Currently my game is single-player only. I've experimented with multiplayer, but my focus has been single-player and other modes are unpolished and probably won't make the release.

I've been considering adding a time-trial leaderboard to add a competitive aspect to the single-player mode and increase replay value. 

However, I built both a first-person and third-person mode. Both are excellent and fun, but it seems unfair to have first-person players compete against third-person, so I'd be looking at two separate leaderboards (or 6 leaderboards if I go with 3 difficulty levels). 

Further, if I separate these modes from a leaderboard standpoint, then I can't allow a player to switch mid-game. I use a checkpoint system for save games and don't currently support multiple save slots, so the only way to play the other mode would be to start over.

I'm also a little worried about cheating on the leaderboard, I don't really think that there's anything within reason I could do to prevent it, though I'd be lucky if my game was popular enough to have that problem.

Do you think a time trial leaderboard is worth the trouble? Is it enough to just have a single-player game that's just fun to play without a competitive reason to replay?

Here's some videos of both modes to add context:

I would renew leaderboards on a weekly basis so cheating has a far smaller effect. Also in my experience the fewer leaderboards the better. Having lots means that players find it difficult to track and lose interest. Having maybe 3/4 for the whole game will keep players really engaged in improving there scores (and weekly resets are a good way to get players to keep coming back!)

On 12/25/2017 at 10:11 AM, Lithic said:

Do you think a time trial leaderboard is worth the trouble? Is it enough to just have a single-player game that's just fun to play without a competitive reason to replay?

In single player with a private best time, there are some players who love it.  Observe online all the recordings of speed runs for various games.

For online games it is difficult to monitor and correct for cheating. They may range from someone discovering an exploit for the game to someone reverse engineering the scoring protocol or intercepting and modifying the value being sent to the servers.  The risk of damage from cheaters goes up as the number of players and popularity increases, since for many people the notoriety of being artificially top of the leaderboard is their own reward.  On the flip side, it may be very difficult to detect when someone has cheated versus submitting an extremely skilled speed run. 

The cost and effort of detecting the problem depend on the game. A relatively simple protection is having the game recording a replay file through the game and transmitting that with the score, but even those are subject to tampering.

Weekly/Monthly leaderboard refreshes, like Dave Fooks mentioned above, are probably the easiest and safest way to discourage cheating (though it's going to happen anyway). What if players were able to view the playthrough?

19 minutes ago, CurryTime said:

What if players were able to view the playthrough?

Interesting thought. While video recording would probably impose too much overhead on the framerate, it would be cheap to record positions, directions, and velocities of game objects and allow for a replay.

Interpolating the recorded data and drawing with the game engine isn't much different than interpolating multiplayer data, it's doable but a fair amount of work to allow replay.

Everyone's feedback seems to validate that it's worth adding this feature. For initial release I'll probably just enforce with refreshes, but maybe come back later and add replay support.

For Cheating, look at GTA V and see how every race win record is 10 seconds which is impossible. Rockstar seem unable to fix it. Now it's pointless to bother even trying so you definitely need anti-cheat if you're taking scores online, player V player. The replay idea is probably the best way to track it and you only need check them after the game has been out for a while, reducing the workload because as the game is released, plenty of players will try for the leaderboard which will push the time completed down to a point where a new "winner" is quite rare and easy to check for cheating.

On 12/31/2017 at 8:17 AM, LorenzoGatti said:

Why? Shouldn't you have a single leaderboard for "view switched as the player sees fit"? If switching view between first person and third person is a meaningful tactical choice, there doesn't seem to be any reason to forbid it; if either view is strictly better, it will be used almost exclusively in the good games in the leaderboard, but it isn't a problem.

The only role I see for forbidding view switching is in multiplayer games where some or all players agree to use the bad view, respectively as an handicap or as an increase of difficulty.

In third person you have a bigger field of view and you can better see where shots are coming from when you're getting shot at from behind. But really the advantage doesn't seem significant and aiming is easier in first person. I think you're right, doing leaderboards only by difficulty level seems to be the way to go.