Games, Part 1

We're trying something new with formatting. Instead of posting each paragraph as a separate entry, the whole series will be one entry that continues via links. Hope it helps. Some of you may have heard my opinions on games before. I�m not a very big fan. I would nearly always rather discuss strategy with my friend than play against him. The word 'against' may strike some of you as inaccurate, but it�s largely beside the point. The thing I don�t like about games is that you rarely learn anything important about your friends while playing. Many of you also know that this is a fairly game-loving crew. How then will I cope? It remains to be seen, but I have accepted that some playing of games is permissible when one lives close enough to see friends virtually every day (after all, how many meaningful conversations can you have on such a regular basis?). Therefore, I have spent some considerable time and thought recently on game strategies. I think this comes from consistently being the worst player on any team in elementary school, and the consequently desperate fear of having no one to hang out with at recess for lack of skill at games. Some of the strategies seem so fool-proof that I would practically assert them as algorithms for winning, in which case, the games again seem hardly worth playing. But then, what do I know? The following games may be won according to my simple plan:

Clue: More Than Meets the Eye

Clue may seem to be child�s play, a simple game of deduction. However, like baseball, with the proper scoring tablature and more than two or three people, it can be much more complex. The trouble with the Clue scoresheet is that it doesn�t have room to write down partial information. I�m sure you have noticed that when someone makes a suggestion no one can answer, everyone�s ears perk up, but where can you record the fact that one of those three cards is in the envelope? Some people put dots to the left of the word, but its difficult to scratch them out as you narrow it down. I usually just write the three at the bottom since there are seldom more than three of these moments in a game. You (and everyone else for that matter), then try to discern which of the three is actually in the envelope by forcing the others to show you certain cards (by including cards in your own hand in the suggestion). This brings up another important point. Anyone who is being systematic about her investigation will have to call on certain people and weapons more often than others because they are the cards in her hand. In fact, there is a very high probability that anyone who uses a suspect or weapon twice has that card. Therefore, I put that person�s initial next to the suspect and weapon he names in each suggestion. This might seem messy, but it is spread out over enough items that it seldom gets untidy. This comes in very handy towards the end when everyone is racing to find the last element.
Clue For Champions
These two records alone should give you a major leg-up, but if you want to go whole hog, you have to keep track of who answers to which suggestions. If, for instance, I were playing with Kristen, Michele, and Gene, and Kristen suggested that it was Mrs. Peacock, with the Wrench, in the Conservatory and Michele showed her a card, I then know that Michele has either Mrs. Peacock, the Wrench, or the Conservatory, n�est-ce pas? Now the trick is to record this somewhere so that later, when Kristen shows me the conservatory card, I can narrow my deduction to Michele having either Mrs. P or the Wrench. This can�t be done with just the clue score sheet; you need a sheet of lined paper with four columns for each opponent: suspect, weapon, room, and asker (or this form). When the same answer shows up twice to different askers, the person probably has it. (The asker column is simply to make sure that if something repeats it is not a function of someone using their card to force a particular answer.) You can also cross out cards you know they don�t have (until you can deduce that Michele showed Kristen the wrench) but this seldom results in knowledge you couldn�t get via one of the other methods and involves cross-referencing your list of who definitely has which cards.
Extreme Clue, a la Celebrity Fear Factor
The other thing that you need for this is a good one or two-letter, non-repeating code for easy cataloging and reference. This is going pretty far even for me, but in the interests of your mirth and Clue-playing prowess, here�s what I got: M-Colonel Mustard, P-Professor Plum, G-Mr. Green, K-Mrs. Peacock, S-Miss Scarlet, W-Mrs. White, N-Knife, CK-Candlestick, V-Revolver, R-Rope, LP-Lead Pipe, WR-Wrench, H-Hall, U-Lounge, D-Dining Room, T-Kitchen, B-Ballroom, CV-Conservatory, BR-Billiard Room, I-Library, Y-Study.

Settlers of Catan: Contingency Dependence for Fun and Profit!

Gene has a new game, and it is great fun, but unlike Clue, it is less complex than it seems. Winning Settlers of Catan has almost everything to do with the initial placement of your settlements. For those of you who have not played, it is much like all "civilization" games where the goal is growth, which is exponential, not linear. Therefore, you need the most resources in the beginning to get bigger faster than your opponents. You get resources from someone rolling the number on the resource space (hereafter called 'hex' for its six-sided shape) your settlement is next to. Therefore, you want to build next to hexes that have numbers that get rolled often (6 and 8, then 5 and 9, etc.). The makers of the game have been kind enough to put little dots on all the numbers that show how often they come up. Since each settlement sits at the corner of three hexes, you can simply add up their little dots to see how often a settlement there will give you resources. In my experience, you have to have the highest total score for your two initial settlements to win.
Settlers of Catan: Small Print and Nitpickiness for Obsessive Gamers
There are other things you have to do too, but getting that high score is the determining factor. Luckily, the way odds would have it, there is usually one most valuable intersection (17 or 16) two or three valuable ones (15) and some okay ones (12 or less). Since the placing goes ABCDDCBA, A is likely to have the most and least valuable and D is likely to get two average picks (see example). If everyone chooses the best available spot, usually at least two people tie for best initial set-up. Picking a less valuable intersection because it has wheat and ore instead of brick and sheep is wishful thinking. Take the most valuable intersection available at every chance and adjust your strategy to suit your holdings. There are other factors that will help you decide between two equal intersections, like proximity to a port, complimentarity to your other resources, scarcity, abundance, room for expansion, etc, but the bottom line is that you have to get the highest total (or tie for it) to be in the running for the win. The ideal situation is to end up with lots of wheat and ore that you share with one other player on each hex and a settlement two road-lengths away from the wheat port. This prevents you from getting the Robber too frequently and allows you to build cities early, which double your production.
Ports and the Yellow-Brick Road
You�ll need a port to save yourself from too many pricey trades at the bank, and I like resource specific ports because it�s much easier to get two of something than three. Pick the port that best matches your biggest resource (and preferably the one that looks like it will be most common in the game so you can 'loan' other people your port on your turn for a one-card fee). From there, it�s pretty much up to the dice. You�ll need the longest road or the largest army to win too, so if you have brick and lumber, connect your roads, and if you have ore and wheat, buy cards. You can win with four cities and the largest army or with four settlements, two cities, and the longest road. There are other combinations, but those are the two most common.


Though I have only played this game once, it is also a game of statistics. You roll five dice and get points for particular arrangements. Except for the straight and the flush, all the other situations reward you for having more similar numbers. Therefore, you only have three possible goals on any roll: (1) all the same, (2) all different, and (3) three of a kind with a pair. The flush usually comes at some point when trying for all the same (a yahtzee), and it will be obvious which rolls predispose you to a straight. When you get nothing, use your ones, your yahtzee, or your chance. There is a bonus that requires that you have at least three of a kind for each number in the top section. A high-scoring chance is a great way to pick up some points, but probably not as many as you would get for the bonus, and you�ll almost never get a yahtzee (odds are 3:1188 or 1:396). Your chances rise to 1:36 if you roll three of a kind on the first roll. I say, take the flush if it comes and then use your next three of a kind to go for the yahtzee, if you're lucky enough to get one.

The Point

In essence, in Clue, Settlers of Catan, and Yahtzee, there is always a single best move and you simply need to organize your the data well enough to consistently make it. Almost never will inspiration or luck overcome the grueling consistency of pattern for manufacturing a given outcome. What fun!


I find it funny that after playing Settlers, only once you feel confident about the fact that "you have to have the highest total score for your two initial settlements to win."

In that Settlers has become like a modern day "Risk" there is plenty of geekery out there regarding strategy. I have a feeling it's a little more complex than merely the point value of the initial placement.


I think your Yahtzee probability is a little off. If you roll three-of-a-kind initially, your Yahtzee chances move to a whopping 7 1/2 %, though taking a shot at a Yahtzee with two dice on one's final roll is indeed a 1/36 chance.

Yahtzee is a great deal more random than the other games you listed, so I think a "killer app" strategy is less useful. Also, some might argue that the killer app stratgey to game playing is "boring", "joyless", or even "something only a dedicated nerd would think up", but to me, that sounds like whining from someone who just lost the game. So what if it's joyless? There's no joy in winning when you're expected to win anyway.

The one thing your advice leaves out is the inter-personal aspect. A killer app strategy is much better when opponents do not realize you're using such a strategy.

I inadvertantly cut off that comment. I will have a more detailed response on my own page.

It's a very good point. After getting "How to win at Monopoly from the library" and reading it, Monopoly isn't fun anymore and nobody has any interest in playing with me.

Seems kinda lame to, in response to not enjoying playing games, attempting to reduce them to simple logic gates. For those who enjoy games, like myself, I'd say leave them be. I regret reading that book because Monopoly was a fun game for me.

Gene, I wouldn't say that *nobody* wants to play Monopoly with you ... I'd play with you!

Of course, this is only after me spending the last 15 years recovering from the terribly cruel beating (at Monopoly) you gave me when I was 8 and you were 10, and you had just finished reading the "How to Win at Monopoly" book. Did I mention that you were TEN when you read that book?

Love you!

Are you serious? You're kidding me. How do you remember this stuff?!

God, it's all a blur to me.

In Reply

Sean is absolutely right about the yahtzee miscalculation. The actual odds of a yahtzee with an initial three of a kind (a 15% probability in the first place)are 10:36 that either die comes up matching (plus the 1:36 chance that both do) and 1:6 that the other comes up on the last roll. In the more likely (25:36) case, you roll both dice again and have a 1:36 chance of getting a yahtzee. All of that gives you a 9.33% chance, according to my calculations, of getting a yahtzee with an initial three of a kind. I may still be off here, but the point is that doing the calculations is more fun than simply tossing the dice and hoping.

With regards to Settlers of Catan, I think if you read those guides (as I did before posting anything), you will find that they are all plagiarized versions of each other and that they all say the same thing: there are a number of potential strategies in use, but they all require strong initial placement and flexibility in strategy choice to suit the board. You will also find, if you run a number of simulations (as I did), that though there are other important pieces to strategy (like ports), no one who begins with a low initial resource score ever wins.

I never liked Monopoly, but am much relieved to know that there is a book that can reduce it too to a "killer-app."

All sarcasm aside, my approach clearly, willfully neglects the human interaction component. This stems from my earlier statement that if it's interaction you seek, talk or investigate or do something, but don't play a game. This view may further be traced to my "never settle" approach to life, whose costs I will leave you to calculate another time.


"doing the calculations"


February 2012
Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29      

About this Entry

This page contains a single entry by published on December 3, 2003 12:33 PM.

What Can You Really Test? was the previous entry in this blog.

Games, Part 2 is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.


Powered by Movable Type 5.04