 10. j4
11. h4 12. i3
13. f3 14. e2
15. g2 16.g1
17. d1 18. h5
19. h1 20. c2
21. d3 22. d4
23. b2 24. b3
25. c4 26. d5
27. c1 28. b1
29. e5 30. f5
31. g5 32. e4
33. d2 34. f4
35. g4 36. a2
37. h2 38. i1
39. b4 40. a3
41. e3 42. k5
43. k4 44. j3
45. j2 46. k3
47. a1 48. i2
49. c5-d5 50. f5-e5
41. g5-f4 52. h5-h4
53. i5-h4 54. a2-b2
55. a1-b1 56. g1-f1
57. c1-c2 58. b3-c3
59. g4-g3 60 f1-f3
61. h4-e4 62. k5-k4?
63. d2-e2 64 c3-e3
65. g3-e3 66. d4-d3
67. d5-d3 68 e1-e2
69. d1-e2 70. i1-h1
71. j2-i2 72. h1-h3
73. g2-h3 74. e5-e3
75. b1-d3 76. j5-j4
77. i2-i4 78. k4-i4
79. h2-i3 80. k3-j3
81. d3-j3 82. resign

In the early setup phase there is always a kind of tension between various different objectives, at least for me. The first objective is, of course, to place rings adjacent to the red Dvonn pieces. As every player piece on the board must be connected through a chain of other pieces to a red ring, the closer your rings are to a Dvonn piece, the safer and more potentially useful those pieces will be. Moreover, you can also hop onto an adjacent red piece and move it off somewhere, though this is something for which you have to plan very carefully (more on that later). The second objective is to place pieces on the edge of the board, particularly the edges in the vicinity of the red rings. Rings that are completely surrounded cannot be moved, and so the more edge pieces you have, the more power you have. There are other little ploys, tricks and situation-specific concerns, but the other main piece of advice I have for the setup phase is that, in the absence of any other more important consideration, players should try to keep a consistent piece-density. Now, I don't say this categorically--it's not going to be true in every case--but big blobs of pieces are usually a bad thing. What it will often mean is that as the game plays out you will be forced to move a piece onto one of your own pieces, and every single capture of your opponent's pieces is precious. It also implies that you have a shortage of pieces in other areas of the board, a deficit on which an opponent can capitalize. Note Mr. Netherlands's density on the right side of the board; this is a good example of what not to do.

When the game starts in earnest, there are some new immediate concerns. First, it's a good idea to try and fix any situation where you might become isolated or be put in a position where one of your pieces' only moves are to to jump your own pieces. Thus my first capture; if black had gone d5-e5 first, c5 would have become pretty useless. Another good idea is to try to "dig out" a group of pieces in the interior of the board, which is exactly what black does with his first move. Yet another important priority is to set yourself up to jump on enemy pieces adjacent to the reds. See move #56: black's two-stack can now jump to either f3 or h3, both white pieces adjacent to a red. Moreover, this move also achieves the strategic goal of taking away my ability to move h1-g1, which would have distanced myself just a little more from the giant blob of black on the right.

For me, Dvonn games in general tend to hit a certain "strategic stage" somewhere between moves #55-#65; you've already fixed the worst flaws in your setup and have made the most obvious jumps, the board has started to take shape a little, and now you start to think about an overall game plan.** My general strategy in this game is fairly simple: ignore column h and everything to the right and concentrate on squelching black everywhere else, where I have superiority in numbers. I am outgunned on the right, and so if I move any of my pieces in that direction, I am only sending them to be captured. However, if he tries to develop things on that side he is either leading towards hopping on his own pieces (after he has taken k2 and j4) or else isolating his own pieces (by attacking column h with column i).

Black is having some strategic thoughts of his own, however, and they are expressed in move #58. Black jumps on the Dvonn piece with the obvious long-term intention of c3-e3-h3. Why? If your pieces aren't where the red pieces are, move the red pieces. This can be a devastating move, very much like yanking the rug out from beneath someone's feet, and so controlling the Dvonn pieces can be quite powerful. However, this maneuver cannot be undertaken lightly, because there is a price: you are missing an opportunity to capture, and your opponent gains a piece on you. Sounds harmless? Sure, it doesn't seem like a big deal when the board is full, but when it comes down to the endgame and every single move counts, you will at some point feel the pain of having one less piece to move than your opponent. Dvonn often boils down to who is able to maintain his mobility the longest, so if you are going to jump a red, you have to be able to make it count. Black does put a fair amount of pressure on me with his gambit, but ultimately not enough to gain an advantage.

So anyway, note my immediate reaction to move #58. There is a real sense of urgency on the board now, as we are both keen to set ourselves up to control one single spot, namely e3. If Mr. Netherlands can move the red here and subsequently come out on top, that red is jumping deep into the black side of the board. First he has three pieces threatening to my three; with move 59 I up it white 4 to black 3; he counters to bring it back to 3 to 3 (and that move also serves to "cover" c3 in case I decide to try to capture it). I change it to 3 to 2 and reduce his mobility in the area.

I really have the left side of the board covered pretty well at this point; most of his pieces aren't very effective in terms of being able to jump on my guys in a useful way--all except for b2, which can take d2 and reduce my ability to threaten the aforementioned e3 and generally make it difficult to dig my pieces out in the middle. However, Mr. Netherlands abruptly abandoned the situation to move k5-k4. This was a strange decision on his part, and I scoot d2 out of the way to really solidify my situation on the left. Note to neophytes: it is just as important to take possible captures away from your opponent as it is to capture his pieces.

With move #64 he finally does go c3-e3, but I'm ready for him. After a quick squabble (moves 65-69) he ends up able to take control of e3, but it will be too tall for him to move anywhere. I, on the other hand, have made up for it by securing many other stacks.

The game shifts gears at this point: the left side is more or less wrapped up, so now black is going to try to form his mass of rings on the right into stacks that can leap across the gap and land on my stacks in the middle of the board. The outcome of the game is crystallizing now, and there is a series of "if I move here and he moves there and I move here and he moves there..." calculations performed on every turn. It is entirely typical that somewhere around moves #70-#75 you reach a kind of critical mass where making the right move becomes absolutely crucial. Often when you look back on the game you can see one single move and say "I lost the game there," and that move almost always takes place in the low 70's. In this particular game the critical nature of the situation drags out to move #81--any misstep and I will lose the game.

The stacks at h3 and i3 become the front lines of the war, as these are the two most likely launch pads for attacks on my holdings (any attempt to perpetrate something along row 2 can be foiled). Notice that I keep h2 free and i3 pocketed for as long as I can. I do not want to lose the four-stack on h3, nor do I want f3 to jump on i3 and then move that stack to the left. When I finally open it up with move #77, I only do so because I can now jerk the red away if he doesn't go k4-i4, and I am then able to h2-i3 (he also could have gone i3-i4, but that solves the same problem of keeping i3 off of h3). The game is pretty much over at this point.


If all this sounds like I am trying to pass myself off as an expert Dvonn player, please believe me when I tell you that this is not the case. At the moment my rating is 1640, while the highest Little-Golem-ranked player, a "thor" from Germany, has a rating of 2023. I haven't played that particular person yet, but I have played #4, a Jose M. Grau Ribas from Spain with a rating of 1889. The last game I played with him he beat me so badly I could have given up after only seven or eight captures, believe it or not (he yanked the rug on me). I've also had the privilege of playing against Kris Burm, the designer of the Gipf series. He beat me, of course, though we did have a brief but pleasant chat about Dvonn and the series in general.

I hope all this has been somehow interesting or useful to those who are new to Dvonn and are trying to decide whether it is worth putting in the effort to explore the game. It is.



* The only bad thing about the Little Golem system is the amount of time a game can take if you're playing with someone who only logs in only once every few days. Each player starts with 240 hours on their clock and each move adds 36 hours to the clock. So, for a game of Dvonn, which typically lasts something like 85 moves, a slowpoke can drag a game out for 73 days--that's two and a half months for a game that you can play in twenty minutes face to face (and that's not even counting the ten days of "vacation" that each player gets per year).

** As I've played more and more I've started thinking "seriously" about what's happening on the board earlier and earlier. When playing against much weaker players I will start thinking strategy even during the setup, as neophytes will often set themselves up to have a big chunk of pieces pocketed or isolated. When playing against people who at your level the game often tends to be more of a war of attrition.