# Talk:Dots and Boxes

WikiProject Games (Rated Start-class, Mid-importance)
This article is within the scope of WikiProject Games, a collaborative effort to improve the coverage of game related topics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start  This article has been rated as Start-Class on the project's quality scale.
Mid  This article has been rated as Mid-importance on the project's importance scale.

## Game name??

This is usually called dots and boxes, isn't it? The linked gametheory.net page calls it such, and so does Richard's PBeM server, and even Elwyn Berlekamp (who has written a book on the game). So I think the page should be renamed. --Zundark 19:28 Dec 21, 2002 (UTC)

I've always known it as Boxes and obviously the same for the numerous previous contributors. Mintguy
I hope you're not really suggesting that anyone who edits a page without complaining about the title necessarily approves of the title. I've edited many pages with titles I don't approve of - complaining about them is usually just too much effort. --Zundark 10:13 Dec 23, 2002 (UTC)
Fair enough. If you want to move the page use move page from the menu or do it from here. Mintguy
Well. At least me and the user who wrote the article first then. Mintguy

## Duplicate article?

Duplicate article, needs merging:

Dots and boxes is a popular children's game, at least among mathematicans. The rules are as follows: The board begins as a rectangular grid of dots; six by six is a common size. The two players alternate moves, connecting adjacent dots with a horizontal or vertical line. If a player completes the fourth side of a square ("box") then she receives a point and must move again. When no vertical or horizontal line is left the player with the most points wins.

Dots and boxes has been carefully studied by Berlekamp, Conway, and Guy. See their book Winning Ways for further information.

## Second players winning in a draw?

The article states that " In games with an even number of boxes, it is conventional that if the game is tied then the win should be awarded to the second player (this offsets the advantage of going first)."

I have never heard of such a convention, what is the source? Besides, I think it is a wild claim that going first is an advantage. This is not "Hex" we are talking about and there is no strategy stealing argument. For example in a 1x1 game, clearly the second player wins. — Preceding unsigned comment added by 90.185.185.92 (talk) 20:19, 10 April 2012 (UTC)

## Java applet?

A few years ago I wrote a Java applet to play Dots and Boxes. What do you think: would it add to the article to have such an applet on Wikipedia? I don't know whether it's technically possible, but I don't see why not. I'll be happy to license it as GPL or even GFDL, but I want to find out what other people think first. dbenbenn | talk 16:55, 3 Feb 2005 (UTC)

Disagree - Although an external link to such an applet would be a useful addition. -Surturz 05:40, 8 January 2007 (UTC)

## An Oddity of the Universe

Isn't it strange how you can be sat next to someone you've never met before, and draw a grid of dots on a bit of paper, put a line between two of them, and slide it over, and they'll know exactly what to do, whoever they are?

## Chains?

I've seen the game before, but never really played much. The article doesn't really explain how a chain is defined. Obviously I don't know otherwise I'd add it myself, so could someone enlighten me and/or edit the article to say what exactly chains are? Thanks. --Ciaran H 19:37, 18 September 2005 (UTC)

## Why doesn't B win 4-0?

According to the article, A player who completes the fourth side of a box earns one point and takes another turn. In step 8 of the example illustrated at the top of the article, B completes the upper-left-hand box, yet somehow A then takes a turn and completes three boxes in a row without B being allowed to take another turn. Once B completes one box in step 8, why does A get to move immediately afterward? It would seem that upon completing the first box, B could then complete the remaining three boxes in three successive moves. --Metropolitan90 07:03, 10 January 2006 (UTC)

If you look a bit closer, you'll notice that in step 8 B completes the box and then draws the line going from the centre to the middle-right dot. In fact, any move B could make would give A the remaining 3 squares (the 4x4 game is actually analysed a small amount in the novel Golem in the Gears by Piers Anthony). However, I would suggest this page needs a little clean-up, possibly giving the rules their own section rather than including it in the introduction. Confusing Manifestation 16:04, 30 January 2006 (UTC)
Thanks, I revised the description of the sample game per your explanation. --Metropolitan90 06:13, 10 February 2006 (UTC)

## Good article?

Are there enough references for this to be a good article? --Jtalledo (talk) 20:15, 26 May 2006 (UTC)

Whaaaat!? The article itself sucks, it is not the references that are the problem. Sorry! Naki (talk) 19:20, 13 February 2013 (UTC)

## 6x6?

Well, at least here in Brasil, when there's nothing better to do, this can be played in huge entire-paper board (with about 1cm x 1cm each box)... And it's not so rare to see this kind of play... 6x6 here is for begginers... The only major problem of this is after the game, when counting the boxes (with ending scores around 120 x 80)... 201.56.56.96 22:20, 24 June 2006 (UTC)

On the contrary, large boards are most often used by beginners, because their size diminishes the effect of a mistake. Smaller boards better lend themselves to analysis and strategic play. On a 20x20 board an expert can simply wait until most of the board has been filled and then establish the correct chain parity with any number of sacrifices because it will not affect the outcome. Also, many strategies become obsolete on the large boards, such as the use of quads, nibbling, and incorrect parity wins. Most competitions and tournaments will use either 5x5 or 6x6. Anyone could play pool with 140 balls, but 9 makes it a more challenging game of skill. —The preceding unsigned comment was added by 74.102.156.147 (talkcontribs) 15:45, 8 August 2007.

## Solved?

Has this game been solved? For all size grids, or only up to a certain size?70.42.112.151 06:22, 31 October 2006 (UTC)

I imagine the game is only solvable up to a certain size. I calculate the number of moves in a game to equal ${\displaystyle (2cr-c-r)}$ where c = number of dots across, r = number of dots down. I calculate the number of possible games equal to the factorial ${\displaystyle (2cr-c-r)!}$ (actually you could probably divide that by 8 to account for rotations & reflections)
--Surturz 05:37, 8 January 2007 (UTC)
I appreciate this comment was made a long time ago, but for future readers, it is my understanding that the game is solved for grids of up to 5x5 boxes, that is to say 6x6 dots. I also seem to remember that winning strategies exist for any size box, involving the "first" player attempting to create a number of chains whose parity is that of the number of boxes. — Preceding unsigned comment added by 137.205.57.215 (talk) 12:55, 21 June 2011 (UTC)
OoooK! So, who wins (with perfect play for both players) on 5x5 grid - player 1 or 2? Unlike 4x4, where the game can end in a draw, with 5x5 the number of boxes is not even, so no draws. Thanks! Naki (talk) 13:01, 14 February 2013 (UTC)

## Double-cross strategy

Does anyone know who "officially" came up with the double-cross strategy of not completing the last few boxes in a chain? The reason I ask is that I thought of it circa 1990. I never published it or anything so I can't prove it (unless Julie Thompson, the girl at my high school to whom I showed the strategy, remembers my demonstration!), but I would like to know if I was the first, for my own ego. For all I know the strategy is hundreds of years old, of course --Surturz 04:45, 14 November 2006 (UTC)

The strategy has been around far longer than that, probably as long as the game itself. The concept relies on fairly simple reasoning, and is generally discovered by a player that reaches a certain level of sophistication and maturity with the game. Mindmatrix 13:54, 14 November 2006 (UTC)
Yes, I've discovered it myself too after playing for a while (Circa 1980). No big deal.
Fair enough. It would be nice to know the first recorded description of the strategy though. --Surturz 05:21, 8 January 2007 (UTC)
“Winning ways – Vol. 2: Games in Particular”, by E. R. Berlekamp and others, published in 1982, already mentions on the chapter dedicated to ths game (chapter 16) the double dealing strategy. -89.180.81.64 23:21, 12 November 2007 (UTC)

I think I remember this article used to have a link to an interactive game. I am putting one in anyway. A reader could learn a hell of a lot more about a subject like this simply from playing a couple of simple games. Why should WP not take advantage of the possibilities provided by the new technology? Just because a hard-back encyclopedia can't do it, does that mean we shouldn't? There is no salacious material here, no "inappropriate" subjects, no advertising, just good, clean, healthy dots and boxes fun. And nothing but that fun. Myles325a (talk) 01:28, 14 February 2009 (UTC)

I found an excellent implementation in main article (but it has been deleted), so I'll just leave the link here -- http://en.dots-game.org/.87.253.31.153 (talk) 02:09 ,8 January 2014 (UTC)

These links are removed per WP:EL because they do not elaborate on the subject, and Wikipedia is not a directory. It is irrelevant if the target site contains advertising. Mindmatrix 18:09, 8 January 2014 (UTC)
False. Nothing in WP:EL forbids showing examples of games, and nothing in WP:NOTLINK touches on examples, either. This is why I insist users quote the specific rule you think is being violated rather than just throw out the initials. It gives your argument a false sense of authority on the topic which you may not actually have.
I could now pull out WP:OWN since your pronouncement of what's allowed in the article feels like that to me. And you'd pull out WP:AGF, and then I'd point you to my talk page where I say WP:AGF is impossible and that I will use WP:IAR to get around it. But that's pointless. Instead policy wonking, let's just sit down and discuss the issue.
I think providing users a version of the game that they can try out does inherently add to knowledge to the game, in the same way that adding an image inherently adds when you are describing something visual. The only real problem here is that there is no one official version of the game, and we don't want to become a link farm, where multiple versions of the game are listed.
So let's work on that, instead of shutting down discussion with policy references. — trlkly 06:37, 28 January 2015 (UTC)

## Dot-centric variant

There's a variant of this game where instead of drawing a line on your turn, you claim one of the dots. Each dot can only be claimed once, and lines are drawn between any two adjacent claimed dots. (Which player claimed any given dot is irrelevant after their turn.) Any move which closes a box claims that box for the player who made the move, just as in this game. Thus, a single move can claim up to 4 boxes at once. There's no chaining - you get one dot per turn, every turn until all dots are claimed. It's not a very fair game - with perfect (or even decent) play, the second player is pretty much guaranteed a win. But then, few of these games are completely fair to both players. I'm not sure what the official name of this version of the game is (if it even has one), but I'd like to see some mention of it on this page, maybe with a little mathematical analysis if anybody has a source for that sort of thing. Lurlock (talk) 23:15, 2 December 2009 (UTC)

## not accepting a sacrifice CAN be to your benefit , mentioned differently in the article

There is a case in which not taking a sacrifice CAN be to your benefit. That happens when not accepting a sacrifice in a short chain results in an additional double cross (since the number of initial dots + total number of double crosses can sometimes determine who wins the game). Can anybody confirm this? Shayanuser (talk) 16:18, 16 December 2010 (UTC)

If your opponent makes a loony sacrifice (along the edge of a short-chain rather than up the middle like you're supposed to) then it's usually better to refuse it (unless you're nibbling). Loony moves are always a blunder if they aren't forced. Maybe someone more wiki-savvy than I can explain loony moves in the article? 24.18.225.26 (talk) 05:23, 23 April 2011 (UTC)
Here's a clear example where it's better not to accept a sacrifice. Suppose there's a 2x2 box and a long chain (more than 4 squares), the rest of the grid being full. Adam plays into the 2x2 box, sacrificing all four boxes in the hope of getting the chain. Beatrice should counter-sacrifice by playing across the 2x2 box without taking any boxes, giving all four boxes to Adam but getting the chain herself. This move would not be available if she accepted the sacrifice by taking even one of the four boxes. 2.25.135.89 (talk) 13:49, 1 August 2013 (UTC)