EDIT/Summary: I think I used too many unclear terms, but this covers most of the issues in captain mode at the moment and addresses my interpretation of why the mode feels "too simple" and "too similar" round to round.
I have played Captain since it was released in alpha. There's been wildly different game-states -- cav dominance, archer dominance, shock dominance, light infantry dominance -- and I've played in them all. This post is a summary of the current state of Captain, which will hopefully be useful as we move into separated perks between Captain and other modes and for overall theory about the mode.
Before getting into the details, I'll summarize why Captain has potential:
Crashes
This is being worked on, but listed due to significance.
(Reminder that if you don't get a crash report to send automatically, you have to navigate to
Class and Perk Balance
No class or perk should be OP to any sort of player group, from a match of random first-time players to competitive 6v6 where half the lobby has every class and best perks memorized. Conversely, no class or perk should be underpowered in all player groups, ensuring that every class and perk is viable and useful for at least one audience. (This is assuming a certain amount of viable options per audience.)
Determining what classes and perks are OP or UP is very easy at the moment in captain, since there's nothing dynamic going on in terms of pick-rate or win-rate. You can look at each faction and quickly determine which classes are OP and UP, and if you're familiar with the meta, you can quickly rattle off which perks are good and which are terrible. Obviously, some massive changes are coming when the captain/skirmish split occurs, so I'll just list the major offenders here.
Overpowered:
Captain (Rambo) vs Troop Balance
At the moment, rambo is a hot issue. Some players claim it's a good demonstration of mechanics, while others claim it ruins the game.
The latter group is categorically right. Rambo ruins the game. You can't have a "Captain" mode be based on parking your bots on the other side of the map and slowly picking off enemy troops, pushing the game to the time limit every round. That's not what the mode's supposed to be about and it's not fun for anyone except the cav player.
And you might as well just remove archers. If you gave archers god-tier aimbot the enemy cav commander would still just ride up, shield up, and distract the archer unit with absolutely zero counterplay while their cav unit assists combat elsewhere. Skirmishers have been terrible for a long time, but I expect the same would happen for them.
However, archers don't have godlike aim, and spear and pike infantry will slowly get picked off by rambo as well. It's a race against time at the moment where only the cav players get to play, and everyone else just watches with some minor movement here or there.
This can be fixed with three changes:
1) When the commander is a certain distance away, their bots automatically follow. This puts their bots closer to danger and allows more counter play.
2) Greater troop counts. Troop count is determined by a multiplier, which is currently 20. A new base of 25 or 30 would be interesting to try. A solo captain now has to kill 20%+ more troops, making rambo less viable.
3) Improved Cavalry AI. Part of the reason why rambo is used is because your cav will probably just die without accomplishing anything. Improved cav AI would allow your cav to get more kills and damage.
Some ramboing can still occur with these changes, but in a less oppressive way.
Faction Balance
The main issue at the moment is how data is determined. For example, just balancing factions to 50% off random battles is useless when Empire has a 1% pick-rate and a 25% win-rate in Captain's League. This is method hurts players as they get better at the mode or start playing with friends.
Conversely, let's say we're in a situation where something isn't that great in Captain's League, but is very oppressive in random matches. Let's say that Berzerker is very strong in random battles, but Captain's League knows how to play against Berzerker and pepper them with archers or javs, so Sturgia doesn't even take them that often.
The solution is as I mentioned above: if something is OP for either audience (casuals/random and competitive/captain's league) it should be nerfed. And if something is UP for all audiences it should be buffed. However, due to only 6 factions being in the game, there should be an additional check to ensure a healthy game-state, such as all factions must be within 49% and 54% win-rate for both audiences. If they aren't, then a rework must occur.
In this case, Empire should be buffed so that they aren't UP for Captain's League, while Berzerkers are nerfed or redesigned so that they aren't a problem in random battles.
Lack of Strategic Depth
I had this at the start, but I figured it would scare people away, so I stuck it down here.
Let's consider what factors lead to decision making, and how players can express strategic skill. Due to the importance of strategy over mechanics in Captain, this is crucial to replayability.
At the start of each round in a matchup (or the Pick Stage), each team determines what classes and perks they will select for this round.
For the 3-5 rounds of a map-half, this is always the same. A team always has the same faction, always has the same opposing faction, and will always be able to choose any combination of classes or perks within that faction. This is a team's "position."
That position does not change for the 3-5 rounds of the map-half, and the inverse is extremely similar after the faction swap, meaning 6-10 rounds of a very similar position. The skill expression in picking the correct troops is a blind guess, based on what you expect your opponents to run based on the factions, map, and what they ran the previous round. So, this expression is blind and not-dynamic for many rounds in a row, only changed to a significant degree when a new map and factions are used.
The response to this is to make strategic decisions in the Pick Stage guided and dynamic. I get to that in the section below.
In the Combat Stage of a round, skill expression contains positioning, timing, mechanics, and adapting to the specific class matchup when it is revealed. I would consider this about equal to the Pick Stage, because games can easily be won or lost from actions in either. The improvements to skill expression in this category are a bit easier:
More maps for more positioning and timing complexity. More motivation for teams to use different parts of the map. And the removal of enemy classes being revealed at the start of the round. There should be a 20s-1m delay where scouting is useful and hiding is viable. This would probably work best at 25s-30s if I had to guess.
Now, back to the Pick Stage:
Adding Depth Through Position Variance
The goal is to have each round offer a complex situation to a team, that the team breaks down and reacts to in a unique way. Again, this adds replayability to a mode that relies on strategy over mechanics.
During the Pick Stage, changing the "position" of each team should increase the amount of decision making. Theorycrafting some methods on how to make more dynamic position results in: weaker, stronger, more resources, less resources, and more weighted towards one playstyle or another.
Pick Stage, Round 1
This means that their positions are different from the start. Team 1 now has a hypothetically stronger army, but no leftover gold. Team 2 has a slightly weaker army, but some leftover gold.
The next round, each team gains 50g passively, and now each player on Team 1 has 50g; and most players have 50g on Team 2, except two players who have 80g.
On playstyle, let's say that you can sell a troop back into the shop for 90% of the original price. So you could sell a 100g unit for 90g back, and with your passive income of 50g, now have 140g to spend. However, you could stay on the troop and buy upgrades for it. Such as with a Tribal Warrior, you could choose to stay on the troop and buy a higher troop count, better armor, a better weapon, a spear, ect. This means that switching classes would result in lost potential value, but allow you to counter-pick the enemy better. It also means that your opponent is likely to bring a similar comp, giving you more information and guided decision making.
Maps
Maps are really simple: we just need more. Opening up mapping to the community would immediately solve this problem, but quality control would be necessary.
I have played Captain since it was released in alpha. There's been wildly different game-states -- cav dominance, archer dominance, shock dominance, light infantry dominance -- and I've played in them all. This post is a summary of the current state of Captain, which will hopefully be useful as we move into separated perks between Captain and other modes and for overall theory about the mode.
Before getting into the details, I'll summarize why Captain has potential:
- Unique compared to any "mainstream" game. There is an appeal in controlling an army, while being on the ground yourself, and working with allies against a team of equally intelligent opponents.
- Best engine of any "in-genre" game. Tiger Knight, Conqueror's Blade, and Blood & Steel all have inferior engines.
- Despite the current state, Captain players keep coming back. People see the potential.
Crashes
This is being worked on, but listed due to significance.
(Reminder that if you don't get a crash report to send automatically, you have to navigate to
C:\ProgramData\Mount and Blade II Bannerlord\logs\rgl_log_xxxxx.txt
and send it to @MArdA TaleWorlds. Also, you can't reboot the game before saving this file or the log will be overwritten.)Class and Perk Balance
No class or perk should be OP to any sort of player group, from a match of random first-time players to competitive 6v6 where half the lobby has every class and best perks memorized. Conversely, no class or perk should be underpowered in all player groups, ensuring that every class and perk is viable and useful for at least one audience. (This is assuming a certain amount of viable options per audience.)
Determining what classes and perks are OP or UP is very easy at the moment in captain, since there's nothing dynamic going on in terms of pick-rate or win-rate. You can look at each faction and quickly determine which classes are OP and UP, and if you're familiar with the meta, you can quickly rattle off which perks are good and which are terrible. Obviously, some massive changes are coming when the captain/skirmish split occurs, so I'll just list the major offenders here.
Overpowered:
- Khuzait Lancer
- Khuzait Spear Infantry
- Sturgia Warrior
- Aserai Skirmisher
- Aserai Beduin
- Empire Archer Militia
- Empire Recruit
- Khuzait Mounted Archer
- Khuzait Nomad
- Sturgia Brigand
- Sturgia Raider
- Vlandia Vanguard
Captain (Rambo) vs Troop Balance
At the moment, rambo is a hot issue. Some players claim it's a good demonstration of mechanics, while others claim it ruins the game.
The latter group is categorically right. Rambo ruins the game. You can't have a "Captain" mode be based on parking your bots on the other side of the map and slowly picking off enemy troops, pushing the game to the time limit every round. That's not what the mode's supposed to be about and it's not fun for anyone except the cav player.
And you might as well just remove archers. If you gave archers god-tier aimbot the enemy cav commander would still just ride up, shield up, and distract the archer unit with absolutely zero counterplay while their cav unit assists combat elsewhere. Skirmishers have been terrible for a long time, but I expect the same would happen for them.
However, archers don't have godlike aim, and spear and pike infantry will slowly get picked off by rambo as well. It's a race against time at the moment where only the cav players get to play, and everyone else just watches with some minor movement here or there.
This can be fixed with three changes:
1) When the commander is a certain distance away, their bots automatically follow. This puts their bots closer to danger and allows more counter play.
2) Greater troop counts. Troop count is determined by a multiplier, which is currently 20. A new base of 25 or 30 would be interesting to try. A solo captain now has to kill 20%+ more troops, making rambo less viable.
3) Improved Cavalry AI. Part of the reason why rambo is used is because your cav will probably just die without accomplishing anything. Improved cav AI would allow your cav to get more kills and damage.
Some ramboing can still occur with these changes, but in a less oppressive way.
Faction Balance
The main issue at the moment is how data is determined. For example, just balancing factions to 50% off random battles is useless when Empire has a 1% pick-rate and a 25% win-rate in Captain's League. This is method hurts players as they get better at the mode or start playing with friends.
Conversely, let's say we're in a situation where something isn't that great in Captain's League, but is very oppressive in random matches. Let's say that Berzerker is very strong in random battles, but Captain's League knows how to play against Berzerker and pepper them with archers or javs, so Sturgia doesn't even take them that often.
The solution is as I mentioned above: if something is OP for either audience (casuals/random and competitive/captain's league) it should be nerfed. And if something is UP for all audiences it should be buffed. However, due to only 6 factions being in the game, there should be an additional check to ensure a healthy game-state, such as all factions must be within 49% and 54% win-rate for both audiences. If they aren't, then a rework must occur.
In this case, Empire should be buffed so that they aren't UP for Captain's League, while Berzerkers are nerfed or redesigned so that they aren't a problem in random battles.
Lack of Strategic Depth
I had this at the start, but I figured it would scare people away, so I stuck it down here.
Let's consider what factors lead to decision making, and how players can express strategic skill. Due to the importance of strategy over mechanics in Captain, this is crucial to replayability.
At the start of each round in a matchup (or the Pick Stage), each team determines what classes and perks they will select for this round.
For the 3-5 rounds of a map-half, this is always the same. A team always has the same faction, always has the same opposing faction, and will always be able to choose any combination of classes or perks within that faction. This is a team's "position."
That position does not change for the 3-5 rounds of the map-half, and the inverse is extremely similar after the faction swap, meaning 6-10 rounds of a very similar position. The skill expression in picking the correct troops is a blind guess, based on what you expect your opponents to run based on the factions, map, and what they ran the previous round. So, this expression is blind and not-dynamic for many rounds in a row, only changed to a significant degree when a new map and factions are used.
The response to this is to make strategic decisions in the Pick Stage guided and dynamic. I get to that in the section below.
In the Combat Stage of a round, skill expression contains positioning, timing, mechanics, and adapting to the specific class matchup when it is revealed. I would consider this about equal to the Pick Stage, because games can easily be won or lost from actions in either. The improvements to skill expression in this category are a bit easier:
More maps for more positioning and timing complexity. More motivation for teams to use different parts of the map. And the removal of enemy classes being revealed at the start of the round. There should be a 20s-1m delay where scouting is useful and hiding is viable. This would probably work best at 25s-30s if I had to guess.
Now, back to the Pick Stage:
Adding Depth Through Position Variance
The goal is to have each round offer a complex situation to a team, that the team breaks down and reacts to in a unique way. Again, this adds replayability to a mode that relies on strategy over mechanics.
During the Pick Stage, changing the "position" of each team should increase the amount of decision making. Theorycrafting some methods on how to make more dynamic position results in: weaker, stronger, more resources, less resources, and more weighted towards one playstyle or another.
- Weaker/Stronger = The army you play next round is expected to be weaker/stronger than average.
- Resources = You can spend resources (such as gold) to become stronger this round, but have weaker upgrades in the future due to not saving up.
- Playstyle = It is easier for you to run an army strong at one playstyle, such as infantry or archers, compared to another.
Pick Stage, Round 1
- Team 1 Army Strength: 600g; econ 0g. (Everyone bought a 100g class.)
- Team 2 Army Strength: 540g; econ 60g. (Everyone bought 100g class except two, who bought 70g classes)
This means that their positions are different from the start. Team 1 now has a hypothetically stronger army, but no leftover gold. Team 2 has a slightly weaker army, but some leftover gold.
The next round, each team gains 50g passively, and now each player on Team 1 has 50g; and most players have 50g on Team 2, except two players who have 80g.
- Team 1 Army Strength: 600g; econ 500g.
- Team 2 Army Strength: 540g; econ 560g.
On playstyle, let's say that you can sell a troop back into the shop for 90% of the original price. So you could sell a 100g unit for 90g back, and with your passive income of 50g, now have 140g to spend. However, you could stay on the troop and buy upgrades for it. Such as with a Tribal Warrior, you could choose to stay on the troop and buy a higher troop count, better armor, a better weapon, a spear, ect. This means that switching classes would result in lost potential value, but allow you to counter-pick the enemy better. It also means that your opponent is likely to bring a similar comp, giving you more information and guided decision making.
Maps
Maps are really simple: we just need more. Opening up mapping to the community would immediately solve this problem, but quality control would be necessary.
Last edited: