Changing the curve

[Aeson]

That said, I have no idea how the issue could be gotten around, because steam power/hospitals are not going to arrive at the same time in every game, and the discontinuity can therefore not be timed accurately.

The same problem exists with territory aquisition. Some playstyles slowly build up to overwhelming force, and then quickly take huge chunks of territory. Others slowly but surely keep expanding. Then there is everything inbetween.

The curve isn't going to be perfect. Generally if big gains have just been made in the game (population/territory) it will be best to delay victory for a while to take full advantage of those gains. That's due to the averaging of turn scores to get the final base score. If everything has been maxed out for some time, or there are no immediate plans for population/territory booming, then the Jason score should be pretty stable (+/- 5%)for as long as the player wants to play it out.

 

[Bamspeedy]

Ronald, the difference between what you would have got if you took the early win vs. the late win is only about 3-4%. You considered that alot, but I don't consider that to be very much. If you weren't managing happiness and laying down railroads/irrigating to increase your score while waiting for the later victory, that can easily account for 4% right there.

Yes, I realize that sometimes a 4% variance can cause someone to lose some rankings, but I sure don't want to go back to the Firaxis scoring system.

 

[cracker]

There are several things that we also need to keep clearly in perspective in this scoring process.

First, we do not talk about it much but the change in how the Points for the Global Ranking are figures plays spme role in this process. The Global Ranking points for a GOTM game are now figured based on the percentage of the high score in the game instead of the old method that used the average scor for the game. This keeps the ranking impact fairly consistent over time.

Second, regardless of slight differences in the individual scores for different victory conditions, the differences between the well played 2050 milked scores and the well played fast finish scores is now down into the 3%-6% range instead of the old system where we had 3 to 5 players scoring 20,000 points and all the other players scoring 6,000-10,000 points or less. There is almost a continuous distribution of victory scores and this means the system is clearly working.

Third, fastest finish of a certain type will always win an award. I am 100% comfortable with a scoring system that provides emphasize to game mechanics that reflect the ability to balance other issues such as territory and population in with the requirements of speed. The point that Aeson makes about the internal point surge profiles that accompany different playing styles in pursuit of different victory conditions is very important.

These are tough issues to balance and for players who choose to focus primarily on speed victories the challenge and risk that a player who is a few turns slower but perhaps several percentage points stronger could end up scoring higher in the game is an important feature that continues to provide incentive to improve while still playing and enjoying the game.

In the extreme implementation of the scoring and awards system, players ARE competing for two different objectives to maximize speed and simultaneously maximize score. As long as these two objectives are not inconsistently implemented it is great that the fastest finish may not necessarily be the highest scoring game of that victory type.

I am very comfortable that we will continue to see lots of exciting and enlightning aspects of game play as we continue to see the distribution of players of many skill levels across all different victory conditions in a single game.

 

[SirPleb]

I've submitted my GOTM19 and want to add some comments here. I'll be vague in some areas to avoid GOTM19 spoiler info.

I went for an early victory in GOTM19 (vs. milking to 2050) and did fairly well. In my game it turned out that:

1) At my finish date the new Jason curve gave me a higher score than the old curve.

2) With both the old and the new curve, my Jason score was increasing every turn toward the end of my game. I.e. I could have increased my final score by playing on for a while using either curve.

3) With the old curve my score was increasing faster (by about 50%) than with the new curve.

So I'd say the new curve is a substantial improvement. If I were to play on, it looks like at some point the two curves would meet and I'd have the same score either way. But the new curve has reduced the enticement to play on - it gave more of the reward up front and held back less which could be gained by delaying.

The reason my score was increasing in my final turns was as Aeson has described - I had made substantial territory/population gains near the end.

Although playing on would have increased my score a bit I'm entirely happy with the result, finishing with an earlier date is a separate goal to some degree, involving the tradeoffs Cracker has described. I don't think a perfect algorithm for every possibility exists. IMO the Jason score works remarkably well.

 

[ltcoljt]

I would prefer more information and less change and surprise from month to month.

Now, don't get me wrong, I appreiciate the efforts being made to make the game of the month a very special event. I recognize by the healthy number of participants that something is going right.

But I want to weigh in on something that bothers me. I don't have unlimited time to play, and when I play all I want is a level playing field. Right now the folks that are spending a lot of time dissecting the scoring system seem to have a clear advantage.

I don't want to have to do a math exercise when I play GOTM. I don't see why the scoring calculator to be used for a game can't be made available with the game package. Period.

There are certain aspects of Civ3 that are less than exciting in the first place, now why should I want to be caught up in "playing spreadsheet" to compete in the GOTM?

I don't really care how the curve is constructed as long as I have a simple way of being able to project "my" score based on what I am trying to accomplish in the game.

So don't play with us as if we are lab rats, giving us mystery map sizes and unknown "best dates"; nor give us a curveball one month and a screwball the next.

Give us the calculator with the save file and just let us play. All this drama over scoring is making us want to puke.

 

[Txurce]

ltcoljt, I would go one step further. Aeson has devised a scoring system that is imperfect but very balanced. It has passed the crucial test - gaining player credibility - with flying colors. I would prefer that he tinker with it in private, and not discuss its workings. Then the players could play the GOTM in whatever manner they consider optimal or most stimulating, rather than trying to squeeze a few extra points out of the current month's equation.

 

[Woody]

My two cents on this is that the way things are I probably will no longer compete in the GOTM. I started playing the GOTM only a few months ago for the purpose seeing how other people played Civ III and how I would do in comparison. Now with all the extra custom units and modifed scoring system it seems as if I am no longer playing a Civ III GOTM but instead I am playing a Mod of the month. I usually try to get in several games of Civ III in over the couse of the month and lately the GOTM is becoming almost a completely different game from my other tries.

 

[ltcoljt]

Originally posted by Txurce
ltcoljt, I would go one step further. Aeson has devised a scoring system that is imperfect but very balanced. It has passed the crucial test - gaining player credibility - with flying colors. I would prefer that he tinker with it in private, and not discuss its workings. Then the players could play the GOTM in whatever manner they consider optimal or most stimulating, rather than trying to squeeze a few extra points out of the current month's equation.

Yeah, but.

But some playing find it most stimulating to squeeze a few extra points out. So you have the situation that people are trying to extrapolate from the information they do have to determine what the map size and factors are, and what the best dates will be; then you have a change in the curve tossed in. Better to just distribute the calculator and that will tone down the weeping, and wailing and gnashing of teeth.

 

[cracker]

Originally posted by Woody
...My two cents on this is that the way things are I probably will no longer compete in the GOTM. ...

Woody, I hate to here you say this out of context when I look at the database and do not see where we have yet supported you well enough to get you up to the point of submitting a game for comparison.

There are many opportunities to participate in the game and some people play in the background and just observe what seems to be the frantic goings on of large pack of fanatics.

The QSC could be something you could participate in with only a few hours of commitment and that sub-event is uneffected by the scoring curve discussion.

For many people, the big picture of all the activities going on and the efforts that we make to include players in 4 different versions of the software can be really overwhelming. You may just want to play games that are just like the games you always play and there is no shame in that. The difficulty is that we have to separate the vast majority of the modification efforts that are done for you for free but are required to keep the games running in a way that makes it possible for you to compare your games to many other even if you never do submit a QSC game or a finished game or find individual players whom you enjoy sharing tales and experiences with.

Try to help us help you, by choosing which of the game events you would participate in and then making the commitmnet to do that in a way that lets you enjoy the games in your own way.

 

[Woody]

Dont get me wrong I think what your doing is great for the top of the pack players that have played in the GOTM for a long time and are looking for different challenges but in the process you are making it harder for newer/less experinced players to get invloved. As the GOTM becomes more different from "normal" Civ III it makes it less attractive for me to particpate.

As for taking part in the QSC that just does not fit my playing style at all. I like to build perfect cities slowy and not do very much conquest/expanding till the late middle ages when I go into a blitzkrieg type mode.