MMH5.5 - Heroes of Might and Magic 5.5

50% defense > 50% offense because 50% defense is half the damage while twice the damage is 100% offense.

The point here is the formula, not the concepts of Offense/Defense, which are equal (in some cases, Offense is better, in others – Defense).
Offense increases damage by (1+0.1) = 1.1 times, while Defense reduces damage not by 1.1 times, but by 1/(1-0.1) = 1.11 times.

When comparing Battle Frenzy and Vitality, we can use a single formula to determine how much damage is increased or decreased. Roughly speaking, Battle Frenzy gives +20% to damage, an increase of 1.2 times (if, as you say, all tiers are equal), while Vitality only provides +10% to health (a reduction in damage of 1.1 times).

In regards to the 15% bonus health - this is a skill I would pick anytime this is ~ 30HP boost for tier 7 units. "stand your ground" and "power of stone" cannot be equal unless the numbers are as high as the sky. Also from the Defense skill picture it seems the player can pick both "Vitality" and +15% hit points which is even more absurd.

Sure, but the point is that Power of Stone and Stand Your Ground are stronger than yours.

Power of Stone grants a cast with +10 or 12 to defense (conditionally, reducing damage by half – to 50%) over an area for a short time (longer if SP is high).
When a creature defends, it receives +50% defense (in WGE, this is similar to reducing damage to 60%). And Stand Your Ground doubles this bonus to +100% defense, reducing damage to 40% – 2.5 times less. Or 1.5 times relative to defending without Stand Your Ground.

In short, Power of Stone provides maximum protection (reducing damage by twice), but briefly, in an area, and at the cost of the hero's turn and mana.
Stand Your Ground gives 1.5 times the protection, provided the unit is defending.
Evasion gives 1.43 times the protection against ranged attacks.
And Vitality only gives 1.1 times the protection, but consistently.

In WGE, skills are very powerful; their estimated strength varies from +1 to +3 (on average +1.5-2) to a 10% parameter, and in 3.33% parameters, it ranges from +3 to +9 (on average +5).
The game is fast, there are few level ups, and there isn't time to waste on weak skills. I assumed that each level up would provide something significant, not just minor improvements.
From my point of view, the player should look at a skill and be amazed at how strong its abilities are, rather than scoffing at the fact that it offers some nonsense. Ideally, all branches in each skill tree should appear unbalanced (appearances can be deceiving). This motivates players to try them out. Good balance is not when everything is spread thinly and therefore there is no imbalance, but when there are so many strong things that they cease to be an imbalance. But this is the approach I advocate in WGE; I am not imposing it on you (although I have noted that weakening skills relative to the original game is bad).

I will once again note the fundamental difference between low and high levels. At low levels, you can do very cool things without being afraid that they will break the game (conditionally, not fearing that a warrior will crush a mage in 1-2 rounds). Because at low levels, the warrior selectively takes attack + luck, or attack + leadership, or attack + light, etc., while at level 40, he takes everything: attack, luck, leadership, light, and even defense as a bonus (which he doesn't need against a mage), and education if available (you specifically provide 8 slots for skills). Under these conditions, many powerful skills will simply sweep away the mage and leave him with no chance (if he hasn't invested in defense, which is rare for mages in your version). For this reason, I don't like levels above 20; they prevent strong skills from fitting into the balance.

Tournament practice (in another mod – RTA) shows that the branch of 15% magic resistance and 15% HP (in reality, there are +10% HP and +2 to 5% protection, but it is equivalent) is no more popular than Vitality + Power of Stone. Although Power of Stone is cast for 0.5 turns, i.e., stronger than yours.

What are the numbers for the other skills in Defense tree?

http://forum.heroesworld.ru/showpost...25&postcount=3
description

Added after 21 hours 4 minutes
As I saw in the manual, some creature abilities have been weakened in your game. Why?
The most annoying examples:
- agility: from +2 to defense to +1 to defense (weakened by a factor of 3!, considering the weakening of defense itself)
- cavalry charge: from +5% to +3.33%

If you needed to weaken the creatures themselves, it would be better to do this by reducing their corresponding parameters rather than nullifying abilities. Cavalry players would like an increase in charge to at least +10% per cell (too much damage? reduce base damage – although knights are not exactly the best unit on their tier).

The trick is that abilities create a certain interest; tactics can be built around them. What kind of tactic can be built around a 3% charge? Does it allow the knight to run one extra cell, or does it not? It's only a 3% difference in damage. The same goes for dancers.


Added after 45 minutes
There are also comments about errors in the original game that were not fixed in your version:
1) vampires: torpor is much stronger than retaliatory strike (it was almost untouched, and -1/-1 to attack and defense is insignificant), so the first grade of vampires is unnecessary
2) rakshasas: whirlwind is much stronger than retaliatory strike, so the first grade of rakshasa is unnecessary

I don't know how it is with penetration (although I suspect that removing counterattacks by single digits is not a problem), but I will never take the first grade of vampires and the first grade of rakshasas into the final battle. There is a drop both in abilities and characteristics.

Suggestions for vampires: torpor factor 1, damage 5-9 (instead of 5-13). For rakshasas: initiative +10 (instead of 8) and defense +5 (to 27). Although I doubt these ideas will be accepted; too many players overestimate the role of retaliatory strike (a minor ability). And arguing why multiple damage to several units (whirlwind) or disabling several units from the game (torpor with a huge chance) is better than situational retaliatory strike (which doesn't work if the counterattack has already been removed, which often happens when focusing on one unit) is tiresome.

P.S. I criticize a lot here, but please don't get offended – this is my specialty :) I always look for weaknesses that could be improved.