Global Ranking by Victory Condition

[Mark Cutt]

I win 95% of my games with domination or conquest. I decided it was time to learn the secrets of Diplomatic or Cultural victories and I started wondering which are the best and more expert players for those winning condition.

To answer the question, I took the most recent GOTM and for each of them, instead of having a single ranking, I extracted one ranking for each victory condition. Then I combined the scores in the same way Aeson did for the Global Ranking (i.e. normalize with the best score for the relevant winning condition, apply the game weight depending on its age and difficulty level and sum all the resulting score for each different victory condition).

As for the global one the resulting ranking are a combination of both good performance and regular play. Of course by splitting the ranking in 6 part there will be more medium level players emerging in one of the rankings just because they always win with the same victory condition.

The result are the following six Global Ranking by Victory Condition:

P.S. There could be some errors but I like these rankings because there are 6 first places, 6 second places, ....

 

[Xevious]

I like these rankings too. But then Diplomatic is my favorite style of play.

 

[rabies]

I like this a lot as well. In fact, it would be nice if something like this was added to the official global rankings..

I have recognized that some people always do well at certain victory types (Kemal with spacehip and Ronald with cultural stick out in my mind). I think a chart like this would give more incentive to pursue other victory types..and not just the Domination/Conquest that seem so prevalent at the top of the results lately.

edit: ...cause I can't spell 'pursue'.

 

[Yndy]

Guess it's easy too see that I’m also a conquest and domination fan. Actually I do win with other conditions but I don’t rank good. I am playing Space Ship this GOTM but I’ll rank lower than if I played domination.

 

[da_greatest]

I like them too. I made the list and my global ranking is 362.

 

[col]

Yes but

I think Dave Mc showed the rest of us the definitive way to a fast space launch in an earlier GOTM although he doesnt appear on the list. I guess he regards that as 'done'. I see he's playing OCC this month.

The top players can win pretty much any way they want!

 

[Yurian]

Questions:
1. How score falling depends of score value ? (See ranking Creepster,Yurian,Shillien …)
2. If I loss in game how many point I lose in my rating and why? (See ranking DaviddesJ,Smirk)
3. [pnp]dredd get +19,876 points in rating, but he had only QSC points – may be ERROR ?

Explain me please: How work Global Ranking 2003?

Sorry my badly english.

 

[DaveMcW]

1. Every month, everyone loses about 15% of their score. The exact number depends on the difficulty level and how long ago the games were played.

2. You always gain points for submitting. But if you did very bad, it might not be enough to bring your score back up from how far it fell in #1.

 

[Yurian]

But work of Global Ranking System is higher than my comprehension.
For an example I shall result the following table:

I do not understand why Moonsinger having 10712 Jason Score have received only 5.151 points in global rating!


For each event Score formula:
(according to New Global Ranking System for 2003)


EventX = (PlayerScore / TopScore) * EventValue * Life

Life = (Lifespan - Age) / Lifespan
PlayerScore = Player's score in the event
TopScore = Top score in the event
EventValue = 100 for GOTM, 15 for QSC, 5 for QSC writeup

6789 Lifespan = 4 + Difficulty.


If its true, then life for GOTM 20 may be calculate use Shadow2k result:

life = (TopScore*Event20)/(EventValue*PlayerScore)

life20 = 11760*52.776/6206*100 = 1

therefore Age20 = 0 (lifespan deity = 10 )

As from this I can receive negative result for DaviddesJ ?

Event20DaviddesJ = 801/11760*100 = + 6.811


Best regards, Yurian

 

[Aeson]

Moonsinger has played 8 GOTM and QSC. 7 of those months games had their lifespan decrease, and 1 was added at full strength.

For instance, GOTM13 results were counting 2/8 for month 19, and then 1/8 for month 20. This month they will stop counting towards the global rankings altogether. Moonsinger's GOTM20 added 91.027 points to her ranking, but the aging of the other games decreased their values by almost as much.

If a player submits consistant results, their global ranking will be pretty steady. Each new game added will be offset by the old games getting older, and thus counting less towards the rankings.

 

[Bamspeedy]

Also, wouldn't the fact that the older games were not using the Jason score affect things? In older games they used the Firaxis score, so some people could have 20,000+ points, but now the Jason score keeps the top players around 9-11,000 pts.

It should even out much better if a player plays every game, and all of those games were using the Jason score, shouldn't it? (the firaxis scoring also skewed the disparity between the top finishers (milkers), and the ones who didn't milk).

 

[Yurian]

If I have correctly understood then

Global Ranking20 = Rank20 +8/9*Rank19+6/8*Rank18+...+1/8*Rank13,

there RankX = rank for GOTMX (including QSC).

But next value Age for GOTM13 = 8 and Rank13=0.

How you calculate Global Rating for GOTM21 ?


Best regards, Yurian

 

[Aeson]

Bamspeedy,

The Global Rankings use top score normalization. Basically the gold medal winner gets 100% of the GR points for that event (Currently 100 for GOTM, 20 for QSC). Someone who scored half as much as the top score would get 50% (50 for GOTM, 10 for QSC). The change in scoring systems and inclusion of the QSC were major factors in deciding to use top score normalization. That way events of different natures or scoring systems can be fit into a coherant system of rankings.

------------

Yurian,

The month a result is added to the GOTM it gets 100% of it's value counted. When GOTM21 results are added, they will start off counting 8/8. The next month they count 7/8, then 6/8, ect... till they count 0/8 on the 9th month. Lifespan is determined by difficulty level + 4.

 

[LeSphinx]

Mark Cutt, I like your rating because I'm in the 10 first in Cultural 100K!
Alverwise, I'm around 45 !
But It's a good way to know which QSC timeline to check according to the way of plaing you want to improve on.

LeSphinx

 

[Bamspeedy]

Originally posted by Aeson
Bamspeedy,

The Global Rankings use top score normalization. Basically the gold medal winner gets 100% of the GR points for that event (Currently 100 for GOTM, 20 for QSC). Someone who scored half as much as the top score would get 50% (50 for GOTM, 10 for QSC). The change in scoring systems and inclusion of the QSC were major factors in deciding to use top score normalization. That way events of different natures or scoring systems can be fit into a coherant system of rankings.

OK, so if someone scored 11,000 Jason pts and got first, they get 100%, and someone who scored 5500 Jason pts would get 50%? With the Jason score, the ranking would be closer than the old games where the top player (milker) got say 20,000 Firaxis pts (100%), and another player (non-milker) got 5,000 Firaxis pts (25%)? So the Jason score helps balance them out, doesn't it? So as all the Firaxis-based scoring system games are fading away/out of the picture, it should be getting progressively closer to a fair comparison of skills/results, and that is why some players seem to be gaining many more points (they hadn't been milkers) while others aren't gaining quite as much (they were milkers), because the old scoring method games are fading away.

This is a good thing. I'm just saying, it just helps explain why some are moving up faster in the rankings than others (because they are benefitting more from the new system, and getting more properly placed where they should have been if the Firaxis score wasn't screwing them over).

 

[Aeson]

Yes, the changes in the scoring system are gradually making their mark on the global rankings. Also, the longer lifespans are bringing up the rankings for players who consistantly submit (and play the QSC), rather than just make sure to get 3 out of the last 5.

Right now we have games 13-16 counting towards the GR using the Firaxis score, and games 17-20 counting using the Jason score. So we are about at the halfway point now. It will be 5 more months until GOTM16 results drop out, and that will be the last of the Firaxis score results.

 

[Xarin]

Why not base the rankings off of precentage change from say the average of the last 5 GOTM top scores of the same type?

Example:

Previous 5 winning Conquest scores averaged 10K pts.
Previous 5 winning Diplomacy scores averaged 5K pts.

Thus, a score of 6K via Diplomacy and a score of 12K via Conquest would both be worth 120%, or equal in the rankings.

Over time, this should easy balance the disparity between the various victory conditions.

 

[Aeson]

The problem is that we don't see a high number of some types of victory conditions. For example, if one person submits a 20k victory, and is the only 20k submission, does that mean their game was as well played as possible (or within a few percent), regardless of how they actually did?

We use the top game because with 150-200 submissions, the highest scoring game should be consistantly close (within a few percent) to the best possible performance on the map. The Jason score is already balancing the victory conditions.

 

[Xarin]

*checks last two months of results*

Top 10s - 12 Domination, 6 Conquest and 2 Diplomatic.

Doesn't seem like victory conditions are balanced AT ALL. I'd say that's the REASON you don't see a high number of some types of victory conditions. Seems to me there's a HUGE problem here.

<<< For example, if one person submits a 20k victory, and is the only 20k submission, does that mean their game was as well played as possible (or within a few percent), regardless of how they actually did? >>>

No. Don't see the relevance. Under my system, if one of the victory conditions is underplayed, people are going to notice and play for that victory condition so they can win. It naturally FORCES people to be diverse in their game play, rather than just play for domination/conquest all the time.

 

[Aeson]

Players are free to choose to play the victory conditions they want to. Many of the Conquest/Domination games that have placed so high the last two months actually could have scored higher by playing out to on of the 'later' victory conditions like Spaceship, or slightly changing their priorities to allow for a 20k or 100k victory.

Most of the top players have just been choosing Conquest/Domination. The time required is probably the main reason.