FTA

[A. Normak]

I`ve noticend that some maps are suffering from the Fist Turn Advantage, especially the small ones. I heard, this was discussed during beta testing. Are you planning to fix this? This may be very important for high level games. Maybe the first player should start with less wits than player 2?

[Harti]

The fact that wits are limited somehow relativises the extent of FTA.

I've played Advance Wars competitively and created lots of maps myself so I do know what you're talking about. In AW, you can move every unit every turn, there is no limit on actions. This is what made FTA a huge problem.
In Outwitters, however, with the 5 starting wits you usually tend to get one bonus point or two and move some predeployed units around. You can't do soooo much to exploit FTA here. Yeah, being P1, you're the aggressor, which is definitely something to worry about. You act and P2 has to react to your actions, in some cases crippling their strategy. But luckily, P1's actions are limited and during beta testing we found that it's easier to defend than to break through a defense.

Except for two (and a half) maps there are no outstanding advantages to be P1, and usually those strategies will only work if the opponent doesn't have at least some kind of foresight and is busy with himself/herself.
If P1 doesn't understand when P2's fended off the FTA exploit completely, they might overextend and get crushed.


Unfortunately, there are not so many high-level games between equally skilled players right now, and there haven't been any during beta testing phase either. The metagame is still developing and One Man Left is always having a close eye on imbalances regarding teams or maps. We had some nice ideas to limit FTA exploitability, so once there's 100% proof for heavy imbalances OML's gonna throw them in.

[stevewastaken]

(07-07-2012 04:26 AM)Harti Wrote:  In Outwitters, however, with the 5 starting wits you usually tend to get one bonus point or two and move some predeployed units around. You can't do soooo much to exploit FTA here.

This was the mindset of HA originally as well. Eventually it became clear that in competitive games, FTA became significantly important to winning. I forget the exact numbers, but I think it was something like 20% more likely to win with FTA, and that's even considering HA has elements of luck and team balance that Outwitters does not.

In an ELO system (which Outwitters uses for all league games), where basically every game will be competitive, it will probably become clear pretty quickly that FTA has major impact.

In the HA player run league, the solution was to start the starting player with 3 Action Points on his first turn, rather than the standard 5. While there is still a statistical advantage to FTA using the 3 AP system, it is pretty slight, when previously it was significant. That same solution would probably have the same results for Outwitters as well, considering both games start with 5 action points.

[Harti]

(07-07-2012 08:38 PM)stevewastaken Wrote:  In the HA player run league, the solution was to start the starting player with 3 Action Points on his first turn, rather than the standard 5. While there is still a statistical advantage to FTA using the 3 AP system, it is pretty slight, when previously it was significant. That same solution would probably have the same results for Outwitters as well, considering both games start with 5 action points.

Interesting because this would be 'our' first shot towards balancing as well. This approach is even used in Trading Card Games like Magic: the Gathering (player 2 has 1 extra card in their opening hand).

I'm positive this will work for Outwitters - but once you've got to occupy 2 bonus spaces with predeployeds and would have to set up a couple of things to prevent STA, you can't (due to the limit of 3). This will require quite a lot of maptesting.

[Kamikaze28]

(07-07-2012 08:49 PM)Harti Wrote:  

(07-07-2012 08:38 PM)stevewastaken Wrote:  In the HA player run league, the solution was to start the starting player with 3 Action Points on his first turn, rather than the standard 5. While there is still a statistical advantage to FTA using the 3 AP system, it is pretty slight, when previously it was significant. That same solution would probably have the same results for Outwitters as well, considering both games start with 5 action points.

Interesting because this would be 'our' first shot towards balancing as well. This approach is even used in Trading Card Games like Magic: the Gathering (player 2 has 1 extra card in their opening hand).

I'm positive this will work for Outwitters - but once you've got to occupy 2 bonus spaces with predeployeds and would have to set up a couple of things to prevent STA, you can't (due to the limit of 3). This will require quite a lot of maptesting.

Alex and Adam said in one of the two podcasts which accompanied the release that they have metrics running to keep an eye on such things.

There is one thing to keep in mind though: is it general FTA, independent of the map or is it a map specific problem? To answer this, we have to be patient and let the game run its course. In this early stage of league play, rushing an opponent as P1 is easier than defending as P2 - therefore we may see a false imbalance towards FTA. Once these rush tactics are known and understood and possibly countered effectively, their use might diminish and we will get more reliable results.

Thinking about general ways to reduce FTA is not wrong, but given that there are several maps and hopefully more to come, I'd argue that it would be best to identify those maps with heavy FTA and change their layout.

[stevewastaken]

(07-07-2012 08:57 PM)Kamikaze28 Wrote:  There is one thing to keep in mind though: is it general FTA, independent of the map or is it a map specific problem?

In my experience, it depends largely on the number of turns it takes to finish. The "faster" a map plays, the more problematic FTA is.

For example, if a game can last only 3 turns, (taking bonus spaces out of the equation) the starting player got to use 15 Wits, while the second player only got to use 10. That would be a HUGE advantage.

Tic-tac-toe takes about that many turns, and if you play properly, you will always win with FTA.

If a game last 100 turns, the starting player got to use 500 wits, while the second player got to use 495. That's an advantage, but a fairly small one.

Chess takes quite a few turns, and FTA isn't a big deal, unless you are talking about unbeatable Grand Masters... which brings us to the next factor: skill.

Even in tic-tac-toe, 2 people who don't know how to play can go back and forth with who wins and loses. While the second player will recognize he is at a disadvantage, he'll still win a few thanks to "mistakes" by the other player.

Most people view a mistake as a serious miscalculation that results in a setback. However, when talking about extremely high level play, a mistake can be a really great move, but not the absolute maximum threat to the opponent. A top level opponent will see this teeny-tiny opportunity to take an advantage and as long as he pushes turns to the absolute maximum threat every turn, he will carry that opportunity to a win.

So in chess, when a Grand Master faces Grand Master, FTA is pretty important. Because if neither makes a mistake, then the win goes to the guy who went first. He got (I don't know) 200 turns, while his opponent got 199.

But games with action points are interesting. Because if you cut in half the number of starting action points of the FTA, it (in theory) mitigates the mathematical issue of who got more turns. What's left behind are questions like "Is it better to be able to decide how things start, or is it better to be able to see what your opponent is doing and know how to counter it?"

In my opinion, that's where you want it. When the question is "what strategy is best?", and not "who got more chances to move/attack?".

[Szei]

There's another important point regarding how in chess FTA isn't as big a deal. In chess you can only move 1 piece on any given turn. One of the major things about Outwitters which makes FTA so strong is the fact that you can move many pieces at a time. This allows the starting player to gain a strong positional advantage from an early stage. You can establish forward/central map presence a turn earlier which can dictate the rest of the game as you'll have better scouting information allowing you to make informed decisions rather than shooting in the dark.

(Although on second thought the same should essentially be true of chess. It just takes more time for the starting player to establish that key position a turn earlier than his/her opponent can since they can only move 1 piece at a time. The real difference is that in Chess there is NO FOG OF WAR so having that central map control earlier, while undoubtedly important, isn't as crucial. In Outwitters, losing out on center map control and the subsequent scouting information you obtain is a major setback)

Eventually a first turn disadvantage can be implemented to offset this FTA (like the 3 wits on first turn or only being able to move 1 or 2 units so you can't immediately move out and claim map control) *or* we can shift away from the smaller maps and into mostly larger maps where FTA isn't as big of an issue. The nice thing about keeping the maps from getting too large is that the games also won't become too lengthy. It's nice how the current Outwitters games don't drag on too long. I'm not advocating HUGE maps, but simply that we gravitate towards slightly larger maps as opposed to the smaller ones like Sharkfood Island where FTA is quite significant.

[ihack4fun]

I'd also like to add that the tic-tac-toe reference is invalid. There is no FTA unless the players are young or unintelligent. It has been mathematically proven that it should always be a draw. It's also very simple to show -- make any move and I can show you the correct counter so u don't lose.

[Cookie]

(07-07-2012 08:49 PM)Harti Wrote:  

(07-07-2012 08:38 PM)stevewastaken Wrote:  In the HA player run league, the solution was to start the starting player with 3 Action Points on his first turn, rather than the standard 5. While there is still a statistical advantage to FTA using the 3 AP system, it is pretty slight, when previously it was significant. That same solution would probably have the same results for Outwitters as well, considering both games start with 5 action points.

Interesting because this would be 'our' first shot towards balancing as well. This approach is even used in Trading Card Games like Magic: the Gathering (player 2 has 1 extra card in their opening hand).

I'm positive this will work for Outwitters - but once you've got to occupy 2 bonus spaces with predeployeds and would have to set up a couple of things to prevent STA, you can't (due to the limit of 3). This will require quite a lot of maptesting.

I have a hard time seeing the problem here. Given that the first player gets any number of wits at all on his first turn, the new STA will be smaller than the current FTA, that is a hard fact. You have to remember that the first turns will play out like 2 5 7 wits so the first player has plenty of wits to react on his second turn. With your argument, the second player will have to spend 2 of 5 wits on his first turn to take bonus squares, right?