You know in a game when one of the players takes the lead that there's no way anybody can catch up, and the game will necessarily end with this player winning.

What is the term for having an end game that can turn around at any moment despite a player having a lead?

aganm said:
What is the term for having an end game that can turn around at any moment despite a player having a lead?

“Still winnable”

Part 2, can you give me some hints to make a “still winnable” end game? :D

aganm said:
What is the term for having an end game that can turn around at any moment despite a player having a lead?

https://www.giantbomb.com/rubber-band-ai/3015-35/

I've always heard this referred to as “comeback mechanics” in the MOBA space (for example, in HotS the team that is behind on levels gets an XP multiplier on enemy takedowns).

Maybe the game should end at that point? ie don't have a game design that goes on longer than necessary once a skillful winner has demonstrated enough skill to win?

You have to be careful with overdoing the “come back” mechanism as it discourages skillful play by the winner, as it will all be for nothing if the less skillful player gets powerups or some advantage for playing badly and can threaten to beat the better player. Imagine how that would feel…

Thanks for the insights

Part 2, can you give me some hints to make a “still winnable” end game? :D

Randomness versus skill

The more random your game's success is, the less early advantage can project to victory. A drastic example would be a game mechanic that switches places with random players. The other extreme is a game where success is only determined by players´ abilities. So the best player will start winning, continue winning, and end winning. Boring.

You decide how much randomness works in your game. That is no easy task. You don´t want good players to lose regularly because of random events. Nobody likes that.

Catch-Up and Wall-breaker

Catch-Up mechanics support weaker players in reaching the front line. This can be the distribution of speed buffs to the last players in a racing game. This can also be crafting mechanics in an MMORPG where high-level players sell high-level items to low-level players.

Wall-breaker mechanics enforce a higher skill on players that advance from the rest. This can be a slow debuff that only applies to the first player in a racing game. This can be a boss that gets weaker within weeks and months in an MMORPG.

The bomb

There is a type of game where players throw a “bomb” around or pierce swords in a barrel with a pirate. The longer the game goes, the higher the possibility of explosions / the pirate jumping out. This game combines randomness and skill because skilled players will avoid having the bomb or taking action the longer the game goes.

More general, the bomb is an external force whose impact decides or strongly influences the game. The bomb can also be an alien force or the king. Players can work out strategies that put them in favor. So until the end, the game is undecided.

Abort

In a competitive setting, you can have an abort mechanic. The losing player can give up. So you avoid boring games that are already decided.

Cut ahead

In racing, the winning driver can lap other drivers. This effectively puts the winner at the end of the queue again. A similar mechanic is found in some games. When players reach the highest rank, they gain a reward and restart with the lowest rank. So this mechanics brings top players to the bottom of the ladder, making them more interesting.

When you hint at the type of game you have in mind, I can develop more specific solutions.

Best wishes!

Making a game “still winnable” up until the end is trivial. Just flip a coin at the end of the game to determine the winner. Or determine the winner in some other way that ignores basically everything that has been going on in the game up to this point.

Making a game “still winnable” up until the end while also allowing meaningful progress is harder, because “meaningful progress” needs to somehow improve your chances of winning. It's still fairly easy though. Imagine a series of 19 skill contests. Normally, you'd award the victory to the first player to win 10 contests, making any remaining contests pointless. But lets say we just count victories. Player 1 gets 18 victories, then messes up the last contest so that Player 2 gets 1 victory. Now, instead of counting who has more victories, we randomly pick one of the contests and only count that one contest. Now Player 1 has a 18 in 19 chance of winning the game and Player 2 has a 1 in 19 chance.

However, if you want to make a that is “still winnable” up to the end, allows meaningful, and isn't completely arbitrary in how it assigns victory, the you're probably out of luck. Because one of the players will be more skilled than the other player, and that player is almost certainly going to win, especially if they already have the lead.

a light breeze said:
Because one of the players will be more skilled than the other player, and that player is almost certainly going to win, especially if they already have the lead.

In a game of pure skill among flawless participants, yes. But in practice players are fallible, so what your standard leader-always-wins system actually works out to, is a system where the first person to make a mistake loses.

Comeback mechanics aren't designed to erase the lead, they are only meant to give the trailing team the means to capitalise when the leading team makes a mistake themselves.