Posts from Analytical articles on Heroes II Gold by Qwerty.

Brief Introduction.
This section is a collection of various reviews that I wrote between the fall of 2004 and mid-2006. They are both reference and analytical in nature. Most of them are by no means indisputable, and are more open to discussion than reflecting the "final truth." Consequently, the value of the materials in this section is also ambiguous – some may find something interesting or useful, while others may find it questionable, to put it mildly. This is due to differences in playstyle as well as differences in skill level. Therefore, let's agree from the start: as many people, so many opinions.
And one more thing: this section is specifically a collection of articles, not a single analytical work on HMM2 Gold. It is quite possible that it will be expanded...




Let's begin the discussion about secondary skills and some ways of hero development.
A hero can have a maximum of 8 secondary skills. Some of them (1-2) are given to the hero initially; others are offered as they gain experience or in special structures on the map, such as "huts."
The assortment contains 14 items. They have three levels of development: basic, advanced, and expert.
Secondary skills can be grouped into the following categories.
1. Travel:
Logistics (affects movement distance by increasing movement points);
Pathfinding (affects movement distance across "difficult" terrain such as swamps or deserts by reducing the penalty for such movement);
Navigation (affects movement distance at sea by increasing the number of movement points on water).
Scouting (affects the area visible around the hero on a closed map).
2. Magical.
Wisdom (provides the ability to learn higher-level spells). Additionally, wisdom may be required to pick up certain artifacts.
Mysticism (increases the amount of magic regenerated between turns).
Eagle Eye (gives a chance to learn a spell cast against the hero by a defeated opponent).
3. Combat.
Archery (increases damage from ranged attacks and, more importantly, allows firing through fortress walls at full strength).
Leadership (at high levels, a unit has a chance to act twice in one turn; at low levels, it may not act at all). Sometimes leadership is required to take an artifact.
Luck (at high levels, a unit has a chance to deal double damage; at low levels, half).
Ballistics (needed during the siege of castles, affecting the speed of wall destruction).

In addition, there are three completely specific skills:
Estates, which brings a certain amount of gold per day into the treasury;
Necromancy, allowing one to add skeletons to their army (if space is available) in a quantity depending on the number of enemy units killed;
Diplomacy, affecting the ability to hire roaming armies on the map, the price when surrendering an army, and the possibility of earlier training in the specific "Xanad" building.

The topic of secondary skills is closely linked to the issue of hero development.
Let's try to provide some classification of heroes depending on their intended use:
1. Main Hero - Powerhouse.
2. Main Hero - Mage.
3. Second Hero – Powerhouse 2.
4. Second Hero – Mage 2.
5. Scout-Collector-Supply Runner.
6. Navigator-Collector-Supply Runner.
7. City Defender.
8. Main Hero - Necromancer.

Now let's try to draw a portrait of each of these heroes.
1. Main Hero – Powerhouse.
Logistics, wisdom, luck, leadership, ballistics. Depending on the map, pathfinding and/or even navigation are possible. If there are serious shooters in the army – archery. Mysticism may be included.

2. Main Hero – Mage.
Wisdom, logistics, mysticism (many will start criticizing me now). Depending on the map, pathfinding and/or even navigation are possible. If there are serious shooters in the army – archery. Ballistics wouldn't hurt, especially if it was obtained from a special object rather than through leveling. If troops from different castles are planned – leadership.

3. Second Hero – Powerhouse 2.
Logistics, wisdom, luck, leadership, estates, possibly ballistics. Depending on the map, pathfinding and/or even navigation are possible. If there are serious shooters in the army – archery. Diplomacy may be included. Mysticism is possible.

4. Second Hero – Mage 2.
Wisdom, logistics, mysticism, estates. Depending on the map, pathfinding and/or even navigation are possible. If there are serious shooters in the army – archery. If the hero goes to unknown places, scouting may be used. If troops from different castles are planned – leadership. Diplomacy may be included.

5. Scout-Collector-Supply Runner.
Logistics, pathfinding, and depending on the map – navigation. Estates, scouting, wisdom (so that in case of an unexpected "attack" by an opponent, one could "fire off" some magic before surrendering), diplomacy.

6. Navigator-Collector-Supply Runner.
Navigation, logistics, and depending on the map – pathfinding. Estates, scouting, wisdom (so that in case of an unexpected "attack" by an opponent, one could "fire off" some magic before surrendering), diplomacy.

7. City Defender.
Wisdom, leadership, luck, archery, estates. Diplomacy is possible (it can be very useful when applying the "hit and run" tactical maneuver).

8. Main Hero – Necromancer.
Necromancy, wisdom, mysticism (again I hear indignant criticism), logistics, pathfinding and navigation depending on the map. Luck and ballistics can also be taken.

Well then, let's try to figure out where these heroes come from—the ones whose ideal (but not real!!!) portraits we tried to draw.

Powerhouse 1. Usually develops from a knight or barbarian.
Let's consider both options.
Barbarian. 3-1-1-1. That is, attack starts at 3. Not bad. Defense 1. Not much, but better than "0". Advanced pathfinding, 7 slots free for secondary skills. If the game starts on snow, for example, that's also not bad. In principle, the "pathfinding" skill is only needed on "difficult" terrain, such as snow, desert, swamp, etc. Attack levels up quickly. The ideal way to level: wisdom at the first opportunity, logistics, leadership, and if there are shooters – archery. Then ballistics can be taken.
Knight. 2-2-1-1. Starts with 2 attack, 2 defense. Not bad. Leadership, ballistics. Ballistics isn't too necessary yet, but will come in handy later. Defense grows quickly during leveling, attack slightly slower. Ideal leveling: wisdom at the first opportunity, logistics, then if terrain is "difficult" - pathfinding, and if there are shooters – archery.

Mage 1. Most often develops from a white mage, black sorcerer, and less frequently from a sorceress.
White Mage. 0-1-2-2. Defense 1 is still better than "0". Advanced wisdom, 7 free slots. And this is the main thing. Ideal leveling: logistics, archery (usually there are shooters), mysticism will likely be squeezed in almost regardless. If there is a lot of "difficult" terrain – pathfinding. Then ballistics can be taken; otherwise, many shooters will die during castle assaults.
Black Sorcerer. 0-0-3-2. Wisdom, scouting. And this scouting slightly spoils an almost perfect picture. Magic power levels up quickly, knowledge slower. Everything would be fine, but while there is no magic, it's hard to fight with this hero – attack and defense are zero. And once you've developed well – scouting isn't very necessary... Ideal leveling: logistics, if there's a lot of "difficult" terrain – pathfinding, mysticism. It's better to wait for the rest; later maybe ballistics and even leadership (if dragons are present). Necromancy should not be taken.
Sorceress. 0-0-2-3. Wisdom, navigation. Knowledge levels up, magic power slightly slower. Can become very strong with magic like "door", "berserk", etc. Usually used at the "home" castle or on a map with a large amount of water.

Powerhouse 2. Differs from Powerhouse 1 by having less strength; usually used in pair with Mage 1. Other differences: the role of leadership and diplomacy increases, ballistics is less needed.

Mage 2. Differs from Mage 1 by having less strength and a greater focus specifically on magic. Works well as a black sorcerer – at first runs around and scouts, levels up for free, then, when magic is built – goes into battle. Generally usually develops from the starting hero of magical castles or is bought in a castle conquered by Powerhouse 1 with developed magic. The role of ballistics decreases.

Scout-Collector-Supply Runner.
Needed to open up territory at the start of the game, collect unclaimed resources, and in the mid-game take over enemy mines, as well as for transferring troops and artifacts to main heroes. Based on this, they must run fast, have diplomacy to pay off in case of an unexpected "attack" by a strong enemy hero; Estates would also be useful for daily gold contributions to the treasury. Level logistics, pathfinding, possibly scouting, diplomacy, estates. Then, in case of powerful magic, wisdom so that if necessary one can "thin out" the opponent's troops. Can result from any hero. For example, a barbarian or black sorcerer.

Navigator-Collector-Supply Runner.
Sails the sea, collects what is in it. Then transfers artifacts to main heroes. If the main heroes are fighting "across three seas" - delivers troops there. Main skill – navigation. Logistics and diplomacy can also be taken. Usually results from a sorceress.

City Defender.
Needed for city defense and collecting resources near the city (such as mills, leprechaun gardens, external creature dwellings). Skills that allow holding a defense are needed: archery, leadership, luck. When the magic guild is built – wisdom. For applying the "hit and run" tactical maneuver, diplomacy wouldn't hurt. If there is good magic in the castle – can be made from a magical hero; in other cases – knight or barbarian.

Main Hero – Necromancer.
Where the necromancer comes from is clear. Stats – 1-0-2-2. Skills: wisdom and necromancy. Level logistics, pathfinding for "difficult" terrain, mysticism, ballistics, and luck if given.


Today I want to return to the story about secondary skills.

So, a map of size "B" or "O-B". We have a castle, in it one hero of the same race as the castle with starting troops. Then another hero is bought, then another. Usually one or two heroes conduct battles; others simply run around with two SEPARATED (mandatory) units with speed NOT LOWER THAN AVERAGE and collect resources. Two units are needed so that in case of a sudden attack by the computer there would be a chance for the hero to escape, preserving him and not giving the enemy hero an extra 500 experience (and artifacts, if they have any). Then one of these heroes can become the second main hero.
What does a hero need for a successful foray deep into unknown territory?
They need to run fast (logistics – extremely important skill) and not be slowed down (if pleasant terrain like snow or desert is visible on the map – they need to become a pathfinder; on sea maps – "navigation").
They need to be able to reach castle defenses in time (magic "town gate", fourth level), catch up with a nimble opponent, or get away from superior forces (magic "door" - "Dimension Door" - fifth level) – meaning they need to become wise by obtaining the corresponding skill. Moreover, if there is a "pyramid" structure on the map, then, after dealing with 2*10 vamp-lords and 3*10 king mummies, you get a fifth-level spell. However, if third-level wisdom is not present, the spell will not be obtained. And if this pyramid was already visited earlier – you get minus two to the luck stat.
But how can one notice mines, resources, enemy heroes, castles, etc., for the first time while advancing through an unknown land? It turns out (now most players will disagree with me) that one needs to see far, therefore the "overview"/scouting skill is needed. I want to note immediately that in the case of advancing with two heroes at once (one with fast "ones" scouts the territory and takes resources/mines), "scouting" can only be held by the scout himself.
Practice forced me to reconsider my opinion about this skill (which is usually called useless, along with "eagle eye", "mysticism", and "diplomacy" - by the way, the last two skills are quite interesting). An enemy hero not seen in time, a pile of key resources, or an artifact can lead to a significant prolonging of the game, and in the case of playing against a human, it can even lead to defeat!
Castle assaults often turn into long, drawn-out battles. And when three invincible shooters (central and two corner towers) plus a moat are on the defender's side, and moreover infantry cannot reach the enemy... In this regard, ballistics would not hurt the main heroes at all!
A few words about other skills.
Leadership. We remember that the maximum possible morale is "+3". In principle, in separate cases, morale "+4" may be useful - if we are playing against bone dragons (these cute creatures are so terrifying that the morale of those fighting them decreases by one). In principle, there are two opinions regarding the usefulness of this skill. Some say that morale can be raised by visiting various objects on the map; moreover, if all troops are from one race – it's already "+1", and there are plenty of artifacts... Others logically argue that you won't visit objects before every battle; moreover, it's not always possible to have troops from one castle – some join along the way, some we bribe on the map or in captured castles. Another very reasonable argument is made that in HOMM2 the number of artifacts a hero can carry is limited, and when all slots are occupied, artifacts obtained in battle, for example, simply disappear. If you have leadership – you can free up some morale-boosting artifacts by transferring them to another hero. In general, whether to take leadership or not is a matter of personal taste.
Luck. Can take values from "-2" to "+3". Also a matter of taste. In my view, if the game focuses more on "power" - leadership and luck are very relevant; if it's about "magic" - they aren't as necessary.
Archery. A lot depends on the army here, specifically on whether it has shooters. When playing as Castle, black dragons or Necromancers don't need shooting much (only one shooter), but others might find it useful. The main advantage of this skill is removing the penalty when shooting through fortress walls.
Mysticism. Well, a lot depends on the map, the presence of wells and other mana replenishment spots. This skill is convenient, although it takes up a slot (the maximum number of secondary skills is eight).
Diplomacy. Unlike HOMM3, where diplomacy is one of the strongest skills (many call it unbalanced and cheat-like), in "Heroes II," diplomacy does not affect the free joining of neutral troops. Moreover, the first two levels of diplomacy often cause irritation: they offer to buy troops at four times the price—what kind of robbery is this!! However, diplomacy, especially expert, has a number of undeniable advantages. Firstly, you can pay off 2.5 times cheaper (let's be honest—it can be relevant, although there is an artifact "feather" that allows paying off 5 times cheaper!), secondly, you can buy some neutrals right on the field (not all!), thirdly, you can visit the special building "Xanad" (increases all skills by one!) starting from level 4! (with advanced diplomacy - level 6, with basic – level 8), whereas without diplomacy, entry is only from level 10.
Estates. 100 gold per day at the basic level, 300 gold at expert. Very good for secondary heroes, especially at the start of the game. However, for main heroes, especially on large maps, a slot for Estates is an unaffordable luxury.
Necromancy. Good only for innate Necromancers. But very good!
Eagle Eye. Very rarely taken by human players. Theoretically, the potential use of this skill is understandable. An over-protected AI castle, almost total absence of resources for the guild (everything goes to building the last level of troops, a meager map), and only in battles with computer heroes is there a chance to learn magic. But nevertheless, this skill is usually taken only as an alternative to completely useless ones (navigation on a map without sea, pathfinder on a map without "difficult terrain", "observation" on a fully open map, etc.).

Let's try to summarize some results. They are paradoxical. It turns out that every hero is relevant depending on the situation, and even almost necessary. Lots of snow or deserts – you need a Barbarian (advanced pathfinder); there is sea and access to it – a hero from the phoenix castle is necessary (navigation); for scouting and early magical combat, a hero from the black dragon castle is highly relevant (observation and good spell power); a universal shock trooper for taking castles – Knight (leadership and ballistics); the self-sufficient Necromancer who needs no recommendations; if a pyramid is visible in the foreseeable space – a titanic hero (advanced wisdom, which very quickly becomes expert). Such conclusions turned out to be unexpected...
To cast a spell, you must know it (it must be "recorded" in the hero's magic book) and have enough "mana points" to pronounce it. The maximum number of mana points is knowledge * 10 (exception – visiting a special source on the field, after which mana=knowledge*20). Mana can be restored to maximum by visiting a well or when the hero stops "for the night" in a town/village that has a magic guild. The secondary skill "mysticism" and some artifacts affect the restoration of magic.
To record a spell in the magic book, it is necessary:
1. Have this book;
2. Learning third-level spells – basic wisdom;
3. Learning fourth-level spells – advanced wisdom;
4. Learning fifth-level spells – expert wisdom.

Spells can be learned:
1) in the town's magic guild;
2) in special buildings on the field, including pyramids (guards – 20 vamplords and 30 king mummies);
3) from a defeated hero, if you have the "eagle eye" skill of the corresponding level and the hero used this spell in the lost battle;
4) HOMM2 Gold – by collecting the artifact "Armor of Anduran" - "Town Portal" magic.

And now let's briefly go through the spells that are most interesting in our opinion.
First level.
Slow (very strong if there are shooters);
Haste;
Heal;
Dispel Magic.

Second level.
Lightning;
Blind;
Cold Beam;
Summon Boat.

Third level.
Animate Dead/raise corpse/Animate Dead – a terrible weapon of the Necromancer, allowing the animation of undead killed in battle. A very, very strong spell.
Anti-Magic (one of the best spells in the game);
Wave of Death – another weapon of the Necromancer, affecting all living things except dragons;
Earthquake – a replacement for ballistics, destroys walls during an assault;
Paralyze;
Battle Teleport – allows moving a unit to a free space during battle. Especially interesting for hydras.
Mass spells: haste, curse, bless. With large armies, they can decide the outcome of the same battle.

Fourth level.
Chain Lightning – one of the strongest spells in the game;
Berserk – good against strong armies;
Elemental Storm – a mass combat spell;
Town Gate – moves the hero to their nearest unoccupied castle;
Mass spells: shield, heal, slow!!!.

Fifth level.
Dimension Door (teleport, portal) – according to many opinions, the strongest spell in the game;
True Resurrection – reviving those killed in battle;
Armageddon – the most powerful mass direct damage spell;
Clone;
Summon Elementals.
As well as the rarest spell in the game - Town Portal (by the way, in HOMM2 Gold it appears for a hero who has collected the artifact "Armor of Anduran").

General rule – with small and medium armies, direct damage spells play a large role (arrow, cold beam, lightning, ring, fireball, etc.), with large ones – mass spells (haste, slow, curse, bless) and berserk. Naturally, spells such as blind and paralyze are especially effective when the opponent has 1-2 super-strong units, and heal and dispel magic – when opponents use the aforementioned spells on your troops.
When fighting an enemy mage, the use of the "anti-magic" spell can be of invaluable help.

Typical cases.
A bunch of shooters + mass slow.
Knight troops + mass haste.
Opponent's Knight/Barbarian troops + berserk.
Necromancer + animating undead.
Necromancer + wave of death.
Dragons (except bone) + armageddon (or elemental storm).
Phoenix + elemental storm.
1. Magic.
In cases where the opponent has no magic-immune troops, the berserk spell works wonderfully.
Mass slow is cast on the first turn.
And starting from the second turn - berserk. Cool, especially considering that the responding unit also removes some of the berserking effect!
2. Shooting with a fast unit.
Idea - shoot with titans, move the unit slightly forward, retreat with titans, shoot again, etc.
Works against slow non-flying troops (or if you can cast slow on flyers).
Main danger - morale for opponent's troops + shooter ran out of arrows (by the way, there is an artifact that allows shooting forever).
3. Hit and Run and its variations.
Essence - attack the opponent with a large number of heroes.
Methodology of a classic hit and run: a hero (ideally with 5 gargoyles) attacks an opponent's unit. Shoots magic as much as possible and pays off. After that, they are bought again and history repeats until mana runs out.
Variations: different heroes are used.
Hit and run is sometimes also applied against magic-immune troops. The point is to attempt a prolonged attack (green, red, black dragons, phoenixes, cyclops) so that the enemy troops burn themselves out.
4. Wearing down the opponent, or guerrilla warfare.
On maps with many castles and an AI hero who has very strong troops, you can try to force them to capture castle after castle. The idea is simple - besides the losses they will incur, they will leave part of their army in every castle. Thus, the army will get smaller and smaller...
5. On maps where the victory condition is taking a castle guarded by a strong hero with a serious army, you can try a hit and run variation.
Approach the castle with, for example, 5 heroes: the main one with an army and four with 5 gargoyles each. And attack. First with those four. They might not deal much damage, but the AI will spend quite a bit of mana. Then finish them off with the main hero (after visiting morale and luck boosters).

It seems to me that it's time to talk about some typical situations in the game. Naturally, they don't always arise, but, as I believe, quite often.

1. Fighting on the field with neutral troops with small and medium armies.

The main feature is the absence of magic for the opponent, as well as the inability to "pay off" (meaning, in case of loss, the troops will be gone), you can only "flee".
So, a typical situation: a mine, passage, artifact or something else is guarded by a group of neutral troops.
First, check how many there are (right mouse button).
Then, in the absence of the "vision" spell, try to assess the chance of them joining your army. If such a chance exists, you must leave an empty slot for these troops (however, if the army already has the same troops, an empty slot is not needed).

Suppose there is confidence that fighting will be necessary.
Then move on to battle preparation.
First, check mana and think about which spells can help.
Then assess the characteristics of your own and enemy troops and arrange them accordingly.

Example:
Starting mage troops (8 centaurs, 3 gargoyles) against halflings.
Magic – curse.
Arrangement: 8 centaurs, 1 gargoyle, 1 gargoyle, 1 gargoyle.
Game:
Curse on the second unit of halflings.
A gargoyle flies and stands between the first and second units of halflings without attacking.
The second gargoyle attacks the fourth unit of halflings from the side of the third unit of halflings. It is countered.
The third gargoyle attacks the fourth unit of halflings without a response from the side of the fifth unit of halflings.
Centaurs shoot at the third unit of halflings.
Then the battle is played out simply.
No losses.

I note that if the halflings wanted to join, an empty space was left for them.

Good day.
The more I write my tips-notes, the more it feels like some dialogue in an empty room... Who am I writing to, who will read this? Why does no one answer me, ask me about anything? Or maybe I'm alone here?

Well then, let's continue.
So, we are talking about some typical game situations.
Today we will continue the discussion of fighting on the field against neutral troops.
I remind you that almost all shooters in HOMM2 (except mages, archmages and titans) deal half as much damage in melee combat.
Thus, you must block them!
Order of actions:
-) assess armies. In doing so, it is necessary to consider the probability of a pleasant surprise appearing in the central stack – an upgraded unit (in some cases this can very seriously affect the outcome of the battle);
-) try to calculate the sequence of moves taking into account the use of mass haste and mass slow;
-) arrange your own so that shooters are blocked as quickly as possible. At the same time, think about whether troops with "slow" speed or lower are needed in this battle.

Tactics for conducting the battle.
-) blocking shooters and neutralizing them with magic or distracting units (like a single halfling);
-) combining blocking with the destruction of individual stacks;
-) moving to complete destruction by shooting or bringing in medium-speed troops.


The topic of this review was born as a result of analyzing gameplay on maps of size "L" and "XL" (large and very large).
So, we have some starting castle, starting hero, game level "impossible" (chess king icon), zero money and resources.
During the first week, we have two or three heroes (or maybe more), 4-5 levels of troops built (in separate cases, when playing as a Knight castle, levels 1-4 and 6 may be built).
What next?
Next, you need to examine the area thoroughly. We will be especially interested in:
1) presence of sea and access to it (neutral ships, castles/villages by the sea, "summon ship" magic);
2) presence of neutral castles/villages and their class;
3) presence of AI castles/villages and their class;
4) presence of creature deposits;
5) presence of objects like "ford on the hill" and "foundry" (for the POL version – also a stable).

Looked around? Now you need to analyze the situation and make adjustments to your immediate plans. You should think about the following:
1) is it time to sail and catch some wood in the sea (a couple hundred logs), some money (several thousand) and some artifacts;
2) is it time to build a thieves' guild and see who guards the villages/castles;
3) assess the possibilities of building up these castles considering current and potential resources;
4) see if there is an opportunity for free upgrades of several creatures in buildings on the field, which will save money not only during recruitment but also by not building unnecessary structures in your own castle!

Practice shows that troops from two castles (titans and black dragons) are practically self-sufficient, while others would benefit from reinforcements.
When forming a combined army, it makes sense to pay attention to the following factors:
1) army speed (which is determined by the slowest creature);
2) the morale of the resulting army3) the possibility of actually building a second/third castle and buying out creatures.

Black dragons would be an excellent addition to Knight and Barbarian armies. However, in reality, only Knights can usually afford to build dragons on most maps. The reason is the need for a large amount of stone to build both a Barbarian castle and a Black Dragon castle.
Thus, we get one combination: starting as Knights, capturing a dragon village, and building dragons there.
If in the first week we built the first 5 levels as Knights (paladins didn't work out), captured the BD village in the second week, and see that there is still plenty of game left and the opponent is very far away, then you should think carefully. What to think about? Whether you should have placed paladins at all!!! Perhaps, having a strong Knight army reinforced by gargoyles and/or griffins, play for a couple of weeks, and in the meantime try to build dragons! Interested in the price? Just 25000 + 40 ore + 30 sulfur + 10 gems – and you have little green dragons. For black ones, however, you'll need a bit more - 35000 + 50 ore + 50 sulfur + 10 gems. And if you consider the almost mandatory statue and basement – add 15 ore, 5 wood, and 4250 gold. The stone quarries will be well-employed! If a village is being built – add another 20 wood, 20 stone (wow) and 5000 gold. What, are the dracos too expensive, do they require too much stone? But what a reward!
Another option, less spectacular, but far less costly and faster.
What do we have playing as a Knight? Good morale and a lack of special resources (sulfur, crystals, mercury, gems) to build the first five levels. What follows from this? It follows that by spending 17000 + 25 wood + 15 ore + 5 mercury + 5 crystal + 15 sulfur + 5 gems (with a village – another 5000 gold + 20 wood + 20 ore), we get cute little dragons, though they consist exclusively of bones. If there is a seafarer – a supplier of wood and money – then you can try to place both dragons and paladins.
Let's go further. How realistic is it to get a Titan castle as an addition to the Knight castle? This depends heavily on having enough leprechaun gardens in the controlled space and whether exactly 5 gems, rather than 500 gold, drop from them. For giants you need 23300 + 25 wood + 20 ore + 5 mercury + 5 sulfur + 5 crystal + 25 gems; for titans - 35800 + 30 wood + 25 ore + 5 mercury + 5 sulfur + 5 crystal + 45 gems. If giants appear in the third week and titans in the fifth, then with the "Archery" skill and cheap Knight troops – it's serious. But a lot depends on gems; moreover, you need quite a bit of gold too.
Phoenixes are good support for Knight troops. The cost of building them (19000 + 15 wood + 45 ore + 20 mercury + 10 gems) is not that great, and they fly. True, they die off pretty quickly, but something is better than nothing. Moreover, crystals are not needed for them (!), which allows one to hope for building both paladins (2nd week) and phoenixes (3rd week) – now that's optimistic!
I believe it's clear that building a Barbarian village from scratch is not very interesting for Knights – competing resource (crystals) and as a result – almost the same unit (crusaders are nearly equal in strength to cyclops). However, if there is a "fort on the hill" nearby, then in the absence of other options, one could think about building up to ogres with the idea of upgrading them for free at the fort.
Now let's talk about Barbarians. Here, you need a lot of stone for building. And this immediately makes phoenix and black dragon castles almost unrealistically difficult to build in parallel. But titans... Imagine titans protected by Barbarian troops during an appropriate attack! And what if there is also the Archery skill? Wow, what power! True, until the titans mature, it will be tough.
Bone dragons – not bad, though morale will suffer slightly. And quite realistic.
Knights... Very doubtful...
And who can reinforce Necromancers? The ideal option is, of course, titans. Remember one of the stages of the classic campaign for Archibald? Titans protected by ever-growing hordes of skeletons – that's power!
Well, what about elves? It turns out they will have to be reinforced with Knight troops, because you'll have to forget about BDs and even Barbarians (stone), and titans require too many things, while 10 gems are already spent. The only option is to place the first three levels, struggle through them somehow, and build titans. Gems aren't spent, other resources aren't either. The problem is that elves don't upgrade without building druids, and those can only be placed after the magic guild. Together it comes to 4500 + 15 ore + 5 wood. And then the elf upgrade – 1500 gold and 10 wood. Summing up – 6000 + 15 ore + 15 wood. But you also need to erect a Titan castle – 5000 gold, 20 wood, 20 stone. That's how it is, although the option is interesting; moreover, own royal griffins will be added to the army.
In principle, by giving up buildings above the third level, one can try to build dragons too. However, in practice, neutral castles/villages are not always visible in the first week and druids may already be built, which makes building dragons very problematic.
And who can reinforce black dragon and titan troops?
Black dragons are ideally reinforced by titans. And basically, no one else is needed. Except perhaps phoenixes for the first move. I believe it's clear that building these troops at the expense of building dragons is hardly a good idea.
Titans are well-reinforced by champions from the Knight castle, as well as phoenixes. To a lesser extent – vampire lords. Why? Because champions will provide the first move and can protect titans from flying units. Moreover, they are VERY cheap. Phoenixes give the first move and shut down enemy shooters. Vampire lords can shut down shooters or protect against flying shooters.

Well then, let's try to see what we have come up with.
1) When building a castle, think about the development strategy, about what will happen after some time;
2) When taking monsters from another castle into your troops, keep an eye on morale, evaluate troops by potential utility (shooters, flying units), and take what is missing;
3) When forming an army, remember that the hero's movement speed depends on the speed of the slowest unit;
4) Don't forget that possessing the fastest unit on the battlefield gives the opportunity to cast magic first, as well as "shut down" the opponent's shooters and/or mechanically protect your own shooters from melee attacks.
5) Before proceeding to build troops in a captured castle, think about whether they will be needed by the time they are built and if there isn't a better investment of money (buying out medusas, for example).

Today we will try to figure out the peculiarities of shooting units in HOMM2 and consider some tactical aspects of the game in this regard.

Shooters are present in all six castles. We will examine them by levels rather than by castles.

First level.
Halflings. This is the weakest shooting unit in HOMM2. Belongs to the titan castle, health 3, attack 2, defense 1, damage 1-3; in melee combat (as with most other shooting units), it is half as much. Speed slow (3), occupies one cell on the battlefield (later we will see why this aspect is important), shots 12, cost 50. The unit is encountered quite often in the game. This is due to the presence of a special object "burrow" on the terrain; upon visiting it, the halflings living there leave their home and go to war with the hero. Moreover, free of charge, that is, for free.
Centaurs. Belongs to the black dragon castle, being the only shooter there. A good unit for its level; according to the computer, it is the third strongest at its level (after skeletons and thieves). However, there are other opinions, in particular, that this unit is not only the strongest at its level but also more useful than some second-level units. Health 5 (the best indicator at its level, which is surprising for a shooter), attack 3, defense 1, damage 1-2; in melee combat half as much. Speed medium (4), only 8 shots, cost 60 (a good indicator for a shooter), however, the unit occupies two cells. And in most cases, this is not good for it.

Second level.
Properly speaking, shooters. An unupgraded unit from the Knight castle. There are no other levels of shooters in the Knight castle. Health 10, attack 5, defense 3, damage 2-3; in melee combat half as much. Speed very slow (2) - and this is a disaster, occupies one cell on the battlefield, shots 12, cost 150. The unit is encountered quite often in the game – it is found in structures on the terrain, where it joins the army free of charge, that is, for free.
Archers (upgraded shooters). The previous unit, only upgraded. The only shooting unit in the Knight castle. Health 10, attack 5, defense 3, damage 2-3, shoots twice. However, it specifically *shoots*! In melee combat, it strikes once, and twice as weakly. Thus, the same damage from shooting is 4 times greater than from melee combat! Speed medium (4), occupies one cell on the battlefield, shots 24 (excellent, it doesn't get higher), cost 200. Not a small amount, of course...
Orcs. Unupgraded second-level shooters from the Barbarian castle. Health 10, attack 3, defense 4, damage 2-3; in melee combat half as much. Speed very slow (2), occupies one cell on the battlefield, shots 8, cost 140. The unit is encountered quite often in the game – it is found in structures on the terrain, where it joins the army free of charge, that is, for free. It should be specially noted that a "fort on a hill" structure is sometimes found on the terrain, where orcs can be upgraded free of charge, that is, for free – as a result, both speed and health will increase by one and a half times!
Upgraded Orcs (such as "Orc Chiefs"). Health 15, attack 3, defense 4, damage 3-4 (not bad, by the way), in melee combat half as much. Speed slow (3), occupies one cell on the battlefield, shots 16, cost 175.

Third level.
Elves. Unupgraded shooters from the phoenix castle. Health 15, attack 4, defense 3, damage 2-3, shoots twice. Only shoots; in melee combat it deals only one strike, and that is half as weak. Speed medium (4), occupies one cell on the battlefield, ammunition is fine – 24 shots. Cost 250 (however...).
Grand Elves. Health 15, attack 5, defense 5, damage 2-3, shoots twice. In melee combat it attacks once with half damage. Thus, for grand elves (as well as for elves and archers from the Knight castle), shooting attack damage is 4 times higher than melee damage. Speed very fast (6) and this is one of the main trump cards of this unit, occupies one cell on the battlefield, plenty of shots – 24, but cost 300...

Fourth level.
Druids. The second shooting unit from the phoenix castle and beautiful Sorceress. Health 25, attack 7, defense 5, damage 5-8; in melee combat half as much. Speed fast (5), occupies one cell on the battlefield, shots 8, cost 350.
Upgraded Druid (such as Great). Health 25, attack 7, defense 7, damage 5-8; in melee combat half as much. Speed very fast (6), occupies one cell on the battlefield, shots 16, cost 400. Expensive.

Fifth level.
Lich. A shooter from the Necromancer castle; there are no shooters at other levels. Health 25 (and this is the fifth level – simply ridiculous), attack 7, defense 12 (colossal!), damage 8-10; in melee combat half as much. Liches shoot fireballs – balls of fire, and in doing so, adjacent cells are also hit besides the target (regardless of whether they are your own troops or enemy). Speed fast (5), occupies one cell on the battlefield, shots 12, but cost 750. Very expensive.
Upgraded Lich (such as "Strong"). Health 35, attack 7, defense 13 (super, though it doesn't help against magicJ), damage 8-10; in melee combat half as much. They shoot fireballs just like their "younger brothers," dealing damage to everyone regardless of whether they are friend or foe. In some cases, this is not very good. Speed very fast (6), occupies one cell on the battlefield, shots 24, but cost 900!!! According to the computer, this unit is the second strongest shooter in the game. There are other opinions.
Troll. Another shooter from the Barbarian castle. Health 40, attack 10, defense 5 (a bit sad), damage 5-7 (low for a fifth level), half as much in hand-to-hand combat. However, this unit has a special property: after each round of battle it regenerates, meaning the number of lives of a wounded unit is full again (40)! Speed medium (4), occupies one cell on the battlefield, shots 8, cost 600.
War Troll. Health 40, attack 10, defense 5, damage 7-9; in melee combat half as much. Regeneration, occupies one cell on the battlefield, speed fast (5), shots 16, cost 750. Not cheap.
Mage. One of the shooters of the titan castle. Health 30, attack 11, defense 7, damage 7-9; does not depend on the type of attack! Speed fast (5), occupies one cell on the battlefield. Shots 12, cost 600.
Archmage. Health 35, attack 12 (excellent!), defense 8, damage 7-9; does not depend on the type of attack! Speed very fast (6) and this is pleasing, occupies one cell on the battlefield. Shots 24, cost 700. Additionally, there is a chance (one in five) to remove a spell from the attacked unit! An interesting unit, though not easy to build up (the problem is that to upgrade mages it is necessary to build the special structure "library," and for that you need gems, just as for building the main pride of the castle – titans).

Sixth level.
Well, here only one reigns...
Titan! The best shooter in the titan castle and the best shooter in the game "Heroes of Might and Magic. Part 2". Health 300 (only black dragons can boast such an indicator), attack-defense 15 (the highest indicators in the game), damage 20-30 (a bit low, less than even bone dragons, not to mention their colored brethren) does not depend on the type of attack. Speed very fast (6), occupies one cell on the battlefield, shots 24. Mind magic does not work (paralysis, blindness, berserk, hypnosis). An excellent unit, but extremely difficult to build up, and moreover, the cost is the maximum in the game: 5000 and 2 gems!

Let's try to summarize some points.
1) Shooters are generally more expensive than non-shooting units.
2) Shooters generally have fewer health points than non-shooting units of the same level.
3) Shooters (with the exception of mages/archmages and titans) deal half as much damage in melee combat as they do when shooting (for archers and elves/grand elves, it is even 4 times less).
4) Almost all shooters occupy one cell on the battlefield (except centaurs).
5) Shooters are a priority target when the computer chooses a target for destruction; that is, they hit shooters first.

From this follow the conclusions:
1) Shooters should be protected in every way from direct contact with enemy units;
2) When storming castles, one must understand that you will likely have to part with some shooters (towers);
3) If there is an "anti-magic" spell in the magic book, consider whether to cast it on the shooters;
4) The "slow" spell (and especially "mass slow") can seriously help when playing with shooters;
5) Blocking enemy shooters is one of the typical techniques in the game, which can significantly affect the outcome of the battle and losses.

In addition, the following factors must be taken into account:
1) the number of shots for shooters is limited (from 8 to 24);
2) if the hero has the "ammunition wagon/infinite supply" artifact, the number of shots for shooters is unlimited.

Soon we will consider issues of tactical struggle for and against shooters.

Let's continue examining the topic
We remember that the distribution of shooters among castles is as follows:
-) Knight – 1 (second level, shooters/archers);
-) Barbarian – 2 (second level orcs/orc chiefs, fifth level trolls/war trolls);
-) Phoenix – 2 (third level elves/grand elves, fourth level druids/great druids);
-) Necromancer – 1 (fifth level liches/strong liches);
-) Titan – 3 (first level halflings, fifth level mages/archmages, sixth upgraded level titans);
-) Black Dragon – 1 (first level centaurs).

We also remember that all shooters except the centaur occupy one cell on the battlefield, as well as that all except mages/archmages and titans deal half damage in melee combat.

Now we will move to an extremely important part: how, actually, to play with shooters and how to play against shooters.

Taking into account all the above factors, we understand that the opponent's shooters must be forced into hand-to-hand combat. In this regard, let's stop in more detail on combat tactics in the game "Heroes of Might and Magic. Part II."

The field in the game is a rectangular grid consisting of regular hexagons ("hexagons"). When you load the game next time, set "show battle grid" in the settings, and everything will become clear. There are 9 of these hexagons in width and 11 in length. That is, when placing five units, one free cell remains between them. And between two opposite units occupying ONE cell, the distance is 9 cells; between units with a width of one and two cells – 8 cells; between two two-cell units – the distance is 7 cells.
What does this mean in practice? That a unit with medium speed (4) needs 3 turns to reach an opposite unit occupying one cell, and 2 turns to reach an opposite unit occupying 2 cells. For example, skeletons will take three turns to get to halflings, but 2 turns to get to centaurs! Or (a less typical example), despite the same medium speed (4), ogre lords will reach titans in 3 turns, and medusas – in 2 turns! And in the case where champions stand opposite centaurs (or dragons, or other two-cell units), they arrive in just one turn!!
Turns are made depending on the "speed" parameter. Within a group with the same speed indicator, the first turn is for the top unit of the side that hasn't moved yet (if no one has moved – for the attacker), then the top unit of the second side, then the second from the top unit of the first side and so on; that is, turns are made as if in a "snake-ladder" pattern. The presence of morale can play a pleasant or unpleasant trick: the higher the morale (maximum +3), the higher the chance that a unit will move again during its turn; the lower it is (minimum –3), the higher the probability that it won't move at all.
In the case of playing with shooters, you need to think about how to avoid melee combat. Usually, this is done by surrounding the shooters with your own units.
Example of a battle with halflings and golems against goblins (without taking morale into account).
1) initial setup: first unit – almost all hobbits except 1, second unit – 1 golem, third unit – 1 golem, fourth unit – all other golems (another option is to divide the golems into three equal parts), fifth unit – 1 halfling (he will die).
2) Started. First turn – all goblins moved. Halflings shoot, golems move up without exposing themselves to a hit.
3) All goblins moved. Halflings shoot. Golems surround the top unit of halflings. They attack the first unit of goblins only if they can wipe out that unit. At the same time, the halfling unit must be completely surrounded.
4) Goblins hit golems and kill the bottom halfling. Halflings shoot. Golems hit only if they can wipe out a unit. Meanwhile, the halflings must remain completely surrounded.
5) And so on until complete final victory. Losses – 1 halfling and a few golems (or maybe the golems survive).

It goes without saying that morale, which might fall to the goblins on the second turn, can very significantly affect the results of the battle, even to the opposite.
Battles with undead are very convenient to calculate (for example, skeletons, upgraded zombies, unupgraded mummies) – they have no morale.

And one more very well-known algorithm. If you have a shooting unit with high speed against non-flying slow troops, you can try to "run around." The essence is simple: a unit approaches – we don't shoot, but run away from it; it follows us – we still run; it follows us – we (calculating that it won't catch up this turn) shoot, then it approaches – we run away and so on. You only need to remember three factors – the number of shots (if there is an artifact providing ammunition – you can forget about this), where to run so as not to be caught, and suddenly appearing morale for the pedestrian.

In addition, it is important to master the "let them surround me" technique. If there are any obstacles on the battlefield, there is a possibility that among several units some will go around the obstacle from one side and some from another. And we will pick them off one by one. This technique, by the way, is very often used in "Heroes of Might & Magic. Part 3".

Now let's try to draw some conclusions:
1. Put a grid on the battlefield;
2. Before attacking, you should think about how to arrange your troops;3. When going on a campaign, remember that battle can break out unexpectedly, so your formation should always be as combat-ready as possible;
4. When fighting large armies, you can sacrifice one shooter by placing them fifth (they will end up at the bottom of the battlefield) – some enemy detachments will be distracted by them, giving your shooters more time to fire;
5. During battle, calculate where the opponent will move on their next turn (you can try to calculate several turns ahead);
6. Use the nuances of the battlefield – this can not only reduce losses but may change the outcome of the battle!


To begin, I will remind you that in HOMM2, the income brought by a castle is 1250 gold per day (CD castle – 1750). For comparison: in HOMM3, with similar unit prices, the income of one castle is 4000 gold per day. Furthermore. The cost of weekly creature growth in HOMM2 ranges from 11 685 (knights) to 25 500 (titans) (the cost of special resources – sulfur, gems, mercury, crystal – is conventionally taken as 250 gold). Thus, it becomes clear that the cost of weekly growth exceeds the weekly income of the castle. Certain problems follow from this; in particular, one must think about what to build, when and why, as well as whom to buy out and why.
Attempts to introduce certain coefficients that would reflect the true utility of troops have been made repeatedly. It is clear that with an equal number of health points, vampire lords are stronger than ogres, and grand elves are stronger than zombies. It is clear that despite the computer rating, an elf is stronger than a pikeman, while a skeleton is weaker than both a centaur and a thief. It is obvious that shooting increases the value of troops (the absence of direct collisions with the enemy reduces losses), while low speed decreases it (helplessness against arrow fire and magic). An unretaliated strike, magical abilities of creatures or immunity to magic, mass attack, double attack, special abilities like regeneration for trolls, restoration for vampire lords, or the probability of removing half the troops for djinns – all of this increases troop efficiency. They become more durable or "biting."
I tried to derive a certain formula that would take into account all the main parameters of a unit, which I include:
-) number of health points;
-) attack-defense parameters;
-) damage;
-) speed;
-) ability to shoot or fly;
-) immunity to magic (necromancer troops, dwarves, golems, giants/titans, phoenixes, colored dragons, etc.);
-) magical abilities (mummies, unicorns, medusas, cyclops, archmagi);
-) double attack, unretaliated strike, extended attack, mass attack, counterattack (griffon), absence of melee penalty for shooters (mages, titans);
-) special abilities (vampire lords, trolls, djinns, witches, bone dragons);
-) unit cost;
-) unit building cost.

As a result of independent calculations and analysis of a number of materials on this problem, I came to the following conclusion: a universal formula that would correctly account for all the above factors according to their "weight" is extremely difficult to derive; moreover, its appearance seems highly unlikely. For example, how do you account for the fact that the strength of vampire lords increases non-linearly with their number—that is, 30 vampire lords are not three times stronger than 10 vampire lords, but at least four?!
It should also be noted here that numerous studies have been conducted to determine the comparative strength of creatures. This is mainly done through "sparring" in one quantitative ratio or another. The results of these studies are quite curious, but nothing more. It is clear that a troll wins almost every "one-on-one" duel, but this by no means means that in practical gameplay he is better than, for example, a phoenix. Also understandable is the mighty power of 100 vampire lords, and the much smaller relative strength of one vampire lord. It is equally obvious that units like dwarves, iron golems, and ogres defeat most flying units of similar levels and at the same time lose without a fight to shooters of lower levels. However, all these inquiries are quite far from practical gameplay. Most results are obvious even before testing, and the fact that "slow-movers" can be used against flying troops, while conversely, flying, shooting, and fast detachments are needed against shooters—and it may not even be the slow-movers' turn—is something any practitioner would say, or even someone who has simply carefully read my previous materials.

So, let us formulate the problem of this study in detail: when playing on "impossible" level (chess king piece, no starting resources or gold), due to the deficit of gold and resources, building up the castle (specifically, creature dwellings) is difficult; moreover, there is not enough gold to buy out all troops. In connection with this, two questions arise: whether to build a certain building, and which troops to buy out.
First, some examples.
In an elf castle, full development can be carried out by bypassing the construction of dwarves. The question of the expediency of building dwarves (requiring a tavern for 500 gold and 5 wood, a cottage for 1000 gold and 5 wood, and in case of upgrading the dwarves another 1500 gold and 5 wood) should be decided depending on the availability of free gold, resources, as well as the potential utility of dwarves in the upcoming struggle. And personally, one of the main arguments for me is the presence of a "fort on the hill" building within sight, which allows upgrading dwarves without charge—that is, for free. Thus, we get normal troops without building an upgrade for them and at a lower price.
In a knight castle, paladins (sixth level) can be placed without building cavalrymen (5th level). As a rule, when the choice arises: build paladins or cavalrymen, the higher level is chosen. This is understandable and in most cases justified. But after some time, the question arises: whether to place cavalrymen. Let's say right away: if a player thinks about this question, it already speaks to their certain strength. Now about the essence of the choice: one can either upgrade other troops or (on separate maps) build a shipyard or place cavalrymen. Here too, many things must be considered: look around for the "foundry" special building (allows upgrading pikemen and swordsmen without charge—meaning money and resources are freed up), evaluate the need for cavalrymen in further struggle, look around for promising villages: maybe it's better to capture one and use those 20 wood to build a castle there... And of course, playing the POL ("The Price of Loyalty") version on so-called "new" or "extended" maps, you must check if there is a "stable" special building nearby, where cavalrymen will turn into swift champions completely free of charge. This is a serious argument in favor of building cavalrymen ...
So, it has become clear that the topic is relevant, but without universal recipes. Nevertheless, I would like to list a number of units whose purchase would be quite interesting.
Witches (Ghost). In 99% of cases, purchasing this unit is justified. Moreover, almost always after buying this unit, building other units is no longer necessary.
Djinns. At an early stage of the game, fast flying troops are a very serious argument. And at a late stage of the game, the probability of removing half a detachment is a huge threat!
Red Dragons. Yes, their construction is not easy. But at a price of 3500 gold and 1 (just one!) sulfur, you get the 2nd or 3rd strongest unit in the game, moreover, immune to magical attacks.
Bone Dragons. Development does not require a huge amount of a specific resource (as in other castles); everything is quite balanced. At a price of 1500 gold and nothing else (no resources), getting a flying unit with 150 health and damage 25-45, moreover immune to mind magic and reducing the opponent's morale by one—that is good! And if you also have "raise dead" magic ...
Champions. A significant nuance here is the amount of wood required for construction. Otherwise—with very high speed (7) and a cost of 375, we get a unit with 40 health! If there are stables nearby—buy cavalrymen and upgrade them in the stable!
Ogre Lords (ogres will also do against flyers). Lots of health and inexpensive. In the presence of a fort on the hill, buy ogres and upgrade them without charge (that is, for free)—and that's almost victory!
In the presence of a foundry, purchasing golems and swordsmen also deserves attention (pikemen look slightly less interesting—here the main and almost only plus is the defense rating of 9).
In the presence of a fort on the hill, buying dwarves and orcs seems quite interesting.
Besides the above units, practically all level 6 units seem quite interesting, as well as rocs (they fly!), wolves (good speed), vampire lords (provided there are at least 23!), griffons (unlimited number of counterattacks), gargoyles (good speed combined with flight and decent defense), specifically against flying units—hydras (mass unretaliated attack allows reducing losses), when playing for the Necropolis—king mummies (speed, ability).
Naturally, these assumptions are far from indisputable; they are not based on fundamental calculations, although they take into account parameters such as unit price per HP (one health point), unit price per computer rating point, as well as intuition and experience—the most important parameters for the magnificent strategy "Heroes of Might and Magic 2".


A continuation may follow ...
Best wishes, Qwerty

Practical Numerology
or
Three Questions of the Second Heroes.

Brief introduction.
At the moment, I distinguish three types of gameplay in the second heroes.
1. Offline tournament play.
2. Online tournament play.
3. Amateur play for pleasure.
The material presented in this article refers exclusively to amateur play for pleasure.

First Question.
1 or 7
("When we acquire castles...")

A situation quite often arises where you can take a computer opponent's castle on the seventh day or the first day of the week. Of course, if it is the last castle and there are no more heroes - you must storm it as soon as possible. However, if it is one of the first castles and many battles still lie ahead, it makes sense to think a bit.
Let's try to analyze the main comparative pros and cons of taking a computer opponent's castle on the seventh day or on the following first day of the week.
If we take the castle on the seventh day, then:
1. An additional 1000-1750 will be added to the treasury depending on the castle and its level of development.
2. POSSIBLY there will be a market in the captured castle or the ability to build one. And this may provide an opportunity, by exchanging resources, to build a necessary structure (for example, the highest level of troops) in ANOTHER castle.
3. POSSIBLY everything is ready for building the highest level of troops in the captured castle and we have the resources for it. Then we build these troops and go out with them the next day (more will be added during the week transition!!!).
4. Opportunity to replenish mana.
5. Opportunity to build the next level of the magic guild and another level of the magic guild the following day. Thus, the hero MAY BE equipped with powerful high-level magic (if you are lucky with spells).

If we take the castle on the first day of the week, then:
1. When playing on "Impossible" difficulty, we can get an additional creature growth.
2. POSSIBLY the computer opponent built a structure important to us specifically on the seventh day (for example, the last level of troops or another floor of the magic guild) and we will be able to benefit from this without spending resources.

Thus, based on the troops of enemy heroes and the castle guard, we can make a rough forecast of buildings and decide on the time to take the castle.

Second Question.
4 or 5
("What if they join?")

It is obvious that increasing the number of stacks (squads) of troops in an army increases tactical combat possibilities and very often minimizes losses (and in some cases, a battle can only be won by dividing into the maximum number of stacks).
At the same time, the joining of neutral troops is a serious reinforcement. In this regard, I have repeatedly observed the following moment among strong amateur players. Neutral troops are standing there. A strong amateur looks at them. Then he arranges himself into 4 squads instead of 5 and attacks saying "What if they join...". As a result of the battle, losses exceed those that could have been with five stacks by one and a half times.
How to try and reduce the number of such situations (remember that ghosts and elementals never join)?
1. Hero without Diplomacy.
Roughly estimate (or precisely, using a guide for the second heroes) the strength of your army from the computer's perspective. If it is twice as strong as the neutrals, think further. If it is less than twice as strong as the neutrals - arrange into five stacks: there will be a battle.
So, we have an army more than twice as strong as the neutrals (naturally, you must also try to estimate the neutrals). Try to remember if anyone has attacked them previously. If they did - arrange into five stacks (there will be a battle). If not - estimate possible losses in case of a battle with 4 of our stacks versus 5 of our stacks (it is preferable to "allow" for the possibility of an upgraded squad among the neutrals). If the difference in losses between different numbers of stacks is critical - arrange into five squads. If not critical - try to recruit them.
2. Hero with Diplomacy. First, the same estimation of your and neutral armies as in the previous example. If our forces are more than twice as large - think further. You need to estimate the approximate cost of the neutrals and compare it with your financial capabilities. If the neutrals cost more than the available cash - switch to the hero without diplomacy option. If less - evaluate the expediency of acquiring them, keeping in mind that for the cost of a neutral army we will get a full squad with Expert Diplomacy, half a squad with Advanced, and only a quarter squad with Basic. If the neutrals have previously participated in battles - they join only through Diplomacy. And then it is as in the previous example with one exception. A diplomat hero with an army more than twice as strong as the neutral one and with a sufficient number of troops ALWAYS recruits neutrals (except ghosts and elementals), but whether they join for money or just so depends on the specific save (more accurately, both on the save and the map) and whether the neutrals have had previous battles.

Third Question.
2, 3 or 6
("Who am I going on a campaign with today...").
One of the most difficult questions in both amateur and online play is which troops to buy under conditions of fund deficits.
Two algorithms typical for amateur play are:
-) buy archers with everything, and the rest as change;
-) buy the last level, and the rest as change.
These algorithms have a very deep justification, but they are not the only ones.
Why are archers so often chosen? Because they effectively attack without retaliation (until enemy troops enter direct contact with them). And this allows for minimizing losses.
Why are last-level units chosen? Because they are strong, they have many hit points, and this allows for minimizing losses.
However, today we will introduce a concept that has long existed in heroes: the "strike stack". A "strike stack" is a squad in an army that allows dealing great damage to the opponent and deciding the outcome of the battle. Examples of "strike stacks". The best example of a "strike stack" are skeletons for a Necromancer on an XL (very large) map. Other options: Ogre Lords, Vampire Lords, colored dragons, Titans, Dwarves, Grand Elves; in different games almost any unit can form a "strike stack" (for example, with several external dwellings - Halflings).
Now we can look at buying troops slightly differently. You need to think about forming a "strike stack". And it may well be that we buy not 4 Cyclops and 6 Ogre Lords, but 14 Ogre Lords. When forming a strike stack, you must absolutely consider the upgrade possibilities on the terrain. If a Fort on a Hill is visible nearby, you can buy many Ogres (only 300 gold), then upgrade them, and you will get a strike stack of colossal power. The same, but to a lesser extent, applies to the Foundry and Pikemen/Swordsmen/Golems. By the way, Golems as a strike stack against non-shooting squads (provided there is a Foundry allowing for free upgrades) perform very well.
The second parameter that makes sense to consider when deciding on army composition is the hero's mobility, which depends on the speed of the slowest creature in the army. It is clear that if there are many Halflings, Steel Golems will no longer slow you down. However, for an army of Grand Elves, Druids, Unicorns, and Fairies, it would be better not to take non-upgraded Dwarves (exception - if a Fort on a Hill is visible nearby or if the Dwarves themselves constitute the "strike stack").
The third aspect that is important to consider is the survivability of the "strike stack". It can be increased using:
-) a non-striking stack of archers, taking most of the hits;
-) the presence of a fast stack, allowing magic to be applied first;
-) the presence of flying troops that "plug" the opponent's archers;
-) the ability to "remove retaliation" by hitting an opponent's squad first with any unit and then with the strike stack (so it hits without retaliation).
And finally, one more way to preserve a strike stack.
Buy no more than 4 squads of troops. This will allow you, if necessary, to reform the army, separating single units to remove retaliation, splitting archers, etc., as a result of which losses may be lower.

Until next time!
Best wishes,
Qwerty

This material was written specifically for the "Herald of Heroes" magazine, No. 8 (April 2012) - http://www.woggrad.ru/ru/vestnik-heroes.html

Mighty Rangers
or
The Strongest Strike.

Today we will talk about the damage dealt by units in Heroes II.
Let's start, of course, with a small theoretical overview.

Everyday Power Meter
or
Strike Formulas.
So, in Heroes II, the damage dealt by a stack depends on the base damage of the attacking creatures (UnitDamage), the attack parameter of the attacker and the defense parameter of the defender—more precisely, on the DIFFERENCE between these parameters.
And there are two such formulas. You can't understand this; you just have to remember it.
If the attack of the attacking (A) stack is greater than the defense of the defending (D) stack, then damage (Damage):
Damage=UnitDamage*Number of units*(1+0.1*(A-D)).
If the attack of the attacking stack is less than the defense of the defending stack, the formula looks different:
Damage=UnitDamage*Number of units*(1+0.05(A-D)).
For both formulas, there are limit values for damage. If the attacker's attack is higher, the limit value of the modifier (1+0.1*(A-D)) will be 3. If A is less than D, the limit value of the modifier (1+0.5*(A-D)) will be 0.2 (it is often mistakenly believed that this parameter is 0.3).
UnitDamage is the damage dealt by a unit. It is specified in the properties of each unit; you can also use the Heroes2info reference by Qwerty. For all units except peasants, damage is distributed within a certain interval. For most subsequent calculations, we take the average damage (the middle of the damage interval). Thus, for halflings (damage 1-3), the average damage will be 2 (in melee 1).

Turning the Page
or
Magic in Alliance with Strength.

There is a whole range of spells that can affect the damage dealt by a stack. Since the formula includes the damage dealt by 1 unit and the difference between attack and defense, the main magical "gadgets" in this case will act specifically on these indicators.
The damage of 1 unit will be affected by (mass) Bless and (mass) Curse spells. They fix the damage at the upper or lower bound (except for undead and colored dragons; Curse also does not work on crusaders)—in this case, the damage will be either at the upper bound (Bless) or the lower bound (Curse) of the "damage interval." We will analyze specific examples a bit later; here I will only say that 10 halflings under Bless, shooting, will deal 30 damage to a creature with defense 2, while those same 10 halflings under Curse when shooting will deal only 10 damage to the same creature (and in melee even 5).
The rest of the spells will act on the difference between attack and defense.
Bloodlust will increase the stack's attack by 3 for 3 rounds.
Stone Skin will increase the stack's defense by 3 for a number of rounds equal to the hero's spell power (excluding artifacts that extend spell duration).
Steel Skin will increase the stack's defense by 5 for a number of rounds equal to the hero's spell power (excluding artifacts that extend spell duration).
Disrupting Ray reduces the stack's defense by 3 until the end of the battle. Furthermore, this spell cannot be removed and can be applied until the stack's defense equals 1.
Dragon Slayer increases the stack's attack only against dragons by 6 for a number of rounds equal to the hero's spell power (excluding artifacts that extend spell duration).

Strictly speaking, there is also "stone" magic cast by medusas; when hitting a petrified stack, it receives 50% damage.
We will talk about Shield spells below.

Indian Fedot Shoots Accurately
or
Peculiarities of Damage Calculation for Ranged Stacks.

First, let's remember who can shoot. Halflings, centaurs, shooters, archers (2 times), orcs, orc chieftains, elves (2 times), grand elves (2 times), druids, great druids, trolls, war trolls, mages, archmages, liches, mighty liches, titans.

All except mages, archmages, and titans deal 2 times less damage in melee. That is, the base damage of one unit will be 2 times lower.
Additionally, creatures that shoot 2 times—archers, elves, and grand elves—deal only one hit in melee. Thus, their base damage in melee will be 4 times lower (and they'll get a counterattack on top of it).
When an attacker shoots through own castle walls that are not destroyed, the damage will be 2 times less (in the absence of the Golden Bow artifact).

The same secondary skill Shooting removes the penalty when shooting through castle walls and increases damage: basic by 10%, advanced by 25%, expert by 50%. Thus, archers with expert shooting deal damage 6 times greater than in melee.

The (mass) Shield spell reduces the damage received by a stack (in the mass version—by all non-immune stacks) from ranged attacks by 2 times for a number of rounds equal to the hero's spell power (excluding artifacts that extend spell duration).

That's it? No, not everything! We must figure out one more nuance. For shooters to shoot, there must be space free of enemies next to the shooter and an availability of arrows. Let's immediately note the "ammunition supply" artifact, which makes the number of arrows infinite. However, a hero may not always have such an artifact, so let's stop in more detail on the issue of the number of arrows.
There are 8, 12, 16, and 24 arrows. First, let's list those units that have only 8 shots, as they are often insufficient for a battle: centaurs, orcs, druids, trolls. Halflings, shooters, mages, and liches have 12 arrows in their quivers. Orc chieftains, great druids, and war trolls have 16 shots in reserve. This will be enough for most battles. And archers, elves, grand elves, archmages, mighty liches, and of course titans have 24 shots. Let's also note that archers, elves, and grand elves can use up their 24 arrows in as little as 12 rounds, because they shoot twice per round (if the target remains alive after the first shot).
A logical question arises: what happens to shooters after exhausting their supply of shots? This is where double standards come into play: for computer troops, these shooters stop being a priority target, becoming one of the moving stacks (by the way, in Heroes II there are no flying ranged stacks), but!, and this is important for the topic of this material, shooters regardless of whether they have arrows or not will have half damage in melee (of course, except for mages, archmages, and titans).

Who is the Strongest in the World
or
The Best of the Worst and the Worst of the Best.

Statistical data on creature damage can be seen in the Heroes2info reference by Qwerty; we will focus on separate nuances.
First of all, we already know that abstract damage does not exist. There is damage dealt against a specific stack with a specific defense.
In calculations, I took this defense to be equal to 9. Why 9? There is an answer to this question, but it falls outside the scope of our research, and therefore instead we will provide examples of units with defense 9. Defense 9 is found in pikemen, veteran pikemen, swordsmen, swordmasters, cavalrymen, medusas, champions, unicorns, hydras, djinns, cyclopes, and bone dragons. Generally speaking and summarizing (although generalizations are not very appropriate here)—defense 9 most closely corresponds to fifth-level troops.
A very interesting question: what should be compared? The damage of one unit? But there is little sense in that.
I see three possible approaches: comparing the damage of a weekly growth, or comparing the damage of a stack with a certain number of hit points (HP), or comparing the damage of a stack achievable by a certain point in the game.

The damage of weekly growth is quite obvious and predictable: 3 black dragons with 14 attack and 25-50 damage are obviously the best (over 168 damage). Red, green, and bone dragons are also among the leaders.
The outsiders are naturally peasants with 13.2.
Damage from a stack with a certain amount of HP is more interesting. However, here a question arises: what does this parameter practically give us? Apparently nothing...
The third aspect is the damage of a stack of a certain cost. Here, of course, a question arises about the methodology for calculating the cost of a stack. Should building construction be included? If so, should castle construction be taken into account? And how to account for external dwellings: what to do with shooters (since rangers are truly mighty, their piercing power when shooting is equal to grand elves, and their growth is 4 higher, plus they come with heroes, plus shooter houses), halflings (whose damage under Bless is quite significant), and even dwarves?
The damage of a stack of a certain cost is a practical question, and its application is obvious. In calculations, we get an indirect dependence. Obviously, the cost of 1 black dragon including construction will be monstrously high, while the cost of 10 black dragons divided by 10 will be an order of magnitude less.
Therefore, we will try to give a very rough estimate of the damage of a stack that can be at our disposal by the third week. Why only the third week? First and foremost, so that this material is read to the end. For certainly, I would like to provide such an analysis for the first week, second, fourth, and upon completion of the first month of play. However, a stream of monotonous information does not contribute to increasing interest. Specifically, the third week is because on most maps by this point it is possible to build sixth-level troops (as for upgrades—that depends on the situation).
Analyzing indicators is extremely difficult, but we will still try to highlight leaders and outsiders.
Taking into account various factors, which we include: potential presence of an external "external dwelling", relative ease of taking a village, cheapness in construction, potential arrival of units with heroes, potential sharp increase due to the "Shooting" skill, susceptibility to the "Bless" spell (and also "Bloodlust")—first place is awarded to the mighty rangers!
When this stack can shoot—it simply has no equal. On the first day of the third week, we could quite possibly have around 90 archers (28 home castle, 24 captured unbuilt village, 12 come with heroes, about 30 from an external shooter house). 90 archers under a zero-level hero will deal 288-432 damage to creatures with defense 9. Naturally, everything said refers exclusively to ranged damage, because in melee archers deal only 80-120 damage; moreover, low defense (only 3) and small number of hit points (only 10) immediately start to take a toll.
But during further analysis, we encounter questions of double attack. If it is possible to strike the opponent's stack first so that it responds, and only then pile on with a stack with a double hit (this tactical maneuver is called "removing the counterattack")—then it seems one should account for double damage. However, when a stack enters direct contact with an enemy, you must be prepared for the fact that we will also receive hits, and the counterattack of any stack with a double attack is only single. That is, if paladins hit—they deal two hits (if neither participating stack dies), but if paladins are hit—they respond once.
There is no counting methodology. There is an idea to use a coefficient of 1.5 with approximate justification: in 1 round with two fighting stacks (without unopposed damage) a stack of paladins, crusaders, or wolves will deal 3 hits and receive 2 (3/2=1.5). Such a methodology is not correct, but would another be better? There is another nuance. Units possessing a double hit receive a counter-hit after the FIRST hit, meaning only those who survived it deal the second hit. Let's look at an example. 4 wolves deal 27.2 damage over 2 hits against a creature with defense 9. So, in a couple of rounds we'll take down a djinn with some to spare? But here is what happens in reality. Wolves attack, deal say 14 damage; in response the djinn deals 45 damage, killing two wolves; the remaining 2 deal another say 7 damage. After this the djinn moves and simply kills the remaining wolves.
Let's try to accept a coefficient of 1.4 (although even it is clearly inflated).
So, sixth level troops.
Let's consider a hard-to-reach stack of 12 paladins (built them in the home castle on the first week, grabbed a bunch of wood and finally managed to build them in a captured village, but already on the third week). Then damage for one hit will be 168-336; with a coefficient of 1.4 it's already 235.2 – 470.4. We won't talk about how to find 40 crystals and 90 wood in 2 weeks (without an external market this is practically impossible on the vast majority of maps), we'll only say that this option is much harder to achieve than red dragons. At the same time, paladins have only 50 hit points, so they will perish at a considerable rate.
Among others, it makes sense to highlight possible bone dragons on rich maps. Now we will not consider the option of building dragons in the first castle on the first week plus in a village on the second—this happens too rarely (although it is possible). Let's consider the option simply with dragon construction on the first week. Then on the third week we have 7 of them. They will deal 210-378 damage, which is an excellent indicator.
4 red dragons, however, will deal 140-280 damage to such creatures; moreover, "Bless" does not work on them, just as it doesn't on their bone brethren.
In the fifth level, in case of a large amount of wood (from the sea, for example), construction of cavalrymen in two castles is possible (and in the case of an external stable they are upgraded to champions for free!). There can be up to 21 of them (13 from starting castle + 8 from built village). The price of the matter—90 wood. They will deal 115.5-231 damage, which is quite serious.
At the fourth level of troops there are very interesting units; however, the question of the expediency of their construction in two castles is not simple. In terms of damage, minotaurs certainly stand out. There can be 21 of them (13 from starting castle, 8 from built village). They will deal 105-210 damage.
Rooks. Here, strictly speaking, it's not about them themselves, but about the construction and purchase of titans. Let's stop on this point in more detail, as it is extremely important. Suppose we are playing as Barbarians. On day 6 we gathered 20 stone and 20 wood; a village was captured. In the vast majority of cases, we will place trolls, not a castle. The same situation on the second week. With crystals collected and at the end of the week (day 6-7), one's hand still reaches to place cyclopes. The same applies even more so to the black dragon castle; a roughly similar situation for Sorceresses. Let's look at Knights. When wood is needed for everything, levels 5-6 troops are built quite easily (if that wood exists); wood is also needed for the key archer upgrade—it's extremely hard to make a decision about building a village into a castle (20 wood is insanely regrettable).
But Necromancers and titans depend much less on stone and wood. To get bone dragons or giants, 25 wood is needed, and even less ore. But if Necromancers have a worthy alternative in the form of vampires, requiring an additional 15 wood (another 5 for lords), then for the titanic castle gold for upgrade and purchase of titans is much more important. One of the gold generators is the castle (especially since many stones are not needed for titans, meaning you can spend 5 on a statue, further increasing income). Thus, in the first week we often simply have no other business than building a village into a castle. Considering that if the village already has second-level troops (and we are considering exactly such an option in this article)—rooks are placed the very next day after castle construction, it turns out that by placing a castle on day 5, by the end of the second week we will earn at least an additional 2000 gold, and further this income will increase; moreover, rooks won't be superfluous (you need something to fight with before titans appear). Based on the above, we can have up to 26 rooks on the third week. However, I suggest considering their number as 24 (there is a high probability of either losing two rooks or simply not having enough days to build a well in the first week). 24 rooks will deal 86.4-172.8 damage. Not so little, especially if you consider the following nuance. To build mages and giants, construction of a magic guild is necessary. Halflings, who are the main weapon in the first and partly second weeks, are extremely sensitive to the Bless spell. Thus, we will obtain and use this spell whenever possible, which means 172.8 damage will turn from a statistical corridor (for undead troops and dragons) into actually dealt damage.
At the third level, a stack of wolves is easily achievable. Suppose by day 131 we have a stack of 31 wolves (19 from starting castle, 12 from village built into a castle on the second week). In terms of gold this is very easy to achieve; the problem is exclusively in finding 20 scarce stones, so needed for building trolls and cyclopes. 31 wolves will deal 79.05-131.75 damage with one bite. With a coefficient of 1.4, this turns into 110.67-184.45.
Under certain conditions, golems can be interesting. The conditions are simple—an external building "foundry", which free of charge, that is for nothing, will turn iron golems into steel ones and at the same time increase the damage they deal by 20%. We could have 26 (or maybe even 32!!!) golems. Here's the thing: titans are extremely expensive, both in construction and purchase, but not demanding of wood and stone. Therefore, under certain conditions we can not only capture a village on the first week (and doing so will be quite easy, as Qwerty wrote in the article "Is there life without titans?"), but also, perhaps, place a castle there and even build golems (although in most cases it makes sense to build rooks). 26 steel golems deal 93.6 – 117 damage; 32 steel golems already 115.2 – 144. It seems that if you look on average, these are quite unremarkable figures. But! The damage of golems when the "Curse" spell is cast on them by an enemy decreases much lower than for others, and defense 10 with reduced damage from many spells makes battles possible with serious troops under the command of mage-heroes...
Speaking of second-level troops, we will assume as a hypothesis that in the village captured by us on the first week they are already built. However, we do not build this village into a castle, but upgrade second-level troops either in the home castle or in an external building "hill fort" (orcs, dwarves).
Besides the leaders of our chart—the mighty rangers—it is simply necessary to consider the super-stack of dwarves very often encountered in practical play. With an external building where they join for free, taking into account those who came with heroes and provided a captured village, their number by the third week can reach 90. Note that if a hill fort is found, they will also be upgraded.
90 dwarves will deal 153-306 damage, which is an excellent indicator.
Among first-level troops, skeletons clearly stand out. Talking about skeletons in isolation from necromancy, I think, is not entirely correct, as it's clear that any growth of first levels is incomparable in strength to higher ones. Let's only note that for example, 150 skeletons (more than realistic by the first day of the third week, even without hyper-active use of necromancy, with a main hero having no necromancy at all) will deal as much as 225-337.5 damage to creatures with defense 9, which is simply an excellent indicator.
Fairies, possessing an impressive hit, only have two hit points, which combined with defense 2 removes them from the list of leaders. Meanwhile, let's note that those same 150 fairies will deal UNOPPOSED 112.5-250, meaning under Bless 250!
Halflings are good fighters in a ranged attack (especially under Bless). 150 halflings deal 97.5-292.5, and obtaining Bless is quite possible (for example, when buying a Sorceress hero). Let's also note that blocked halflings deal 48.75-146.25; meanwhile defense 1 and only 3 hit points will quickly reduce their number.

Of course, speaking of damage, one must also speak of receiving damage, at least for non-ranged troops. But this will be the topic of a separate material dedicated to the legendary strategy—"Heroes of Might and Magic II".