Anatomy of a Game: Big Brother
by Shannon Appelcline
Yesterday night was the season finale of Big Brother 4 on CBS. Yes, it's one of those trashy reality shows, but it's also something else, something rare: a true game on television.
Sure, there are all kinds of "game shows" on television: Who Wants to Be A Millionaire, Jeopardy, Wheel of Fortune. But for the most part they either amount to glorified trivia games or games based on almost pure luck. There's no gameplay here, no strategy, no design.
Big Brother, and its closest Brethren, Survivor, which I discussed back in Trials, Triumphs & Trivialities #74, Anatomy of a Game: Survivor, are really something different. They're examples of good game design that we, as game designers can learn from.
And, if the ratings are any indication, a number of you secretly watch these shows. So these week I'm going to look at the recently departed Big Brother and try and lay out what makes its game design good & interesting. (And, if you came here looking for the third part of my series on the history of design in roleplaying games, I'll be back to that in just two weeks.)
A Voting Game
Like Survivor, Big Brother is a voting game. Each one of the 13 initial contestants comes into the house with equal power: one vote. Each week the contestants then vote one player out of the game. Somehow each contestant must parlay their singular vote into a majority that keeps them in the house until there are only two people left. Taken in the abstract, this is a fairly common strategic game design: you start players off with exactly equal resources and give them some opportunity to turn those resources into victory. Chess, Backgammon, Checkers, and any number of other games all follow this ideal in the abstract.
However, this core design also immediately reveals its biggest flaw. If you have n people in the game, a group of n/2 (rounded down) + 1 people can dominate the game. In Big Brother, with its 13 players, 7 people can thus dominate. This can result in both a boring game and boring television. Without refinement an alliance of 7 houseguests could have quickly formed in Big Brother, then spent the next 6 weeks voting out their initial targets.
In order to combat this game design issue, the producers of Big Brother have made changes which dramatically reduce the power of a majority.
Head of Household: Breaking the Pattern
Every week in Big Brother a Head of Household is elected. This election is determined not by votes, but rather by competitions--perhaps physical, perhaps mental. Rather than having an open vote for whom is evicted, instead the Head of Household chooses two candidates for eviction. Only one of the elected candidates can be voted out of the house.
Assuming that each player now had an exactly equal chance of winning the Head of Household competition, that means that a 7-player majority might win Head of Household 54% of the time, and a 6-player minority might win 46% of the time. Since an intelligent Head of Household will always elect members from outside of his alliance for eviction, suddenly a one-vote majority becomes marginal. A majority has a slight advantage, but can still go down fairly easily. Increases in majority size make the chance of victory increasingly large, but never 100%.
What Big Brother decided to do was, effectively, take an entirely arbitrary vote system and introduce a random element into it to make the game more interesting through the introduction of chaos. This is exactly why you want randomness in a game, as I discussed back in Trials, Triumphs & Trivialities #106, Designing Strategy: The Random Factor. When you compare the outcome of Big Brother 4, which bounced back and forth between two alliances, with the boring and expected person-by-person destruction of a team that you saw in Survivor 1 and Survivor 2, it's pretty obvious that the designers at Big Brother made a good choice.
Of course, any change you make in game design is going to have other effects, some of which may not be desirable ...
Kill the Strong!
Winning Head of Household insures survival for not just yourself, but also your team. Thus, the best way the you can make sure the opposing team doesn't survive is by taking out their members most likely to win Head of Household. This means that week by week the strongest competitors are the most likely to be booted out of the game until your are left only with a well of mediocrity.
And, to a certain extent, this is exactly what we saw in Big Brother 4. One of the alliances was "The Dream Team" a group of 3 men: Justin, Jee, and Robert. Competively, Justin was a major threat, Jee was a serious threat, and Robert was a medium threat. And, that was exactly the order that they were voted out in. The other major alliance in the team was somewhat more fractured, but we did see two of the strongest competitors, David and Nathan, voted out early on.
Every game system change has serious results. That's always an important fact to remember in game design. Potentially, this particular result could be deadly, as it turns upside down the expected results of competitive play.
(There's another result of the whole Head of Household system, which is that decisive players will tend to be voted out earlier than indecisive players. You can know for certain that a clear enemy will put you up for eviction, but will never be certain of a "floater" player, thus you take out the clear threat before the uncertain one. But, this result has less of a worrisome effect on the game overall.)
Veto: Saving the Strong
So, looking at the game design thus far: players form into alliance to create unequal allocations of their initially equal starting positions. Strong players help a team to win protection from eviction by electing two strong players in an opposing team to be voted out. In future weeks the random factor will change the balance of power such that the previous Head of Household is likely to find himself up for eviction for similar reasons of strength.
Enter the Golden Power of Veto. An additional competition is held after the Head of Household competition wherein each player competes to try and win the veto, which allows one player to be taken off the eviction block; if someone threatened with being voted out is on the block and wins the veto, they can veto themselves out. Thus, some of the strong remain.
Overall, the strong are still slowly winnowed out of the house. As likely as not, two strong players will be up for eviction each week; even if one saves himself, the other strong player will still go. However, the Power of Veto does give a player at least the opportunity to carry themselves all the way to the end. In Big Brother 4 Allison, one of the floater players, did exactly this. Through alternative weeks of winning Head of Household and Veto she ensured that she couldn't be voted out, even after it was obvious that she was a threat.
This sequence of game design decisions that originated with the power of a majority is an intriguing one. First, a large problem was revealed: the absolute power of majority could lead to a boring game. It was fixed (with the Head of Household's ability to pare down who gets voted out), but this produced a medium problem: strong players were actively selected against. So another solution was introduced (the Golden Power of Veto).
In your own game design, expect to follow this pattern of introducing solutions that slowly reduce the magnitude of problems until they're either gone or insignificant.
Playing for Second Place
I feel like there's one other interesting bit of game design in Big Brother: the way in which the winner is selected (which is functionally identical to the same selection in Survivor). At the end of the game there are just two players left. At this point, some number of former Houseguests then get to vote for these final winners (in BB3 10 Houseguests voted, while in BB4 only 7 did).
Practically this means that if you backstab someone or betray them, even if it gets you further in the game, it could prevent you from winning.
From a game design perspective, I believe that this means that you can only play for second place. Because, ensuring that you get to the final two will probably cause you to ruffle some feathers and betray some people. Just considering my initial analysis that you need a majority of 7 to take power, it should be immediately obvious that this means that you're going to have to cut out 5 people who thought they were on your team. Or, to put it another way, if you do the strategic moves necessary to get yourself to the end, you're much less likely to be the type of person that people will vote for, in the end. Conversely, someone who closes their eyes and depends upon luck to carry them to the final two, will likely have ruffled less feathers, and thus be more likely to win.
This has proven itself in both BB3 and BB4. In BB3, one of the final two, Danielle, was a brilliant and careful gameplayer who stabbed player after player in the back. She lost the final vote 1-9. In BB4 there was likewise an untrustworthy backstabber who played a great (if despicable) game. She lost 1-6.
The strategic game player does sometimes win, as in Survivor 1 or Survivor 5. However, this generally depends upon the willingness of the evicted members to reward someone for gameplay over honesty. Some juries are willing to do this, but it's a toss-up.
The only truly strategic solution is a sort of Prisoner's Dilemma. If there is another strategic player in the game, you have to backstab just a little less than them. This is the only way to assure yourself of #1; play good enough to stay in, but not quite as dirty as the other player. However, like the question of jury mentality, this is a toss-up. If the right player isn't in the game, you can't play this way.
The fact that generally there isn't a solution for how to assure yourself 1st place in the game, rather than 2nd is, I think, a flaw of these games. Playing for second might be acceptable, but if a winner regularly comes from an ultimately undeserving group who really didn't play the game well, the game will eventually lose its luster. Imagine playing a multiplayer game of Chess or Backgammon where you, the best player, always came in second, while first place either went to crazy Uncle Ralph or your four-year old sister, Beebee. It'd be frustrating, and you'd question the game design. Yet, "unworthy" winners have come out of these reality games, the winners of Survivor 4 and Big Brother 3 among them.
I don't know what the best solution for these games is; the idea of a jury system is interesting, because it forces gameplayers to offbalance betrayals with cost. But this particular flaw may eventually spell the doom of these shows.
One of the most interesting ways to learn about game design is to analyzee existing games. It's just like any craft: you learn from what's been done. Big Brother and its brethren form interesting and unique possibilities to see large what is discussed here every other week.
In two weeks I'll be back with my final look into the history of RPG design. See you then.