We've seen more than one poster suggest that the permanent Hall of Fame modify the Civ4 scoring algorithm to something of our own choosing. As these posts are rather scattered but persistent, let's have the concentrated discussion in this thread.

First off, should we? And if so, modify it how? (i.e. post your proposed formula).

Major drawbacks in firaxis scoring system for me :
- Early victory dates dont supply enough bonus points
- Population influences the score to much

Domination:no change
Milking: no change
Conquest: Set your land score to 100%.
Other: Add every surviving opponent's land and population to your base score.

There wasn't a choice of other so i picked no...

I don't think Firaxis has it right - now. I think we should give them a chance to correct it in the future.

As far as a modded version of scoring, I think victory type should be a bigger factor in relation to time (ie - conquest and dom should have a decending curve mod, while ss, diplo, and ??? should have BIGGER decending curves for a potential of bonus(and should last stretch out longer). And of course - a time victory should have no bonus at all). and reduce pop impact on score.

Definately need to mod it or plain out ignore it.

The fact that a domination victory gets huge land and population scores for doing what is inherent to the game variant only means that domination is the "preferred" victory. Even though Civ3 gotm lived by that standard for 2 or more years I've always found it very cumbersome and boring. This is less the case in Civ4 where fast expansion actually has elements of skill beyond pure numerical and micromanagerial so its a worthy goal, its just not at all balanced with any other.

I think there is a different curve for each victory type, but its not sufficient to equate any two games with different victory conditions.

But, how do you compare two games? Personally I'd just go with the fastest date for all victory conditions and then have a Firaxis score table as well with them all mixed. But I wouldn't highlight the builtin scoring system by having that as default table shown. I would show the top 3 from each victory(except Time, boo!) and then having fuller lists for each condition.

The scenario I would like to avoid is purposefully delaying a win by milking. To me this isn't playing Civ at all, its playing the scoring system. So, what do you wish to promote, Civ and its varied strategies, or a flawed scoring system?



Anyway, having fastest finish and a derived scoring system for each victory type might be a worthwhile approach. But, gathering the data to score each victory is gonna take a bit of time.

Domination:no change
Milking: no change
Conquest: Set your land score to 100%.
Other: Add every surviving opponent's land and population to your base score.
I dont understand?
Score doesn't matter for fastest finishes only for game actually played for score in which case the victory type doesnt matter no?

I basically agree with Smirk.
I don't really care about score, for me its more the fastest finish.
People can still play for score using the current or a to be designed scoring system, but it should not be emphasized, specially if we keep the current system, as Smirk pointed out.

Originally Posted by DaveMcW
Domination:no change
Milking: no change
Conquest: Set your land score to 100%.
Other: Add every surviving opponent's land and population to your base score.

I don't like this. I think your score would increase the later you finished. Since you'd effectively have 100% of the land and all those cities would be growing until the end of the game. I don't want to see 2000AD space race victories scoring higher than 1500AD ones.

I think that for conquest victories should take into account what score your opponents were at before you started to war them. Warring a civ with a score of a couple thousand is harder than warring a civ with only a couple hundred, regardless of how advanced you are.

I also think there need to be a larger link between victory type, map size, and skill level.

I don't think Firaxis got it right, but I recommend to go along with Firaxis. The HoF should be a place for people from around the world to get an idea of how their games in compare to some of the top the score. Without ever having to play anything specific for the HoF, they would still get an idea how their game in compare to some of the top games. If they have to play a certain mod in order to know where they are standing, there is really no point to it. May well call it the "Hall of High Score Mod Game".;)

Well, HOF is mostly about 2 things: Fast finishes (FF) and high scores (HS). The FF part of it obviously doesn't need any score modifications, because we're trying to finish asap and not have the best score.

In the HS part, however, score does matter. I'm not a milker myself, but I know several people like to play a very long game and get a score that is worth it. My favourite option is to keep the system as it is, but if you don't like it, a good idea would be using the avreage score, just like in civ3.

I also voted to keep Firaxis's scoring system. I don't think they implemented the scoring system for civ4 correctly, but I don't think it has any impact on the Hall of Fame. Generally people will either go for a high score or a fast victory with any given game. So why does it matter if that fast victory scored highly or not?

I'm rather surprised by the opinion expressed by a few to keep Firaxis score system. I thought it was pretty much agreed on that it favours population way too strongly; doesn't take into account map size correctly and is in essence pretty simple compared to what a modded score system could take into account.

Or am I missing something here?

The reason is I haven't seen an alternative scoring system that does a better job. Everyone has different opinions of what should go into a high score. I don't think we'll ever all agree on one. If we changed the scoring system to something else you would probably have an equal amount of people saying that it's not a good scoring system. At least with Firaxis's you don't have to try and defend it.

Good point Shillen.

The more I think about it, creating a new scoring system is really just changing the scoring to suit the way you want the game to be played. I think that before you change the scoring system we should try and get the rules worked out a little better. If we change the number of opponents per map size, and figure out the other settings, then reset the HOF, it may change the high scores a little bit.

Although the only thing that I think would be a huge point on the side of changing the scoring system, is to weight different map sizes differently than it currently is. Beating deity on duel with 1 opp is a lot easier than beating deity on standard with 6 opps, however the smaller map will be a way faster finish and result in a higher score. I think the smaller the map the less impact a quick victory should have on a score. ie, winning a conquest on diety, duel, 1 opponent in 2000 BC should score the same as beating diety, standard, 6 opponents in 1000 BC. (or something equivalent)

Shillen:
I agree with not having seen a better system and everyone having different opinions on what should go into a high score. I'm not (yet) convinced though that there isn't a system possible that would have a substantial higher number of people that agree with how it scores.

Currently, there seem to be a number of issues with the current system:
- Population being a too big influence on how a game is scored.
- Space / Culture / Diplomatic victories seem to score substantiantially lower.
(note that these two issues are definitely related)
- Map size not being enough of an influence.

The first important question is whether or not these are flaws. Ergo: Is it desirable that the scoring system works in this way or is it something that we'd want fixed by using a different scoring something? (no, I don't know the answers myself ;))

Secondly, we'd have to define the so called 'better' scoring system before polling whether or not to switch from the old. Now the poll is pretty much measuring whether or not people are satisfied with the old system (which isn't necessarily bad).

Note: Using the huge amount of statistical data from the GOTM games might be useful for defining a good scoring method. With 600+ games of statistical data per month (on the same map!) there's a wealth of information to use as input for other formula's.

I definitely agree though with a good alternative being a necessity for changing from the current system.

The more I think about it, creating a new scoring system is really just changing the scoring to suit the way you want the game to be played.

This makes me wonder what we consider a good scoring system. To me it'd seem logical that if 7 equally skilled people would play the same map, with 7 different victory conditions it'd be 'a good system' if they'd score similar points for their victories (assuming equal amounts of luck, lol).

If that's the criterium for a good scoring system, the current system is most likely not optimal.

Note: There's definitely a case to be made for simply waiting a few more months to see what Firaxis will do with it's next few patches.

This makes me wonder what we consider a good scoring system. To me it'd seem logical that if 7 equally skilled people would play the same map, with 7 different victory conditions it'd be 'a good system' if they'd score similar points for their victories (assuming equal amounts of luck, lol).

I don't understand why this is needed for the HoF. I understand if you were arguing a GOTM scoring system, but why do different victory conditions need to score equally well for the HoF?

I don't understand why this is needed for the HoF. I understand if you were arguing a GOTM scoring system, but why do different victory conditions need to score equally well for the HoF?

Exactly. If that isn't important for the HOF, it isn't important that i.e. space races score less than domination. Thereby making that particularly issue not something that's important for the HOF. --> No change necessary on that account.

Playing devil's advocate, reasons I can think of:
- It'd be useful for the 'score' condition in the HOF tables. Currently that score list is pretty much dominated by conquest / domination (although I do see more other types than expected).
- It'd make sense for the HOF & GOTM to use the same scoring system.
- The score should reflect the skill of the player, not so much the type of victory he is pursuing.

I'm rather surprised by the opinion expressed by a few to keep Firaxis score system. I thought it was pretty much agreed on that it favours population way too strongly; doesn't take into account map size correctly and is in essence pretty simple compared to what a modded score system could take into account.

Or am I missing something here?

for me personally - I agree that it favors pop too much... I am hopefull that they will change the formula - does anyone know if it is on there list of things to do?

if it is definitly not - then yeah lets mod pop value down and give extra bonus to non-con/dom early victories

I voted to leave it with firaxis system for HOF. The Hall of Fame is to try to det records, and see how good you are. It is impossible to balance all the victory conditions evenly in terms of difficulty and scoring. The only difference that would make since to me is to make map size more of a factor.

Exactly. If that isn't important for the HOF, it isn't important that i.e. space races score less than domination. Thereby making that particularly issue not something that's important for the HOF. --> No change necessary on that account.

Playing devil's advocate, reasons I can think of:
- It'd be useful for the 'score' condition in the HOF tables. Currently that score list is pretty much dominated by conquest / domination (although I do see more other types than expected).
- It'd make sense for the HOF & GOTM to use the same scoring system.
- The score should reflect the skill of the player, not so much the type of victory he is pursuing.

Most of these conditions were present in Civ 3 and werent very important.
You have to understand that fast finish and score are completly different.
If I play for culture, I don't care what my score is.
The way to get a good score in civ 3 was fast almost domination and then milking.
Now it's different but the point is still that score is its own game.

Now if somebody argues that the way score is done makes for bad score games sure, but comparing score of different fast victory types is besides the point IMHO.

Also I disagree about Gotm Hof comparison.
In GOTM you are compared MOSTLY my score (sure fastest finish get an award and all but the point is the main criterion for placement is score).
In HOF that is not true at all except for score games.

I don't understand why this is needed for the HoF. I understand if you were arguing a GOTM scoring system, but why do different victory conditions need to score equally well for the HoF?

Because if they do not then its no longer a highest Civ score HOF, its a highest domination score HOF. Where domination may vary based on diff and map size, but on in a minor way like with Civ3 (conquest or milk).

Well that's kind of my point. Civ3 was always a milking high score also. But that's why there are the fastest finish tables. A well-played spaceship or cultural game still gets to the top of the table. If I'm looking at the Hall of Fame and I want to see who is the best at spaceship victories I'm not going to sort by score. I don't see why a good spaceship victory has to be at both the top of the fastest finish table and at the top of the score table. A skilled domination victory doesn't end up on the top of the score table either. Only a milked game will.

Because if they do not then its no longer a highest Civ score HOF, its a highest domination score HOF.
It's not actually called the "High Score Hall of Fame" anymore, we changed it to "Hall of Fame" quite a while back. Someone suggested that maybe we shouldn't default to score on the main page, and maybe they are right!

Most of these conditions were present in Civ 3 and werent very important.

Why were they not considered important? Because there's good reason to or because people just accepted the negative aspects of the scoring method because there was nothing better?

Well that's kind of my point. Civ3 was always a milking high score also. But that's why there are the fastest finish tables. A well-played spaceship or cultural game still gets to the top of the table. If I'm looking at the Hall of Fame and I want to see who is the best at spaceship victories I'm not going to sort by score. I don't see why a good spaceship victory has to be at both the top of the fastest finish table and at the top of the score table. A skilled domination victory doesn't end up on the top of the score table either. Only a milked game will.

Let me put it in a different way: If it were possible to make a scoring method that would make it possible to compare equally skilled space & conquest games, why wouldn't we want that? (note: in both cases a milked game would probably still score substantially higher)

It's not actually called the "High Score Hall of Fame" anymore, we changed it to "Hall of Fame" quite a while back. Someone suggested that maybe we shouldn't default to score on the main page, and maybe they are right!

Interestingly enough, right now the "Hall of Fame" pretty much is the "High Fast Finish Hall of Fame". Six of the seven choices in the drop down list are oriented towards fast finishes, while only one is oriented towards score. Till yesterday I had actually assumed all of them were default sorted by score as that seems more logical for a hall of fame.

<offtopic, interesting idea though>
It seems to make sense to have two dropdown lists instead of one, one for selecting the victory type and one for fastest finish / score. Right now, score doesn't really seem to belong together with the victory type's.

This would also have the advantage that you could easily find out which game had the highest score for a specific victory type. Which isn't possible right now.

Disadvantage seems to be that there'd be 12 hall of fame's instead of the current 7. Although that may not be bad? ;)
</offtopic, interesting idea though>

Why were they not considered important? Because there's good reason to or because people just accepted the negative aspects of the scoring method because there was nothing better?



Let me put it in a different way: If it were possible to make a scoring method that would make it possible to compare equally skilled space & conquest games, why wouldn't we want that? (note: in both cases a milked game would probably still score substantially higher)

This is not a practical problem.
It's a problem of defintion.
What if I want to use fastest victory as "skill".
Then we have fastest finish.
Implied in any defintion of score is an idea of what a "good" game should be.


I dont even understand how you know a milked game would score higher that a conquest game using a not-yet-defined imaginary scoring scheme?
Certain scoring schemes are not milkable, for examples those that rely VERY heavily on early date....


Scoring systems don't have "negative aspects". It's stronger than that. They force a certain playstyle on people playing for score.
Defining a scoring system is implictly defining a way to play the game if you want to play for score (quick dom and milking in civ 3 etc...)

Since we will not come anything close to a consensurson such a topic I vote to use the Firaxis system for score, with maybe putting a little bit of pressure on them to improve it a bit if possible, as this gives legitimacy, people can be impressed even those that don't play the HOF that much (just like ppl were with SirPleb's and Moonsinger game because it ws 70 and 80K firaxis points not, 358 HOF points).

I vote "No". I prefer the focus on fast finishes, and keeping scoring simple.

It would be neat if the fast wins and score tables were limited to one entry per civilization. The fastest Indian, Incan, and Aztec conquest wins are quite different. As it stands, you should probably choose Incan.

It's not actually called the "High Score Hall of Fame" anymore, we changed it to "Hall of Fame" quite a while back. Someone suggested that maybe we shouldn't default to score on the main page, and maybe they are right!


You're talking semantics but I'm not. "Because if they do not then its no longer a highest Civ score HOF, its a highest domination score HOF."

Are we playing Civ, or Civ domination, thats my point.


And that someone was me, also maybe LulThyme said so as well.

You're talking semantics but I'm not. "Because if they do not then its no longer a highest Civ score HOF, its a highest domination score HOF."

Are we playing Civ, or Civ domination, thats my point.

What I was trying to point out is that we changed the name because there is more to the HOF than score. Score is the default filter on the main page for historical reasons, not because it's the most important. For the other victory conditions score isn't really that important. There's certainly no need to get close to the domination limit when going for another victory condition just to get a high score.

Can we change the default filter on the main page? Maybe display the best 1 or 2 games from each victory condition.

There's certainly no need to get close to the domination limit when going for another victory condition just to get a high score.

I'm talking explicitly about the score list not the fastest finishes.

In support of a modded score that list would be able to compare divergent strategies and victories. Its safe to assume at this point that one particular victory type and strategy condition will rise to the top of all score based lists, with no real statement about the quality of the game and skill of the player. That is, other than that particular victory type.


Now this wouldn't be an issue if it weren't the list thats held high above all others and shown as default. But it is, thus its the first impression and assumed to be what HOF is all about.

Can we change the default filter on the main page? Maybe display the best 1 or 2 games from each victory condition.
We're actually debating that in the Rabbithole now...

We're actually debating that in the Rabbithole now...

Be wary of the resulting page becoming too complex. With 7 victory conditions and 9 difficulty levels you'd get 63 distinct tables on one page. That's probably too much, unless the player distinctly 'asks' for that amount of information.

Some ideas:
- Not loading any victory condition by default. Instead offer a blank 'search' page with choice controls for all of the options. And a search button to actually go to the page the person wants to go to. You'd get roughly 15 choice buttons on a very simple start page. And on the resulting results page you'd get exactly what you want to see.
- Use check boxes instead of choice controls. That way people can select multiple conditions & multiple difficulties at once; greatly increasing the amount of game results they get (and the amount of flexibility they have).

Might be useful to make a separate thread for this discussion though, pretty off-topic ;).

Note: What's the rabbit hole?

Note: What's the rabbit hole?
A pub Dianthus and I hang out in a lot. ;)

Can we change the default filter on the main page? Maybe display the best 1 or 2 games from each victory condition.
Main page has now changed. It took us a while to decide what to change it to, but we (the HOF staff) think this is pretty good :).

Like the new main page. Easier for me to tell how I am stacking up based on map sizes, difficulty or whatever else I want to compare to.

Thanks much.:goodjob: