Update!!

[zoraster]

(01-30-2013 03:51 AM)EkoFox Wrote:  

(01-30-2013 03:36 AM)Ja Karta Wrote:  

(01-30-2013 03:29 AM)Harti Wrote:  So you're saying placing a 3 HP "Soldier" (in terms of defense) hypothetically anywhere on the field at the cost of 1 (without having to move to the destination) is expensive?

A half-cost soldier that can't attack and moves like a heavy (with you paying for each move)?
I'd call that expensive, yes.

Don't you pay for each move of any unit? Why did you feel the need to make that point?
It sounds like you're not managing your wits and are upset about losing your bramble patches a lot because you aren't protecting your parents.

---

(01-30-2013 03:33 AM)zoraster Wrote:  Maybe I'm misreading this, but the new league system seems a bit off. Let's say that I'm at the top of my division and I'm at about 50% win rate. The matchmaking system is working, right? But here's the problem: maintaining that seems mostly about churning out games. In theory, those who play 35 games at a time will have higher ranks.

Compare two players that have about equal success: an aggressive player and a defensive player. The aggressive player's games might last a significantly shorter period of time, so he can churn through 2 games in the space of what a defensive player goes through 1 (on average, obviously).

If players are earning more simply for playing games, then this situation does create incentive just to play as many as possible -- even to a degree at the expense of playing well.

If a player can beat his opponents twice as quickly as a different player, then maybe Player A is just better? It seems justified that a player who can win often by playing aggressively should have more points than one who wins by playing defensively. In my opinion, playing defensively is much easier than being aggressive.

I don't think taking turn count into scores, whether directly or indirectly, makes sense. There is value in being careful and thoughtful. The effort should be to make people want to win their games.

But more to the point, such a structure actually defeats the purpose of the change: to get players at equal skill levels playing each other. I don't know the formula used right now, of course, but let's say by playing an average game my expected point gain is 10, rather than the 0 it was before. So I go along and I keep climbing the ranks. At some point I start losing more and more matches. But my expected value is still net positive. Until it turns negative, much later than my skill level would indicate, I keep climbing ranks DESPITE LOSING more often. Just because I can churn out games faster than the people around me.

Think about the perverse incentives this provides as well. Say you're really trying to game the system. You've got 35 games going. One ends and you start a new one. You get a map that plays quickly or average, you play it. But we all know there are certain maps that tend to last far, far more turns. Perhaps you see it's against a defensive team too. You COULD go ahead and grind it out with an expected value of 10 (using the guess above) that you think might take a two weeks to finish. OR you could resign (known value of, say, -5) and start a new game, hoping to get a quick map against a less defensive team.

I don't think MOST will try this sort of thing. But it is a structural weakness of the system.

Last, while I think we want to encourage people to be active, we don't really want to encourage people to play way more games than they're comfortable doing, possibly leading to a lot of burnout. If someone with 1 game is disadvantaged versus someone playing 10, I don't mind. But if someone playing 10 games at a time is disadvantaged against someone playing 35, I think we have a more serious problem.