"Game Theory 1.1" Article from the Games Journal

OK, this seems as good a time as any to spout off some of my latest theories...

I like the Game Theory line of articles, but I don't see things quite the same way. I look at significant events in the game as "breaks."

Think of tennis. One player is serving to another. The serving player is considered to have the advantage. Since players alternate serves every game in a match, if we look at it from a strictly probablistic view, the game will never end because the serving player will always win.

Obviously, this is never the case. At some point, the other player "breaks serve." This is always a big point in the match, because it often dictates who will win the set, and possibly the match.

This idea of "breaks" translates into any other gaming activity. It's a clearly defined moment when a player's actions will win the game - not the actual moment of victory, mind you, but a single event that, seen in retrospect, is a crucial turning point in the flow of the game. In Puerto Rico, it's often the purchase of a significant building, or a captain phase where one player gains much more VP than his opponents. In Settlers of Catan, it's a big trade that gets a player the resources he needs to build a city, or acquire longest road, and so on. In Chess, it's often the capturing of the other player's queen. In Monopoly, it's when a player lands on the one property that makes him bankrupt.

I have to disagree with Jonathan Degann, at least through my own subjective opinion. When I finish a game of PR, I don't feel as though I've "lived through an entire family saga of immigrating to a rough land and building an empire on it." I see it much more mechanically, as a game, a series of choices and consequences. I'm sure there are those people who are drawn far enough to feel a connection to the theme; but this is probably more the case in games like "Titan" or "Advanced Squad Leader," games which have such an epic scope and such detailed chrome that players are much more easily swept in.

Nevertheless, I see things differently. What Degann sees as a "story arc," I see as a series of breaks. In any game, players will encounter several breaks. The key is, in a well-designed game, they don't know which breaks are the most important until the winner is revealed.

There are a few ways of ways of doing this. First is to have one of our favorite acronyms here... MPTV, or Multiple Paths to Victory. If a game is a simple points race where everyone sees everyone else's status, then a runaway leader problem can emerge. But if the race takes place on several different axes, each of which ends in an instant victory, then a player losing on one axis can shift his efforts to the next. So a PR player shut out of shipping switches to a building strategy, or a SoC player, seeing the last development card taken, knows he won't get Largest Army and attempts to build cities instead.

Another is to hide a player's status (i.e. victory points) during the game. In PR, this often leads to shocking decisions ("How the hell did I win?" is a familiar refrain, second only to "How the hell did you win?!?"). In SoC, this is done by placing some VP in development cards that are hidden from other players. As long as players have a general (but not exact) idea of everyone's status in the game, this is an effective device.

Finally, most well-designed games feature their most important break at the end of the game. In PR, these are the 10-doubloon buildings that can give the builders an advantage over the shippers. In Princes of Florence, these are Prestige cards, coupled with the rising minimum values of works throughout a game. In Ra, these are the monuments that only score after the third epoch. And in SoC... well, there are many people who claim that Settlers of Catan can have a problem with a runaway leader, and this is why. There is no special mechanic to give a large break at the game's climax, so its endgame is not terribly dynamic. (This is why I said "most well-designed games," not "all well-designed games.") :)

It's interesting you bring up racing. I've tried designing several racing games, and they all wound up with a runaway leader problem. According to my above criteria, there are two problems. First, there aren't multiple paths to victory. Fastest wins. Secondly, everyone knows how everyone else is doing. A player's position in the race is not hidden information.

A couple of neat solutions to this: Wolfgang Kramer's Top Race (and Daytona 500, et al) features cards that move all cars. So when you play a card, you're moving everyone's pawn, not just yours! Since you don't know the other players' cards, you don't know which player can play a highest-valued movement card for your car or cars at the wrong time, wasting movement on redundant spaces and possibly costing you the race. That's a potentially huge break that can happen during anyone's turn.

Also, these games have players bid on their cars, and pay off money to each player at the end of the race relative to how that player finished. You can bid lots of money for a car of which you have lots of cards for, but if you don't place well in the race, you probably won't win the game; whereas a player who gets his cars for dirt cheap and places them high in the finish will often win. This can be roughly equated with hidden status during the game. Despite the fact that the racing order is public knowledge, no one's sure of who won the actual game until the money is counted.

Another racing game to look at is Formula De, which makes proper cornering of vital importance. If you go into a corner too fast, you lose valuable points from aspects of your car. Your car can handle a little bit of abuse at the beginning of the race; the question is whether to play it safe, go slow, and avoid damage, or to lean on the accelerator and hope you make it out of the race with your head still attached.

This means that the race leader can build up a seemingly insurmountable lead, but if he put too much stress on his car to get there, he could have to pull back (or even put himself out of the game by wrecking), and allow someone else back into the race.

So in terms of racing, I don't know if you can put a literary-type Story Arc into the game. However, you can have an arc that resembles an actual race, and the best way of doing that is probably by paying attention to the breaks players earn during gameplay.

I hope this helps!

Hey Dark... there are some good points you're bringing up here.

which seems extremely similar to Degann's definition of a "bomb":

Yeah, at this point, I think Degann and I are using different words to describe the same thing. Perhaps he means "bomb" to be a single, epiphanic moment, whereas I'm thinking of breaks as being several points in the game; but that's quibbling. The central gist is the same.

Hey, I'm learning too. So keep the good ideas coming. :)

This is where theme comes in. If you have a fixed idea, like if you're only interested in simulating F1 open-wheel racing, then you've created a very specific environment in which to work. On the other hand, Zaiga has been workshopping a game in the GDW that has shifted from auto racing to dragon racing to velocoraptor racing. :)

If you're starting with a clean slate like Zaiga is, you can lay out your game design goals and work your themes around them. But if you're set on putting your game in a specific setting, it gets tricky. I don't think it's possible to design an F1 simulation game with distinct multiple paths to victory. There may be different techniques players may try (aggressive racing, lots of quick pit stops, etc.). But since F1 racing is a skill-based sport where strategic thought doesn't have a huge impact on the race (it has some, yes, but it usually doesn't outweigh a driver's performance; an inferior driver who makes good decisions will lose to Michael Schumacher making average decisions), your MPTV options are diminished from the get-go. Ultimately, everyone's strategy is the same: go fast.

So a fantasy racing game could have two dragons dashing around different kinds of terrain, the strategic selection of which can prove to be victorious; and you can design an alternate path to victory that involves charring knights and capturing maidens. But you can't do that in an F1 sim; you're bound to your theme.

Wolfgang Kramer skirted this with his bidding mechanic for Top Race/Daytona 500/etc.; but as you can probably tell, the mechanic came before the theme. There are no real-life racing leagues where teams bid for cars just before the race. It plays well as a board game, but is not intended to be a simulation (Aside: I was teaching Daytona 500 to an 11-year-old NASCAR fan, and he scrunched his nose in disappointment when I told him that you couldn't run another car into the wall).

So: is your game going to be realistic, or are you more interested in a novel mechanic? There's nothing wrong with either direction, but deciding now will make design decisions easier.

Yes, this is very important in most games. An important note: the longer the game, the more this matters. But if the game is short (i.e. under 15 minutes), don't worry about blowouts. To introduce a couple of non-racing examples, Mag-Blast and Nuclear War are two games where a player can be eliminated from the game before he even takes his first turn. If these games took an hour to play, it would be an unforgivable design flaw. But since each game runs about 10-15 minutes, and feature "take that!" mechanics, extremely direct interaction, and lots of lucky card draws, it's not a big deal. They're a pretty specific slice of game.

Now if you're looking to make your race game take a good part of a night, then it's in your best interest to make sure everyone feels like they're involved in the race until the very end. You may want to avoid elimiation scenarios, or even worse, kingmaker situations (when a losing player's effectively decides through his actions which player is going to win).

The polar opposite of "runaway leader syndrome" is "bash-the-leader syndrome." You may not want to tilt your game too far in this direction. This will feel like the opposite of a race, as every player rushes to put their pawns in last. :)

You'd also be pulling your design away from a strategic game, and make it more of a tactical game, where each player's movements are based on the cards they have when they take their turn. It's not a terrible mechanic, but without a strong strategic feel, you may not want the race to last very long. It's difficult to have 90 minutes of "play the hand you're dealt" remain interesting for everyone.

Again, this would be very tactical, but if you're designing a long-format game, you may want to avoid a game where the winner is the player who was dealt the nastiest cards. Also, what's to prevent the first-place player from playing his evil cards on someone in distant second?

Here's one way to make the game strategic... some of the cards players collect can be converted into different types of upgrades that can make the player's pawn grow faster, or breathe fire out of the dragon's mouth, or let a player move twice in one turn, and so on. Each upgrade would require different sets of cards to be traded in, so you'd have to be specifically saving up for that upgrade in order to get it. Different parts of the track would be easier approached with some upgrades, whereas others would help in a more general way.

Of course, there might be another path to victory, where a player could forego the upgrades and attempt to move his pawn "naked" through the course. If the other players get too sidetracked aquiring their upgrades, this could be a valid alternate path to victory.

It's not the best idea, but I suppose it's a ballpark example. :)

The horse racing game Turfmeister avoids runaway leaders by limiting the amount of spaces the top three horses can move. When I first heard about this rule, I was a little hesitant; it sounded like the leader of the race would be unjustly punished, and the whole game would feel unnatural.

But I played it, and was pleasantly surprised. The game feels like horse racing. In order to get into first place, you need to make a charge. But since you are limited in the number of spaces you can move once you get to the front of the pack, you're more likely to hang back and wait until the home stretch to make your charge - just like in real horse racing! It's a fantastic example of a rule both enhancing play and fortifying the theme.

I'll tell you a little about the state of my racing game now. It's called Star Racers, because it involves spaceships racing around a star. When I put together the first prototype, I was dead-set against arming the ships; I wanted this to be "NASCAR" in space, and wanted to avoid any direct conflict between the participants. The idea would be, it was racing only, and people who wanted to shoot stuff up could play Star Warriors.

I'm on my third or fourth design of the game, and I've given in to the idea of arming the ships. Why? Because it's a great device to keep players from jumping into the lead. The ships are armed with forward-mounted lasers that can only fire in a line, coming straight out of the nose of the ship. So players aren't firing willy-nilly, but have to manuever their crafts onto the leader's six. There are times you'll want to fly behind another ship, to get a clean shot off; whereas without the weapons, no one would be afraid of charging ahead.

Even with the weapons introduced, I had another problem. Suppose there were three ships at the end of the race. The craft in third had no chance of winning; it was between the other two ships. But the third-place player has a clear shot at one of the two leading ships. This would be a problem; the third-place ship has to play kingmaker. The ship he doesn't shoot is the one that wins the race.

I solved this by deciding that laser fire could hit more than one target, as long as they were lined up correctly. So if the two leading ships are tail-to-nose, the third-place player can blast both of them, slow them down enough to pass them, and win the race. I don't think it breaks theme too much either, if I can sell players the idea that these laser cannons are powerful enough to punch through the crafts' hulls and out the other side. :twisted:

The game is far from finished, and who knows, I might scrap this design too. But it feels much more promising than the previous designs.

DarkDream

You too. Good luck with your game! :)

What do you think?

You seem to have some great ideas, DarkDream, and I share your philosophy that games should be short, simple and fun. Welcome to the forum!

To answer your question. I like the idea of some in between racing action, to add another layer of strategy, but I'm concerned that the fact that the winner gets the most money and that money also is a resource in the game. This could potential lead to a runaway leader syndrome. The player who wins the first race will have the most cash, which allows him to buy better upgrades, which allows him to win the next race even more easily, etc.

I think a better mechanic is if you somehow make the resources and the victory points different things. A way to accomplish this in a chariot racing game (just an idea) is to introduce the concept of damage (the resource). To win a race you probably have to take a lot of risks, which means you take a lot of damage. You need money (victory points) to repair this damage, or take the decision not to repair the damage and keeping your money, but then you run the risk of crashing during the next race. This way you introduce the element of balancing resources versus victory points and you bring in an element of risk management.

Now, there's a bit of a problem of a chariot crashing. A crashed chariot will probably not get any money, but it will need the most money to repair his chariot! This is kind of a reversed runaway leader problem. You could solve this by giving that player a brand new chariot for free or you could just let it be, in which case taking huge risks during the race becomes a very tricky matter, but maybe those all-or-nothing situations make for very tense and exciting moments and a lot of fun. Just make sure that the game is short enough in that case, so that players who are out of the race don't have to wait too long.

You could still keep the upgrade thing and the betting, these add just another layer of strategy and decision making to the game. You might want to check out Royal Turf, a game about horse racing, which also features betting.

- Rene Wiersma

I found the article fascinating, but I agree that the thematic tie-in doesn't work for me either. I see the whole idea of a "story arc" on a more abstract level as in: early game -> mid game -> end game where in each phase different things are important.

I agree with this, although I think that every decision point in the game should be a "break". Some are obviously more important than others, but I do think that every decision should have an effect on the outcome of the game. Otherwise, what was the use of the decision?

I don't think "breaks" and "story arc" are mutually exclusive. On the contrary, I think you need to incorporate both elements to get an exciting game.

This sounds a bit like the "bomb" effect in Jonathan's second article. I agree that as the game progresses the stakes should become higher, although not so high that they automatically render any previous decisions useless.

Perhaps "SoC" is not such a well-designed game after all? :wink: Really, I have formulated a whole theory about the difference between VP/time based games and binary ending games and their respective pro's and cons. One of the cons of binary ending games (when someone reaches a certain condition he wins the game and thereby also terminates it) was that the endgame of such games are usually not very dynamic, as you put it. When I find some time, I'll write an article about it.

- Rene Wiersma