GOTM 2.0 Brainstorming

[Frederiksberg]

Feel free to point out what I have missed.

dV

Some ideas regarding new competitions derived from the basic GOTM has been presented. I put forward the idea of a (yearly) GOTM Championship and there has also been discussions on special challenge games, problems and QSC.

On the GOTM rating issue I have been thinking about the following model: Let's assume that a player each turn contributes a certain amount of skill, s, every turn and after T turns the accumulated skill exceeds the difficulty, D of the specific game and victory condition and results in a victory. This could be represented as a simple formula:

D=s*T

Given that all players are assigned a skill number (rating), s, the difficulty, D_est, of a GOTM can be estimated by averaging s*T over all players wining with a particular VC. Next step would then be to compute the skill performance, s_p, for each player using the formula once more, i.e.

s_p=D_est/T

Finally the new rating could be computed as something like this:

s=(1-x)*s+s_p*x

where x<1 is a forgetting factor weighting the most recent performance against previous ones.

There are, of course, a number of practical issues that still need to be solved:

1. The rating system must be able to handle lost games.
2. There should be some mechanism to ensure that in most cases it's better to submit a poor performance than none at all.
3. The rating system should be checked for inflation/deflation if possible.

Lost games might enable the calculation of a rating for the AI as well, which is interesting for comparison.

Several people have suggested different sorts of training games and it gave me the following idea: Why not turn in into a competition? A number of "coaches" could be found among GOTM award winners and the "students" who have signed up for the game would then be somehow evenly distributed among the coaches. Each coach and students would then be given a thread where they can discuss the training game starting screenshot (like pre-game discussion) and also whatever tricks the coach would want to teach. The students would then independently play and submit the game. Best student would be found as in a normal GOTM and best coach would be found by some average of the student performances.

 

[nokem]

There are plenty of long-running and ad-hoc training games in the Strategy and Tips section of the forum - such as the celebrated (Lonely Hearts Club). The Hall of Fame also provides a space for competitive play and hosts many challenge games. I certainly wouldn't want to discourage anyone from creating training games - but I think that might be a better place to host them.

The point of GOTM is that it's a game that happens every month. I think we'll be more productive if we focus on suggestions for improving the content of the existing games and displaying the results rather than increasing the number. I struggle to play them all as it is.

So Fred's ranking formula is a good idea (if it works - my eyes glazed over ) because it might be fairly easy to swap for or add to the current algorithm. I also like ideas that can be run by the community - like 'best spoiler' awards - without loading up the hardworking (although now more numerous) mods.

nokem (conspicuously failing to provide any new ideas of his own)

 

[Frederiksberg]

Maybe a good idea to repeat...

The ground rules for a brainstorming are:

1. Focus on quantity
2. Withhold criticism
3. Welcome unusual ideas
4. Combine and improve ideas

So please avoid posts that limit creativity. It will anyway be up to the staff to find out what is feasible and what is not.

Edit: And thanks to all who have shown interest and posted in this thread

 

[Deckhand]

Goals: encourage participation, minimize additional work for staff, increase fun level

I'd like to see more awards for newbie participants.
These would require comparision with records from past games:

First Timer - for all those submitting their first xOTM game.
Best Newbie - for highest score among first timers
I'm so 1337 - best score from someone that has not won this award before and has never won a gold/silver/bronze medal
FTW - first win at a level > a level where you submitted a loss
Movin' on Up - each time you win at a level > highest level previously won at

more comparison between submittals (as others have mentioned):
comparison graphs
tech paths
where/when 2nd & 3rd cities founded
key reasons for lost games

add optional field to submittal form: what I liked best about this map (encourage designers, provide positive feedback, let them know what we like)

 

[da_Vinci]

I think that Deckhand's ideas about additional "awards" are interesting ... perhaps more as "honorabe mentions" listed in the results thread, but not necessarily immortalized in the pantheon of heroes?

Because the number of them could get overwhelmingly large very fast, especially if the idea is extended to newbie speed (which is then VC-specific).

The "I'm so 1337" seems by definiton to be a restricted 4th place award ... Deckhand, was that your intent?

(If so, I like the name "tin cup" for just missing the medals with fine metals ... stir that tin cup with your wooden spoon ... )

"FTW" and "Movin' on Up" would likely have considerable overlap ... so maybe both not needed?

(Anyone else struck by the irony of reading Deckhand's wonderful list of new ideas, which is then closed with his signature? )

dV

 

[Deckhand]

dV, thanks for positive response

I think that Deckhand's ideas about additional "awards" are interesting ... perhaps more as "honorabe mentions" listed in the results thread, but not necessarily immortalized in the pantheon of heroes?

Absolutely. Like cow and ambulances. Fun and Encouragement that anyone can achieve.

The "I'm so 1337" seems by definiton to be a restricted 4th place award ... Deckhand, was that your intent?

Not exactly. It could theoretically go to the person with the worst (wining) game - if everyone else playing had already won it.

(If so, I like the name "tin cup" for just missing the medals with fine metals ... stir that tin cup with your wooden spoon ... )

As this is brainstorming, my award names aren't strong suggestions for the final award name, but meant to convey the award's meaning.

"FTW" and "Movin' on Up" would likely have considerable overlap ... so maybe both not needed?

Agree. Intent of FTW was to encourage people to submit lost (or abandoned?) games.

(Anyone else struck by the irony of reading Deckhand's wonderful list of new ideas, which is then closed with his signature? )

to paraphrase ms, "These posts are meant to be educational, amusing, and include multiple meaning or irony."

 

[da_Vinci]

Not exactly. It could theoretically go to the person with the worst (wining) game - if everyone else playing had already won it.

Indeed ... I only thought about where it would be for the first few games, but a hundred games later it would be likely to be awarded at the bottom, at least sometimes.

If it has a meaning or value that is so variable, is that a useful property for an award?

I am thinking maybe not ...

dV

 

[Deckhand]

If it has a meaning or value that is so variable, is that a useful property for an award?

hmmn, probably not.
The intent was to recognize good play/score from someone that is unlikely to get a medal. Now sounds similar to an Employee of the Month award, that initially goes to exceptional performance and then devolves to those that haven't won it before. OTOH, assuming new players are always joining the game, let's assume that after many many months just half the players are eligible for it. Then it indicates best of the non-elites and you have cracked the barrier and have joined the 1337s. There could be other criteria as well to keep it from being meaningless.

 

[dalamb]

If you want to have a "skill score" for someone losing a game, it has to be lower than the lowest score submitted by anybody else for any victory condition. So maybe use Frederiksberg's formula with some combination:

• Pick a "difficulty" much lower than the lowest difficulty of any of the other victory conditions.
• Pick a "time" much longer than the longest time actually submitted, or possibly much longer than the time victory would have been.

However, if that fits into a long-term average, I'd be very strongly tempted not to bother submitting at all and wait for the next game I'd actually won. OTOH I rarely submit games anyway; I think we need to look at is what happens to the score of someone who keeps losing games but submits anyway. If I win consistently on Warlord and about half the time on Noble, but submit losing games for all difficulties above that, does my long-term score keep declining from all those high-difficulty losses?

 

[da_Vinci]

hmmn, probably not.
The intent was to recognize good play/score from someone that is unlikely to get a medal. Now sounds similar to an Employee of the Month award, that initially goes to exceptional performance and then devolves to those that haven't won it before. OTOH, assuming new players are always joining the game, let's assume that after many many months just half the players are eligible for it. Then it indicates best of the non-elites and you have cracked the barrier and have joined the 1337s. There could be other criteria as well to keep it from being meaningless.

Maybe the idea would be to set a score threshold ... highest score above the threshold in that particular game (for instance, I am still looking for my first 100,000 score) who has not won it before.

Maybe we are really talking about some personal best recognition ... the 50,000 club, the 100,000 club, etc. ... which would not be limited to one person per game.

ON the speed side, could have 18th century diplo club, 17th century diplo club, BC conquest club, etc.

And am I the only person who does not understand why "1337"?

dV

 

[Niklas]

1337

 

[Mundungu]

For scoring and motivation of less proficient players, how about an extra scoring system / award with handicap similar to Golf ?

This would however force the staff to keep track of a player ranking...

I am not sure how to make a formula that would work though...

 

[da_Vinci]

1337

Ah, the old guy is showing his age ... a whole language I am not aware of! And a whole library of articles about it ...

The rating modeling is interesting ... we have the situation where the score mongers (or medal chasers if you will) actually reach the brink of victory, then milk the game before tripping a victory condition. So a time to reach D_est model does not reflect skill in a game played for score?

Maybe we need a second model, for score ... where rating would be related to score adjusted for turns spent, by some formula?

Then, performance is a game is the better of the two systems?

dV

 

[Frederiksberg]

For scoring and motivation of less proficient players, how about an extra scoring system / award with handicap similar to Golf ?

A rating system similar to the one proposed earlier in this thread would automatically enable a handicap calculation since:

s1*T1=D=s2*T2

It means that the number of turns, T1, player 1 should use if player 2 uses T2 turns is:

T1=s2/s1 * T2

So the ratio of the rating (skill) numbers, s1,s2 is also the ratio of the number of turns spent. An award for the player who exceeds his skill level by the most points is another similar idea.

The rating modeling is interesting ... we have the situation where the score mongers (or medal chasers if you will) actually reach the brink of victory, then milk the game before tripping a victory condition. So a time to reach D_est model does not reflect skill in a game played for score?

Maybe we need a second model, for score ... where rating would be related to score adjusted for turns spent, by some formula?

Then, performance is a game is the better of the two systems?

Including score in the rating system looks a bit messy. The problem is how to determine the objective of the player: Speed or score. Anyway the rating could be coined "speed rating". Players going for score would then have to live with lower rating and they would have medals as a consolation.

 

[da_Vinci]

Including score in the rating system looks a bit messy. The problem is how to determine the objective of the player: Speed or score. Anyway the rating could be coined "speed rating". Players going for score would then have to live with lower rating and they would have medals as a consolation.

I would not try to blend speed and score in one formula. Instead, I would suggest a speed performance rating for the game, and a separately calculated score performance rating for a game.

A player would have the greater of the speed or score performance rating for the game be that games contribution to their overall rating.

Or we could have speed ratings, score ratings, and combined ratings, not to unlike the fact that we have three catetories of global rankings.

Hmm ... which makes me think of a potentially simple thing we could do to add a level of competitiveness:

The current global rankings reduce a game's contribution with age, and sum over the last 9 brackets ... clearly rewarding recent participation.

Maybe we should have a ranking that is calculated based on a players best 5 (or 10, or whatever) unaged performances ever. Kind of like immortalizing that one season in baseball where you batted .400 or had 60 home runs, even if your life average or recent performance is not that good.

It would also allow comparing currently active players to past stars that may not be active currently, bringing a history to the game (Babe Ruth vs. Roger Maris vs. Hank Aaron ... we will skip the juiced boys of summer ... as the baseball analogy).

Could call this the personal best rankings?

dV

 

[Frederiksberg]

I would not try to blend speed and score in one formula. Instead, I would suggest a speed performance rating for the game, and a separately calculated score performance rating for a game.

A player would have the greater of the speed or score performance rating for the game be that games contribution to their overall rating.

Or we could have speed ratings, score ratings, and combined ratings, not to unlike the fact that we have three catetories of global rankings.

Separate speed and score ratings should be doable, it's only when you want to blend them it starts getting complicated...

Anyway, the system doesn't have to be perfect. I think the main virtue of a rating system is that it gives all players the goal of increasing rating and this can be a factor motivating people to play the GOTMs.

 

[GoodGame]

Unusual ideas: Scenarios that play against the trait strengths (isolation, surrounded by aggressive opponents, your UU countered by neighbor UU). Start with AI ally/vassal. Start with an overstretched empire.

 

[da_Vinci]

@ GoodGame: That signature is almost the leading causes of death in rank order (at least for developed countries) ... do you deal with that for a living?

dV

 

[MarkM]

I would not try to blend speed and score in one formula. Instead, I would suggest a speed performance rating for the game, and a separately calculated score performance rating for a game.

A player would have the greater of the speed or score performance rating for the game be that games contribution to their overall rating.

Or we could have speed ratings, score ratings, and combined ratings, not to unlike the fact that we have three catetories of global rankings.

I haven't really followed this thread (was away for most of it) but I just have to say that I will probably go to my grave never understanding this obsession with speed, and why even in the current ranking display we have a ranking based on how many minutes you spent with the game running. I pay so little attention to those rankings that I just leave my game running during dinner, when I get a phone call, or even overnight on a weekend. When I used to play when travelling a lot, using a laptop, I think I would just suspend the machine with the game running for days at a time.

Do people really save and exit their game every time they get a phone call? Those of you that play chess (another turn based game) really think that the world chess federation ranking system is incomplete because it doesn't keep track of how much of the chess clock the player uses? Shuld I be deemed to have played a "better" chess game than an opponent that beat me, head-to-head, because I made my moves much quicker than the relative chess piece margin that I lost by? Really? Can't you just go play any other 3x game on the marekt in the last 10 years (e.g. Age of Empires) if you want emphasis on how fast you play. Why this compulsion to use the score/ranking system to compel people feel they have to play turn-based civ like an RTS when there is no logical reason to in the design of the game itself??? I don't get it.

 

[AlanH]

I haven't really followed this thread (was away for most of it) but I just have to say that I will probably go to my grave never understanding this obsession with speed, and why even in the current ranking display we have a ranking based on how many minutes you spent with the game running.

Where do we rank people based on the time the game runs? We quote the time played, as there is no reason to withhold it from our more statistics-obsessed players. But we don't rank it!