It seems simple, but can be deceptively tricky. I think this black and white gem warrants a thread of its own!
EDIT: I'll add something substantive later, but I'm not really the most expert. Hugh? Mongrel?
Oh boy. This might take a while, I just posted a big one (not about this) on another thread, and I have dishes to do. Hugh? Alpha?
Perhaps if I put up something controversial, maybe the others will respond.
Black has the edge.
I don't find this controversial. We might disagree on white's best defense, how to play beginners, and certain maneuvers I've seen you make as black :) Perhaps we should begin with very specific openings and argue from there.
Hugh wrote:I don't find this controversial. We might disagree on white's best defense, how to play beginners, and certain maneuvers I've seen you make as black :) Perhaps we should begin with very specific openings and argue from there.
My moves are airtight, and I will fistfight anyone who says otherwise.
By the way, leaderboard.
How about we label the board 1 2 3 4 5 for the rows (1 bottom, 5 top) and the columns A B C D E F G (A leftmost)
As black: You get the first move and control over some of the board development before "contact". You can get to the center space D3 first, which holds a ton of power.
During the first two moves (as black): Grab control over the center of the board. I take B3 C3 D3 and one other space that's not D2 or D4- if white is also
aggressive they can take a reasonable shot, snuffing out the bonus that you would get at the beginning of your next turn. So protect your bonus while staying aggressive. I really have no desired location for the 5th space, I don't think it's terribly significant in subsequent rounds (I can't remember thinking if only I had put it THERE 5 rounds ago). So, C2 or C4 if you want to go slightly aggressive, B2 or B4 for slightly more defense and some bonus protection, A2 or A4 for bonus protection and defense in later rounds. In most cases your just picking a row for an aggressive white to go down in later rounds (the row that does not have a black space, since it takes you a turn to put up any roadblock to their advance).
General philosophy:
I think the big obstacle to overcome is the nonstandard dice: with 5v8, to roll against a space with confidence one should have a little more than 3 times the number of attacking units as is defending. So 2 units in a high defense space is well protected during the first leg.
This game is a carousel of worry between position, bonus, and unit distribution. To advance in one category, you sacrifice elsewhere- and in certain rounds it makes sense to focus on one over the other. It's really a thing of beauty, and I'm surprised more of the regulars haven't tried it out. From what I gather it's the least biased duel map on the site, and has a following outside the usual fanclubs, or rank hunters who know the loophole. There is a learning curve, but you arrive a functional understanding in 3 to 4 games (unfortunately, this is about as much time it takes someone to say F*** this board). And, most importantly, it's fun and surprisingly replayable. Sure, the dice will blow up a few games, and this blow up can happen early, but no more than what is the accepted whims of lady luck.
As to turn bias, I think black does have an edge, barely. 55%/45%, and I've lost a lot of games as black, just ask Hugh.
D5 vs D8, 100 rolls of 3 dice vs 2 dice:
Attacker loses 140.92
Defender loses 59.07
2.38 times as many for parity. My statistics are rusty so I can't give you a 95% confidence level =]
I'm not surprised that that map is catching on, it looks like it has a chess/checkers/go type quality about it.
To everyone who makes maps other than Nygmongrel maps. Sorry. In retrospect, perhaps I'm being a bit too harsh. I'm under a lot of pressure finishing up the dissertation and looking for yobs, and I think I'm taking it out on the forums. So, group hug, XOXO, Mongrel.
@ Irolls: Didn't take long before my arbitrary definition was called out. "with confidence" is greater than a 64% chance of capture (with continuous attack). I have the exact numbers but am too lazy to go on campus computer to use the compiler.
For low units, rolling "with confidence" is
Pr(4 against 1) = 64%
Pr(7 against 2) = can't remember but > 64%
Pr(10 against 3) = can't remember but > Pr(7 against 2)
or, a little over 3 times as many attacking units as is defending.
My guess is your simulation is right in the large, but things are particularly delicate when dealing with extremely low units. Direct calculation shows that the chance of success rolling 3 against 1 is only 40%.
I generally just run some sims using this guy: http://prestopnik.com/warfish/rollSimulator/. Works really well.
From running a few 3 v 1 simulations, it appears that the simulator calculates go-for-broke odds: the chances of capture on any of the 3v1, 2v1, 1v1 rolls.
So, for low units, and low # of turns per round, it's way off. Plus, it's extremely laggy/browser breaky.
Easy argument for chance of capture with 5v8 dice and a single 3 against 1 attack: If the the defender rolls a 5,6,7,or 8 the attacker automatically loses. If the defender rolls a 1 2 3 or 4 there are still ways for the attacker to lose. So the chance of capture has to be less than 50%.
So when is the strategy discussion going to begin. If you were waiting for me, I will be watching this discussion now.
Oh, right. I was going to post something or other at a later date. Well, might as well post something now.
In my experience Spy v. Spy games break down into three stages:
Before getting into a discussion of those particular stages, there are a few basic strategy points.
1. Openings (where both players still have only +1 bonus)
The opening strategy for white and black can be quite different, so I'll take them in turn. Perhaps the most common move for black is to advance to 3c on the first turn, though variations have been seen where 1d or 2d are taken instead to keep unit-producing territories out of the line of fire during mid-game. The goal of the more direct strategy is to take 4c on turn 2, thereby gaining control of the powerful position in the middle of the board, and forcing white to take defensive or evasive maneuvers instead of launching a counter-offensive.
While it is usually most advantageous for black to take the offensive early, white has choices. As white, you can either choose to meet black in the center, setting up for a defensive blockade strategy (with variations including counter-attacks), or you can choose one of a series of evasions, designed to lure black into a frontal assault while gaining ground on the edges and keeping the unit-producing territories safely on the back lines.
That's all for now. Next installment I'll start getting to
2. Mid-game (where both players have +2 bonus or more)
-and-
3. Finish Him (where one player occupies his opponent's kill square)
Kjeld wrote:
- Early in the game, the central 4c square is important to control
that black can reach this square first is a big reason black has to stay aggressive (and why black has a slight edge).
I agree that my opening for black has solidified, but I can get creative with white.
That image post thingy would do wonders here.
Thanks for doing this Kjeld. Writing an articulate synopsis would take me a day, probably all week. Good read.
Just a few more quick basic strategy points (in blue). I'll get to the mid-game and finish him stages later on, but they'll take a while.
In my experience Spy v. Spy games break down into three stages:
Before getting into a discussion of those particular stages, there are a few basic strategy points.
1. Openings (where both players still have only +1 bonus)
The opening strategy for white and black can be quite different, so I'll take them in turn. Perhaps the most common move for black is to advance to 3c on the first turn, though variations have been seen where 1d or 2d are taken instead to keep unit-producing territories out of the line of fire during mid-game. The goal of the more direct strategy is to take 4c on turn 2, thereby gaining control of the powerful position in the middle of the board, and forcing white to take defensive or evasive maneuvers instead of launching a counter-offensive.
While it is usually most advantageous for black to take the offensive early, white has choices. As white, you can either choose to meet black in the center, setting up for a defensive blockade strategy (with variations including counter-attacks), or you can choose one of a series of evasions, designed to lure black into a frontal assault while gaining ground on the edges and keeping the unit-producing territories safely on the back lines.
Excellent post kjeld. This is discussion that shows why spys is such a fun board to play albeit frustrating at time when you lose attacks that were in your best interest to take.
While this thread makes me really happy, my strategy needs a little work. It usually involves having excellent defense to make it so that the opponent has to wait forever until they can finally eliminate me.
http://www.wargear.net/games/player/16718
Thought I played pretty well until I lost to guess who.
Well, the results don't always indicate the quality of play. There are two changes I would make on my side - move 2 I should go in the opposite direction since my plan was to prevent the "running strategy" you've been employing lately. And move 4 I should just allow you to have a 55/45 shot at the win and keep a guy on offense. Nonetheless, the defense and timed shot idea was interesting to try.
I thought it was masterfully played by both sides. Even hitting the 2 v 1 puts me in all kinds of trouble.