Skip to Content

Runaway Leader

10 replies [Last post]
GamingNerd's picture
Joined: 03/26/2012

In a playtest of my game last night we had a Clear Winner. And we knew he'd be the winner (more than likely) from fairly early on. In the end his win was not a shut-out, but it was a 10 point victory.

In the end he won because everything went his way. He got the cards he needed, I couldn't draw the card I needed for the life of me. There's some bonus points that show up randomly on the board, in one instance 3 of these showed up in his direct path. I picked up two for the entire game.

There's also a few other considerations the biggest of which is that the game has spell cards which can be used to help you along, or to hinder an opponent and I was intentionally NOT buying these while he intentionally bought as much as he could - the idea being to see how the extremes of play strategy would work out.

We worked it back and without the luck of the bonus tiles and the spells, he won by 2 points making it a much closer race.

My question is: what do you think constitutes a Runaway Leader problem? I want to be aware, but I don't want to over-think this, any game with a certain amount of luck has the possibility for the luck to run almost entirely in one player's favor.

- Andrew

JHouse's picture
Joined: 04/03/2010
Some things to think about

Before I start there's one thing that I'm confused about: Did you enter the game planning on testing the 'Extremes' of play or did you start this after the leader had emerged? If it's the latter, then you should test play again because the original play is tainted. IE. had you used spells could you have caught up? If so, you may not have a run away leader problem - just an early leader.

If you do have a runaway leader problem, here's some stuff to think about (with my limited knowledge of the game):

Are bonus points necessary?
- What do they add?

Can bonus points be given in a more controlled way?

Is there a limit on spells a single person can buy a turn?

Is there a way to limit the leader on purchasing spells?
- NOT stop, but limit - ie. Spells cost one more resource for the leader.

In almost any game there's a slim chance that someone can just run with it, however making that chance as small as possible is the best solution. An average person playing their first game expects to lose but if they get stomped they're less likely to want to try again.

Izraphael's picture
Joined: 01/29/2010

I think the best thing to do is someway limit the leading player (as JHouse suggested). Make things harder for him: he gets less points from bonuses, everything costs more, he do something "less" (draw less cards, etc).
Question: maybe the spells are overpowered? It seems that ignoring them is not a "fair alternative", you don't gain as much as buying them.

MarkKreitler's picture
Joined: 11/12/2008
Test some more

Hey Andrew,

If this is your only data point, then test more. If, after 5 or 10 tests, you're seeing a pattern emerge, address it then. The additional data should give you a better idea how to fix it.

Along those lines, you might want to play without the random bonus mechanisms to get a better idea if your deterministic systems have a problem.

Based on what you said, it sounds like it would've been a close game if not for the bonus points. That suggests you're fine, but one data point isn't enough to make any conclusions.

Good luck!


Joined: 01/17/2011
Positive feedback loop

I agree with the others who say playtest a lot more. You shouldn't form any solid conclusions from a single playtest. Nevertheless, there is some static analysis you can do on your game.

The runaway leader problem is normally created by a positive feedback loop. An example of a positive feedback loop is investing in fixed interest bonds -- the more money you invest the more money you earn, which means you have even more money to invest next time. If this "fixed interest bonds" was a game, it would be rather boring because you know exactly who is going to have the most money at the end of the game without even playing it.

Random variation does not remove the positive feedback loop, it only obfuscates it. Let's say you replace "fixed interest" with "variable interest". At the end of each investment period you earn 1d10 percent interest. Now the winner is unpredictable because the person with the most starting money could roll really bad, but that doesn't mean that your positive feedback loop has gone away. You've just hidden it behind some dice rolls.

There is not enough information on your game to determine whether there is a positive feedback loop or not. Random points on the board are probably not a positive feedback loop (unless they spawn in front of the players on a single track -- which would mean that the person in the lead would always get to them first). The spell cards you describe are probably not a positive feedback loop either, it's just a matter of balancing their cost vs effectiveness.

Some individual cards could create a positive feedback depending on what their effects are. Take for example a card which costs 3 gold and lets you play it for 5 gold. While you may not have anything so blatantly obvious, check the individual card effects for feedback loops and decide whether you want them or not.

Positive feedback loops are not always bad -- it depends on what the winning condition is for your game. For example, Monopoly is premised a positive feedback loop -- the more money you have the more properties and houses you buy, which earns you even more money. Since the winning condition of Monopoly is to bankrupt the other players, the positive feedback loop is necessary to bring the game to a conclusion otherwise it would drag on indefinitely (I mean, even more than it already does).


Horatio252's picture
Joined: 03/13/2011
Negative Feedback Loop

I agree with Kos. Yes, playtest it more. If the runaway leader problem persists, then consider adding negative feedback loops for the leader. A negative feedback loop will mean that the leader will have to be exponentially better to gain a linear amount of points. It should take three or four times as much skill to beat me by 10 points as by 5 points.

Some examples of negative feedback include Dominion and Power Grid (Power Grid might be the poster child in my mind). In Dominion the more victory point cards you have in your deck, the less efficient each of your hands is (instead of drawing useful cards, you draw lots of unusable victory point cards). There are several negative feedback loops in Power Grid but a couple are that the leader gets fuel last (when it is most expensive) and the more cities a person powers, the less they gain in income per city ( powering 4 cities gains you less than twice as much money as if you powered only 2).

Think about how you might do similar things in your game to slow down the leader.

GamingNerd's picture
Joined: 03/26/2012
Thanks everyone! This was

Thanks everyone! This was great advice.

We've play tested a half dozen time now, with another play test tonight with 6 people. This play was also tainted, as JHouse observed, because we did intentionally play with one person using as many spells as possible and one person using none. In typical play there's more of a balance, but we wanted to test the extremes.

I don't think we have a positive feedback loop in effect here, but I'm definitely going to give some thought to if there needs to be a negative feedback loop. One thing is that it costs 2 tokens to buy a spell, which may be useful or may just block other spells, but at the end of the game unspent tokens are worth 1 point each.

One thing I think I'll change based on these comments is the bonus points. Right now they come out randomly, and can repeat their location. This led to a situation in the game where twice they happened to come out next to the leader, in locations they'd already been. I think limiting them so they can't come back out in the same location twice might be a smarter approach.

In addition I think we're going to give each player 1 spell card at the start of the game, randomly. Any other spells will have to be purchased, with the hopes of giving someone who goes for tokens over spells a bit of a boost. We'll see how it goes in the play test tonight. I also have one person playing who has never played it before, so I'm really excited.

Thanks again for the great comments.

- Andrew

GamingNerd's picture
Joined: 03/26/2012
Well it happened again. In

Well it happened again. In the 6 player game this same guy won by almost 20 pts. The interesting this is everyone else was grouped within 5 points.

We identified a few reasons he could have won by so much (game and non-game design related) and are making some modifications to the point system for next play. Some simple changes, left over tokens are the end are worth half their value rounded down instead of 1 point (he had 14 left over which helped account for his lead, but wasn't entirely the cause) and spells are going to get more powerful to allow people who choose to go the spell route a bigger advantage since they lose the bonus of the points at the end (spells cost 2 of the previously mentioned tokens).

Time to tweak then on to the next play test!

Joined: 07/14/2011
One other consideration for

One other consideration for your playtest would be to assign a different player to maximizing the spell buys. It's possible that the player that's creaming everone is already a really good games player. Switching the assignment to different players each time you playtest should level that potential effect out.

Joined: 11/25/2010
I agree with Tiberius

When I happen to have my friend Ira playtesting a game, I'm not at all concerned if he wins, because he always wins. (He once calculated his win-rate over a 6-month period in 4 player games to be "about 70%"). The question is, was it fun? Did everyone else feel like they had a reasonable playing experience? Is there something I can do to make it so that the same strategy Ira picked will not win every time?

GamingNerd's picture
Joined: 03/26/2012
What you both said holds some

What you both said holds some truth to it. The winner both times was the same guy, and he tends to win a lot of the games we play. Taking him out of it, the rest of the players were all within a 5 point spread of each other. We've got some more tweaking to do, and a couple more play tests along with it, but then we're inviting 6 people who've never played to test it out for us. I think that will help us figure out if this is a real problem, or if that jerk (I mean friend) is just that good at beating me.

- Andrew

Syndicate content

forum | by Dr. Radut