Last week I wrote about Broken Picture Telephone, a site on which I wasted far too much time in June and July when I should have been getting work done. I thought I'd follow that up with an article on Kdice, a site on which I wasted far too much time in August when I should have been getting work done. (As opposed to wasting it writing Calendar articles.)

I gave up on Broken Picture Telephone because the site was troll-ridden and buggy even before it broke completely, and its absentee creator and sole maintainer seems not to care. I gave up on Kdice because even though the site seems like a solid piece of programming, I eventually had to conclude that the game itself is fundamentally flawed. But before I get to my objections, let me first describe the game and the aspects of it that drew me to it in the first place.

Kdice is Ryan Dewsbury's multiplayer adaptation of Taro Ito's game Dicewars. The site generates a map, as in the picture above. Territories are randomly distributed among seven players, and each player's dice are randomly distributed among his or her territories. The goal is to control every territory on the map. This is achieved thusly: click the territory from which you wish to launch an attack, then click your target. The dice in your territory and the dice in the target territory are rolled; the highest total wins, with ties going to the defender. Obviously, the more dice you have, the higher your total is likely to be, but this isn't always the case; it's not uncommon for five dice to successfully defend against an attack from eight, and technically, one die can fend off an attack from a stack of six. (It's a 1-in-279936 chance, but if the defender rolls a 6 and the attacker rolls six 1s, the defender wins.) If you win, the enemy dice in that territory are removed, and the territory is turned over to you; your dice automatically move into the targeted space, with one left behind to hold down the fort. If you lose, your attacking stack is knocked down to one die, and the enemy stack is unaffected.

At the end of your turn, the computer counts up your greatest number of contiguous territories and adds that many dice to your forces. This is one of the cleverest rules in Kdice (and in Dicewars before it), as it gives you a clear goal early on: connect up your territories! And in so doing, you will of course trample over some of your opponents and be trampled upon in turn, giving you grudges that will shape your strategy for the rest of the game. Note, however, that the dice you receive are added to your territories randomly. You have to cross your fingers and hope that most of them happen to wind up in a big stack next to an opponent you plan to attack.

Kdice's sister site is Gpokr, a fake-money Texas hold-'em game, and the mix of luck and skill involved in the two games is similar. Yes, the same people populate the leaderboard on Kdice month after month, just as there are people who can compile a high enough win percentage in poker to make an income at it. But the outcome of any given game is largely a matter of chance. Let me put it this way. As a go player, I am 5 kyu online (closer to 10 kyu over the table). This means that when I play other 5 kyus, I have about a 50/50 chance of winning. (I'm a weaker 5 kyu, so maybe I win more like two games in five.) Against a 4 kyu, my chances are much slimmer — maybe I win one game in ten. Against a 3 kyu, my guess would be one game in a hundred. Against a stronger player than that I have basically no chance, short of my opponent suffering a midgame stroke or something. Now, consider poker. I've never set foot inside a casino, have never played a real game of poker in my life. I've watched it on Youtube, and I've played a few hands on Gpokr but found it boring. Nevertheless, in any given hand, I have a very good chance of being able to beat the top player in the world. Sure, over a significant span of time he's going to beat me. But in one hand, if I get A-J, he gets 7-4 offsuit, and the flop comes down A-J-2 rainbow, I've got a 99.4% chance of victory. The way he shows his skill is to fold quickly (and with a 7-4 he almost certainly already has). Much the same is true with Kdice. Your initial position is a matter of luck. In one game you might start off with four contiguous territories and the first move; in another you might start off with four disconnected 3-stacks, each of which is next to the 4-stack of someone who moves before you. As in poker, the trick to sustained success is maximizing the value of your lucky breaks and minimizing the damage from your bad ones.

Kdice, you see, is not winner-take-all, which is the other thing I like about it. The point distribution on the 100-level tables, for instance, is as follows: 1st place, +175 points; 2nd, +83; 3rd, +38; 4th, -8; 5th, -45; 6th, -72; 7th, -100. So you might well look at your initial position and discover that you have no real shot at ever taking over the entire map... but that's no reason to give up, because you can still try to eke out 4th. The central strategical question Kdice players have to confront is this: how ambitious should I be? Say two players have distinguished themselves from the rest, with you in a distant third place, three others reduced to one or two territories, and one already eliminated. One of the juggernauts, fighting with the other, has left his border with you sparsely defended. Do you attack, trying to grab first or second place at his expense, and risk having him surrender to his main opponent in order to win the freedom to knock you down to sixth in retribution? Or do you officially declare that you're happy with third? And if you do, will the superpowers respect your white flag and allow you to maintain the territory you've secured?

The answer to that last question is "usually, yes." For while the third major element of Kdice, on top of luck and tactics, is diplomacy, conducted via a chat window in the bottom right corner of the screen, Kdice isn't Diplomacy. The game Diplomacy contains no element of chance; the primary way you gain an advantage over your opponents is to gain the trust of one of your neighbors and then choose the perfect moment to stab your supposed ally in the back. It's a vicious game, and bad feelings run high. The last game I played in real life, back in college, ended with Turkey in tears and France ending the game by flushing the German pieces down the toilet. The last game I played online imploded after Austria quit in a huff after Fall 1901 and France quit in a huff after Spring 1903. In Kdice, by contrast, breaking your word is generally considered a major breach of protocol. Sure, it happens — on a regular basis, even — but even though the rules say that respecting surrender flags is entirely optional, players tend not to accept being attacked while flagged as part of the game. They complain about it vociferously and litter each other's profile pages with reviews like, "Asshole! Doesn't respect flags! KILL ON SIGHT!" The only practice that attracts more scorn is the formation of pre-game alliances.

Aside from this difference, though, Kdice is quite similar to Diplomacy, and that's not a compliment. I've told the following story before, but it's one of the most memorable things I've ever observed, so here it is again: in my ninth-grade history class, the teacher had us play a game that was supposed to demonstrate how shifting alliances work. He divided the class into seven groups — dubbed Britain, France, Germany, Belgium, Italy, Austria and Russia — and, every few minutes, declared a "battle" between two of the countries. Then there was a negotiation period, during which we all were supposed to walk around the room making deals. Whichever warring country collected the most allies would win the battle and a certain number of points to divvy up with its allies. The idea, I think, was that countries in a battle would try to win over the wavering countries by promising them extra points to jump aboard. But that's not how it worked in practice. Three or four guys decided among themselves during the first negotiation period what the outcome would be, and told people whom to vote for. And the others just shrugged and did as they were told. These guys had decided that Germany would win, followed by France, Britain, Belgium, Austria, Italy and Russia. The first battle was France vs. Russia. Germany and Britain both signed up on the French side. Austria and Italy, realizing that if they just went along with the plan they'd come in 5th and 6th, joined up with Russia. That left it up to Belgium. I was on team Belgium. I voted to give our vote to the Russian side, because that way at least we weren't doomed to come in 4th. And no one else on my team went along. They meekly gave their points to the French side. After that, there was no contest. Britain vs. Austria? 6-1, Britain. Germany vs. Belgium? 6-1, Germany. (And we could have beaten them if we'd just formed a bloc with the other three losers!) The teacher noticed that Germany and France were always on the same side and declared Germany vs. France. Outcome: 6-1, Germany. These three or four guys were able to just impose their will on a class of 40 students. No carrots, no sticks, just "here's what will happen" and everyone else nodding.

In Diplomacy the same sort of thing tends to happen, at least in the games I've played. There always seems to be one player whom the others passively obey. In 2001 I served as the moderator for an online game, and was astonished to watch everyone just roll over for Germany. By the Spring 1903 turn Germany was actually writing Italy's orders! In Fall 1903 Germany commanded France to commit suicide by moving his units out of his supply centers, and he did! It was ninth grade all over again, with people just falling in line. After the game I asked France what he had been thinking. "I knew I wasn't going to win, so I just wanted to have an effect on the game," he explained. By helping the player who just treacherously attacked you defeat the player trying to come to your aid? "I wanted to have an effect on the game." Uh-huh. Researchers such as Michael Gazzaniga have demonstrated that the brain contains elaborate mechanisms to rationalize the things we do. Show a split-brain patient's right eye the word "laugh," and she'll do so; ask her why she's laughing, and she'll smoothly come up with a spurious reason. "Oh, it's funny that people make a living doing experiments like this." I have to think France's explanation falls into the same category. Why did France commit suicide on Germany's say-so? Hard-wired primate dominance behavior strikes me as the much more likely answer. You know, the same sort of thing that makes armies function. Marching off to your death on a commanding officer's say-so is a big part of a soldier's job description.

The same sort of thing happens in Kdice. It manifests in two ways. One is identical to how it works in Diplomacy: someone starts barking orders and the others inexplicably follow them. But in addition to the informal dominance hierarchy that seems to form whenever you put a bunch of animals together, Kdice also supplies a formal hierarchy with its 1st-through-7th in-game standings. And Kdice culture demands fealty to the hierarchy. There are various customs that make up the generally accepted rules of fair play in Kdice — don't attack people who have flagged to you, never attack people to whom you have flagged, let lower-ranked players settle disputes without interference, etc. But the number one rule is this: fair is whatever #1 says it is. Say you're playing red. Blue owns about half the map and is in first place, while green and brown are fighting on the other side of the map from you, and the other three players have all been eliminated. In theory, whichever player wins the green/brown war should take second, while you automatically get third as the third surviving player. But say the war begins to turn green's way. Brown isn't likely to fight to the bitter end; he'll undoubtedly say, "Okay, I surrender to green and flag third." What's your proper response? "No fair! That was clearly a battle for 2nd vs. 4th!"? "No fair! At least let me fight brown for 3rd!"? Certainly you'll hear both of these if you play for any length of time. But in the culture of Kdice, the proper response is the following: "Blue won the board, so it's blue's decision." Deference to hierarchy is the supreme virtue. This is one of the things I dislike about Kdice.

Another is the game's propensity to produce stalemates of various sorts, another fault it shares with Diplomacy (whose map is notoriously riddled with stalemate lines). Consider the map above. Teal is in first place, purple in second. But what about red vs. green? The doctrine of non-interference says that they should settle 3rd vs. 4th between themselves. But this is impossible. This puts teal in the position of having to decide. But teal might be totally agnostic about red vs. green. One solution would be for teal to allow red and green to advance towards each other across the top and fight it out that way. But then teal would drop below purple in strength and risk falling into second — or, hell, fourth! And the thing is, I didn't have to look very hard for a map like this. I went to the Kdice site looking for any kind of sample map to show off the graphics, and there it was. A lot of games work out like this. And most games fall prey to another kind of stalemate. See, as the map suggests, no space can hold more than eight dice. Remember how I said that tactics don't play a huge role in Kdice? In most games, players eliminate tactics entirely by dividing the map up amongst themselves, letting every space "stack up" to the 8d6 maximum — then waiting yet another few turns to build up some replacement dice — and only then do they venture an attack. So, you capture an neighbor's territory with an 8v8 attack... and he captures it back. You try again and fail... he captures one of yours. You capture it back... you're back where you started. Eventually someone will win a few 8v8s in a row, but that's just luck. Incredibly tedious luck. Once the scramble to connect is over, Kdice isn't actually, y'know, fun.

But even that isn't the main problem with the game. The main problem is the same problem that plagues every unmoderated space on the Internet: people suck. Now, let me be fair — are most people on Kdice unfit to interact with others? Nah. But are enough people on Kdice unfit to interact with others that, when you put seven of them together, at least one of them is likely to make playing the game about as pleasant as scuba-diving in the Pacific Trash Vortex? Yep. And it's not just the psychotic rage cases who start screaming "FUCK YOU! YOU FUCKING ASSHOLE I AM GOING TO FUCKING RAPE YOU FUCKING NIGGER FAGGOT!" as soon as the game turns against them. For every one of those, there are several more who don't sound like Krieg Mockery but still manage to pepper the chat window with casual racism, sexism, and homophobia — more than anything else, I quit Kdice because I'd had my fill of seeing the word "fag" thrown about more frequently than the word "the." Or they'll clearly be using the game as a way of getting out their aggressions, declaring "BAM! ha ha ha ha ha" because their imaginary dice just produced a higher number than someone else's imaginary dice. Now, one interesting thing about Kdice is that at the beginning of every month, everyone's score is set back to zero. For a couple of days the chat windows are full of complaints from the regulars about being sent back to the 0- and 100-level tables and how they can't wait to escape the "noobs" and get back to playing with the regulars. And for a short while I thought that maybe if I worked my score up high enough I'd be able to escape the sort of junior high bullshit I've talked about above. But then I did work my score up pretty high — as I mentioned in a pop-up note, I even won one of the tournaments — and it turned out that the objectionable players I'd been trying to escape were the regulars.

So with Broken Picture Telephone still broken and the appeal of Kdice having faded, I need another game to waste time on. Hmmm, what's this? "Save your lover"? Sounds promising...


Return to the Calendar page!