The long delayed Medieval 2: Total War patch came out around the third week of June a few weeks after the Labor Day weekend it was promised. I’ve been playing it for a few weeks to exercise the new artificial intelligence (AI).

The AI and game rules still need some major improvement to be a challenge to the expert strategist and tactician even at the highest difficulty levels. I was able to satisfy the victory conditions for the ‘long’ campaign playing the Holy Roman Empire on turn 8. What follows is strategy, tactics and a recount of how to do it. I’ll point the way for how to do it by turn 7.

Strategy Highlights

• Mercenaries provide an instant, powerful army in the early game and allow your generals to be in two (or three! places at once)
• Buy mercenaries either next turn (if short on cash and not in danger of crushing attack and can’t attack your target this turn), immediately (if not available on the other side of the border), later (to preserve movement points and avoid zones of control) or multiple times (to increase the chance of winning a mercenary captain)
• Pillaging cities which loses key population and income potential in a long game provides a very high amount of initial capital in a short game
• Bypass some close targets to arrive at the furthest targets in time to win and send reinforcements from the rear (or the front!) to take the bypassed targets
• Buying ships to move along the coast will sometimes save a turn versus moving them there overland
• Disembark ships with units with most movement points first
• Bring generals around to the far side of cities to position them to attack the next city and to allow junior troops to form new generals
• Break sieging troops into groups so that some of them won’t be near the wall when others are attacked by a sally when facing cavalry troops
• AI opponents will not buy up mercenaries to starve your advancing generals of new units
• If your general’s unit or horsemen move to attack a city and moves independently from the rest of the force, it can be joined later by the rest of the army exhausting their movement points to build siege equipment to take the city or to attack the city if the siege equipment is already built or is part of the army
• The same attack/join strategy can be used to add units with movement exhausted to an attack on a village without walls by ‘laying siege’ to the village rather than immediately attacking it
• Units part of a sieging army can break off to save movement points to attack the next city or be pre-positioned as a non-sieging supporting army
• Not much point in building a lot of economy if the payoff has to be in 7 turns or less
• It’s often cheaper and faster to liquidate units in one location and build them in another rather than move them
• If you attack a unit next to a city, the city army sallies to support it. If the enemy army is sufficiently depopulated before being routed, the city will be empty and may be captured in one turn without a spy and potentially recruit several units that turn
• Spies did not appear to be a good investment in such a short game. Even the opening spy is probably best killed off
• Often the AI will attack you to sally on its turn especially when faced with an equal or inferior force in its (flawed) opinion
• If an army without a general breaks into several regiments, only some of them will turn rebel; if each subgroup can’t defend itself, it’s easy to create a man of the hour from the remaining units by attacking the rebel (that also doesn’t move)
• If you are at war with lots of factions, your generals seem more likely to win a mercenary captain so consider attacking factions for the first time before buying mercenaries
• Liquidating military installations within castles is a great source of capital for that last unit or two
• A single cavalry unit can lay siege to a city that a very large army of foot soldiers cannot break
• A single fast foot unit (archers or sometimes peasants) can lay siege to a city that a very large army of foot soldiers cannot break
• If you give enemy diplomats tempting targets they can’t bribe, they will often prefer them to other targets that are easier to bribe
• It’s often possible to retrain units in a newly conquered city even if no new recruits are available
• Recruiting cheap units can allow garrisoning expensive units to head to the front (or be liquidated–but that’s a three or four turn payoff) making mustering halls one of the few useful builds
• Carefully combining veteran units with casualties can create more experienced units, move experience to multiple units where it can be retrained, move experience to smaller units where they can be retrained to create a unit with more net experience, or moved to a larger unit to better leverage the value of the experience, but this risks losing movement points or postponing the turn the unit can be retrained
• If two cities are a couple squares less than a single march apart, rear garrison troops can come to the front to join an attack, then a unit from the victorious sieging army can return to the rear to garrison the rear city
• Armor is not a good investment in such a short game
• Move your capitol to the front to allow princesses’ husbands to appear at the front if your finances aren’t too severely affected in case your princess finds someone suitable
• Setting city management to ‘financial’ and leaving extra garrison troops in cities to boost morale maximizes revenue minus cost
• Pay attention to a city’s ‘free’ units
• A mine in Vienna can pay for itself in 5 turns which is probably a bad investment, but all other mines take 12 or more turns to pay for themselves and are terrible investments
• If a ship is attacked, it can often reach its destination (or a better one!) sooner
• AI will not refuse a battle against comparable faster dominating troops (e.g., archers accept battle with horse, melee troops accept battle with archers); the rules should probably prevent such refusals because horse should be able to force engagement with slower troops in short pursuits, but the AI doesn’t take advantage of the fact that the rules don’t
• An effective one-two punch is to lay siege to a city with part of your force, whittle down the external defenders, remove the siege and attack the defenders now outside the city and blast the city’s defenders when they sally to the aid of the routed defenders; night attacks can also be used to prevent enemy support
• The first (and sometimes only) build should be a road or a port
• Land clearance, grain exchange and town hall might pay for themselves in 7 turns, but all three-turn builds seem obviously quite bad on this time scale
• Units disembarking ships can break through enemy zones of control
• The Pope and other powers do not react or react too slowly to your instant unipolar power grab; the Pope often only excommunicates you after you attack him and the next power in your path is more likely to propose a trade agreement and give you map information than to declare war on you before you attack
• There were no counterattacks in such a short game
• Basic strategy seems to continue to be effective from 1.0 and 1.1
• It still remains easier to attack a target with tall walls than a palisade–the rules should allow ladders to be used against palisades; I want some pole vaulters too
• Leave a city empty with a pending unit build (or even without one if there are no enemy nearby; even a very hostile city usually takes a couple turns to successfully revolt)
• Block a hostile mobile army from breaking your siege by detaching a force to a bridge or defile that can be supported by the sieging troops or building it a fort in the way
• Break through a hostile screening or blocking force by bring the bulk of units to meet it on the path to the target, then detaching one unit to initiate the attack which preserves movement points for the main force while often creating a man of the hour
• In the early game, fire forces that are too far from the front, force them into action that will reduce the number of soldiers to pay, or substitute them into armies instead of foot soldiers and fire the foot soldiers (especially the two initial units of heavy cavalry)
• Archers, peasants and rump depleted veteran units are cheaper to garrison than full units of soldiers or cavalry
• Archers and crossbowmen are the best value for the money if cavalry can be kept off of them because they are cheap, they retain their experience well, they lose few forces in an engagement, they have great special attacks (super range for crossbowmen and fire for archers), they can capture some enemy routing troops, stay out of range of foot soldiers, engage foot soldiers so they can be decisively flak attacked by cavalry and they can be used as skirmishers and support troops to mop up after killing the bulk of the enemy
• Destroying a faction caused all of its ships to sink
• Siege factories take too long to build in such a short game
• You can sometimes move ahead behind enemy lines to buy mercenaries then come back to the front (or bypass) in the same turn
• Build a fort or a chain of forts between cities (most useful between cities east of Hamburg and Prague and in France)
• Count backwards from when you want to win and what you can get to the front in time; this avoids wasteful building and unit builds
• Move some or nearly all of besieging troops forward to the next target before the attack if the new target is not in the same direction as the attack if you have sufficient reinforcements to bring forward to take over the siege

Tactics Highlights

• Dropping ladders and picking them up with a unit with 9-35 soldiers breaks the ladder group into two or more groups allowing for example four different units to have a single ladder
• AI sallies units out en masse and does not adjust to a barrage of arrows or bolts; this can smash the sallying force’s best unit or two such as cavalry or a ranged attacker when the enemy has no cavalry
• The AI will now go to loose formation better in city and village defenses which suggests alternating ranged/melee tactics
• The last unit out of a city still leaves the door open
• The AI still sallies with its entire force even when facing cavalry when it has no cavalry allowing the simple tactic of running to the corner of the map, then capturing the square
• Cavalry defending cities still does weird stuff like going outside when all units are inside or sallying to attack alone against a group of attackers
• A strong general can hold the central square and fight off entry into the square from one direction for thirty to ninety seconds
• The AI will enter and will not break off an uphill battle until it gets routed
• Archers on the city walls cannot be attacked by horse and with a sufficient morale advantage cannot by assaulted by melee troops either
• Note well where enemy reinforcements will enter the field from the strategy map (center on defender in meeting engagements, the city in city attack/sally engagements) so that this group can be cut down immediately in detail by well positioned archers and medium to long range and flanking melee and cavalry troops
• The AI gets confused by a group of melee troops splitting into two directions to scale the walls
• The AI will not use sally tactics when it is defending an attack so archers are usually safe outside the walls in siege attacks
• Archers and crossbows in a high altitude defense can use a combination of obstructions, screening melee troops, fire arrows and flank and rear attacks by cavalry to rout close attackers and steadily kill the rest through superior range
• Certain maps have walls that allow ranged attacks on units defending central squares which allows ranged troops to best cavalry if they make it to the city wall because the cavalry won’t stay out of range
• It’s possible to position troops down alleys with melee troops defending a slope or outside the castle on the Palermo and Ragusa maps to attack the central square
• With the exception of putting your crossbowmen at point blank range in front of a breached wooden wall, almost all basic tactics from 1.0 and 1.1 seem to work
• Defenders still can’t tip over ladders even if there are no troops on them
• Pay attention to where your general is in your general’s unit and face him away from the battle (typically, unless you want to kill him off)
• The AI still gets confused by assaulting troops with ladders and will eventually leave an open wall to scale if you move the troops around outside
• Your cavalry and melee troops can’t attack units that are routing in front of doors, units that are routing off of walls or units that have not yet entered the map by right clicking them all the time, but you can position your troops to fight and kill them or shoot some of them with range attack troops
• If you can afford the damage, you can poke lots of holes in a palisade with a ram to allow your cavalry to enter the city without a frontal assault
• Mutually supporting separated range troops with overlapping fields of fire are a very good unit placement and works almost as well as it did against the Spartans in the Peloponnesian War

Coming soon, play by play.

This entry was posted on Saturday, July 14th, 2007 at 1:00 pm and is filed under Strategy, Ancient Wisdom, Game Review. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.