The ethics of milking (post Jason)

[Ribannah]

Moonsinger,

You are missing the point. The rabbit played poorly and never crossed the finish line.

Let me try to clarify my thoughts.

Player A and player B choose different paths to victory, both intending to continually make progress towards that goal to the best of their ability. The only difference is that player A emphasizes speed, while player B emphasizes the state of the empire.

Case 1. A and B both play great. A finishes earlier, and B has the most impressive empire. IMHO, they should get equal scores.

Case 2. A and B both play a mediocre game til the moment A finishes. After that, B improves and plays an excellent final part. Here, player B should score higher.

Case 3. B fares a little better than A. A still finishes earlier, but after that B suddenly does badly in the remaining part. There should be a chance that B scores lower.

Case 4. A fares a little better than B, and finishes quite a bit earlier. For the rest of the game, B continues to play at the same level. Player A should have the higher score.

My estimate at this point is that the Jason score still favours player B's approach to a point that player A tends to lose in case 1 and 3, and even in case 4. The favouritism is way less than before, to a point that exceptions are possible, but still very present.

But that's after only one month. I could be wrong. We should get a lot of info from this month's game.

The question I'm asking is how we want it to be.

Do we want to give only special awards for the speed games but reserve the medals for games that balance speed and empire - or even focus on empire only, as before - or do we want all approaches to have an equal shot at the medals, if played well?

From your last post it seems that, like me, you'd want all approaches compete without favouritism, but your earlier posts sounded more like you'd prefer to give extra points for the player who emphasizes the empire, because the (in your opinion, I don't agree) extra effort should be awarded.

 

[Bamspeedy]

To go along kind of with what E-man was stating. Before the Jason Score was introduced, whether or not you milk a GOTM was a no-brainer (if you wanted a shot at a medal) on most maps. Some of the non-milkers believed (wrongly in many cases) that the milkers were only milking to inflate their rankings and score. There was doubt in some people's minds about if these top players got their high scores by just the fact that they had the time to milk, or were they that much better of a player?
You could compare the milkers games to each other, but you couldn't compare a 2050 AD victory to an 1100 AD victory.

Under Firaxis's scoring system, it sometimes makes it look like the top players are 10 times better than other players. They are great players, but not *that* much better. With the Jason score, scores are more closer to reality, with many scores within just a few percentage points from each other, instead of being a 1000-2000% scoring difference, like their was under the Firaxis scoring system.

Now that we find that milking won't inflate our scores that much (or not at all), the real question is:

Is it still worth the time to milk the game?

The players who do enjoy milking a game, will continue to do so from time to time. Players who don't like to milk, will certainly have much less (or no) pressure on them to milk (if they care at all about rankings).

GOTM17, nobody milked. This was due mainly to 2 things, IMO.
1. everybody wanted to test the scoring system and see how fast victories scored. Unfortunately, there wasn't any milked games to compare against
2. Finally a game, where we didn't 'have' to milk in order to score well. So, I'm sure some players that had milked every game for the last X months, were relieved to have a change of pace and do something different.

 

[Yndy]

I mostly agree with everything that you have said so far. But I want to add another perspective here. A classic milked game is a fast domination game up to the domination threshold. It is not the fastest domination game either because you still have to mind the happiness of the population while fighting. So in my opinion, under a fully fair scoring system (which may never exist by the way)
Earliest domination (ED) = Early milked domination (EMD) = Fully milked game (FMG)
There are three issues here.
1. Equaling the finish bonus difference between the ED and EMD with the score difference.
2. Equaling the finish bonus for EMD with the score difference between FMG and EMD.
3. Equaling these differences at every intermediary step between them.
Based on Ribannah’s cases I would think that the Jason score is pretty good for cases 1 and 2 and leans towards the milking victory for cases 3 and 4 but only marginally.
I intend to study this problem a little deeper in GOTM 20 when I will have time for a fully milked victory. I would like to study my score at the earliest victory turn and every turn afterwards should win. Ideally, the score should remain unchanged. In practice it would probably vary either above or below the earliest victory score. To make things clear I won’t reload every turn but calculate the score that I would have if I won that turn.

 

[Moonsinger]

Originally posted by Ribannah
You are missing the point. The rabbit played poorly and never crossed the finish line.

I concur! I did previously mention that I lost the point some where. Therefore, it was just a nice story of the race between the rabbit and the turtle.

 

[alamo]

Originally posted by alamo
Can we get a description of the Jason scoring system on the GOTM page?

The stuff is there, just not linked to the main page.

Calculator
Jason Calculator

Description
Jason Description

According to the calculator my GOTM18 SS victory in 1958 gets reduced from 2878 to 2133. I guess the balancing cuts both ways.

There are lots of interesting points that I don't fully appreciate. I'll probably never have the time and patience to be a milker.

What is more important - winning or how you play the game?

 

[Bamspeedy]

The Jason score does go up and down from what I've seen.

GOTM18 (no spoilers mentioned other than score) I kept track of my score every turn. Unfortunately, my computer crashed and I lost all the information on my spreadsheet from 1932 A.D.- 2034 A.D. I had a similar crash around 1200 A.D. or so and had to guess what my score had been during a 10 turn or so period.

I haven't calculated the jason score for every turn, but taking some samples every 10-20 turns or so, I see some patterns.

My Jason score peaked at one point around 1600 A.D. This was a while after I was maxed out in population and territory. I had been gaining a few points/turn from the point I 'could have' won the game via conquest up until 1600 A.D., then after that I started to lose some points (anywhere from less than a point/turn to 5 points/turn). My Jason score continually dropped a few points every turn (some fluctuations in score, but it was generally decreasing). But then at some point in the post 2000 A.D. era I was gaining points back, and eventually passing my 1600 A.D. Jason score.

Overall, I only increased my Jason score by less than 10% by milking for 1,000 years (not worth the effort). The firaxis score, on the other hand, I had more than doubled. More evidence of how great this new scoring system is.

 

[mad-bax]

Bamspeedy:

I am glad you made this post as it answers the obvious question raised by the post you made about your completed game.

If what you say is true then the effort that goes into milking a game is still rewarded a little if you can be bothered. I personally am fairly neutral about this. The only thing I'm not sure of is whether or not you could have had an even earlier "theoretical" finish date if you had gone all out for the earliest domination/conquest win, in which case you would have had a higher Jason score. Would you have played any differently if you had no intention of milking the game? If so then milking really isn't worth it under the new scoring system is it?

The obvious thing is that the Jason system is a really good "leveler". My only concern is that it might completely remove the "milking skill" from the GOTM and I don't want that to happen.

 

[Bamspeedy]

Whether or not you are going for a milking game does change when you could achieve other goals. If I wanted just a quick conquest, I would produce much more military instead of settlers and workers. I would have only needed enough workers to improve my core cities and to road ahead to my next victims. No need for 300+ workers or 400 cities if fast conquest was my goal. Fast tech progression helps milkers (because of railroads) more than fast conquerers (unless astronomy/navigation is needed or you need a cavalry rush), especially if using a powerful early UU. Quick domination, then I would have built a little more military and space my cities a little further apart (and keep the tundra cities).

So, if I go strictly for speed, I would have a higher score in the 'best date' part of the Jason score, but my population/territory would be lower. IMHO, the best Jason scores would be from either beating the 'best date' (or come really close), or to be continually adding population/territory as you progress towards your victory, even if it slows the victory by a few turns. And that is what I did in that game with having settlers constantly chasing after my military units on the front line.

There is still the 'cow' award for highest score, and some players do enjoy milking, so you don't need to worry about the 'milking skill' dissapearing.

 

[Capt Buttkick]

I have never milked civ3 games cause I got bored of milking in civ2 (I once ended up with a negative percentage after milking a game thoroughly). I now haven't got the time nor the stamina needed to milk games.
That said: I'm, like mad-bax, pretty indifferent to whether it's marginally better to milk a game than not too. In fact, milking should preferrably be just a few points better than not milking so that people, like Bamspeedy did here, starts questioning whether milking a game is worth it. That's why Jason imo is so good.

Like someone else said: when you have a good setup and the initial worker-joining has been done, there's not much skill in milking a game. It DOES take an awful amount of time, though. If people want to put in days of toil to secure the extra 50 or 100 points that will give them a medal, I say let them. I won't loose any sleep seeing milked games ahead of me in the rankings (and that's NOT because I'm far from the top. I plan to get there, if it can be done without milking lol).

 

[Ribannah]

Originally posted by Bamspeedy
Overall, I only increased my Jason score by less than 10% by milking for 1,000 years (not worth the effort). The firaxis score, on the other hand, I had more than doubled. More evidence of how great this new scoring system is.

I'd say it may still be worth the effort. The difference in points is much smaller, but that goes for everyone.

The Jason formula levels scores, but that by itself means little.

The question is, does it also change the order, or will the fully milked game still end up ahead (albeit by a much smaller margin)?

If that is still he case, then the new scoring system merely clouds the issue by making it seem that non-milked games can be competitive, while in reality they aren't.

Of course it is also possible that your milking skills are superb and that is why you were able to eventually crank up your Jason score.

Hopefully the results of GOTM#18 will give us an indication, as quite a few of the top players are keeping track of how the scoring formula works out.

So well done, Bamspeedy!

 

[Bamspeedy]

Well, my 2050 A.D. Jason score was less than 10% higher than my 1020 A.D. Jason score (I won't give the exact %, since I already posted my 1020 AD Jason score elsewhere), but my final Jason score was only 1% higher than my 1600 AD Jason score.

But that is only for my game and playstyle. For other playing styles, people may not see a decrease in points. At the very end of the game I was increasing 21 points/turn (Firaxis points). Someone who didn't move into the 'milking phase' until much later in the game than I did, would probably still be earning more than 21 points/turn, so they would see a different curve.

Everyone is going to have different degrees of increase in scores (conquering alot of land very quickly, railroads-if lots of irrigation is in place, hospitals, accessing luxuries, etc.), that the curve will affect everyone differently, and no formula could possibly balance that out perfectly for everyone.

 

[Shillen]

I think milking is more beneficial to the less skilled players or if you have bad luck and don't get close to the best finish date. I didn't reach conquest until around 1365AD which is well after the best date. By milking I was able to increase my Jason score by 25%. But I also did waste time because I planned to milk it all along. If I had just gone for the fastest conquest I'm sure I would have finished a hundred or more years earlier. I also noticed the fluctuation in score which was very weird to say the least. My firaxis score would go up very steadily, but my Jason score would go up by 2 one turn, then 6 the next, then 5, then 3, etc. And very surprisingly I noticed the same thing as Bamspeedy. After the year 2000AD my score started increasing rapidly, even though I was long done setting up my milk. So I don't see why it should have gone up so fast since all conditions remained the same. In the beginning of my milking period my Jason score was going up by about 5-6 points each turn, then it dropped down to 2-3 for a while in the 1800-1900s. Then in the 2000s it was going up by 10-12 points per turn.

 

[Moonsinger]

Well, I really want to say something about what I learned during the GOTM18, but I can't really say it until the end of the month.

 

[Capt Buttkick]

You can say it in the spoiler thread, Moonsinger and we can go there

 

[Moonsinger]

Originally posted by Capt Buttkick
You can say it in the spoiler thread, Moonsinger and we can go there

Well, since we are trying to keep the spoiler thread clean, going there isn't a good idea.

 

[Capt Buttkick]

I'm not even going to comment on that