1) The game can be broken by bouncing Alchemy, Enchanting and Smithing off each other recursively. But it can’t be broken as badly as previous TES games and it remains brutal on harder difficulties even then.
2) The perk tree provides a lot more character customization than any previous TES game. Far nicer to change up the way a character interacts with the world than just to make numbers go up. Plus it gives the player much more control over what their character is effective at and greatly diminishes the need for grinding.
3) This game is far more forgiving of muddling about in different skills and going a jack of all trades route than any previous entry, at least in the normal difficulty range. In previous TES games it was necessary to gain exactly 5 ranks in skills on two specific attributes to get +5 to each of them on level up. Failing to do so would generally cause attributes to level half as fast or less. This, combined with enemies auto-leveling rapidly (and improving dramatically at combat even if the character had only been pursuing social skills) could easily lead to combat becoming unwinnable even on lower difficulties. Autoleveling has been largely fixed too, as now getting to level 50 will cause tougher and tougher variants of bandits to be much more common, but you’ll never find bandits in diamond platemail waving around lightsabers.
4) The graphics are positively gorgeous even on lower settings. This in comparison to previous entries which required a dozen or so mods to become even half pleasing to the eye. Now only elves have horse faces instead of everyone having them. Unfortunately the environment is generic. It isn’t half so aesthetically pleasing as Morrowind was with its carapace cities, forests of mushrooms, ash storms, 30 foot tall stilt striders for transportation, and floating islands. Those did a far better job of evoking a sense of wonder than simple snow and trees can manage.
5) Want to survive on Master mode? Here are some things to bear in mind. Armor and resistances are additive and max around 560 armor and 80% resistance. Your first 200 points in armor will not help you in a significant way nor will it help at all against elemental damage which is common, meaning you’d best focus elsewhere early on unless you want to rush to max them out. Going mage early on is not effective as enemies can take more of a beating than you can dish out without running out of mana. Being a mage has less to do with skills than with getting 200%+ to regen with found gear or 75%+ reduction in mana consumption from crafted gear. In the endgame though lightning is very effective at both passing resistances and stunlocking multiple enemies with “impact”. To be effective early on I’d recommend getting sneak up high enough to unlock backstab, enchanting a set of smithing gear, smithing yourself a nice sword, and then enchanting your gear with +% one-armed damage. This is particularly effective as enchantment and perk damage bonuses are multiplicative while smithing can provide flat bonuses which easily double the damage of even a mediocre sword. End result being that quadrupling damage output is far easier than achieving even a 50% increase in defensive survivability. For this reason companions and conjured pets are vital. Past this keep any fire or frost specific resistant gear you find as it’s easiest to stack this to effective levels early on and without it wizards and dragons will toast you. Broad spectrum resistance is achievable too, though very difficult without going Breton, getting the alteration magic resistance perks, and wearing magic resistance gear.
6) The balance issues listed above could probably be fixed relatively easily by cutting both enchantment damage multipliers and the effectiveness of smithing over 100 in half (100 is the skill cap but enchantments make it easy to reach 200 effectively), capping the maximum effectiveness of potions & enchanting (so one can’t make better potions to make better enchantments to make better potions, etc) then changing defenses from additive to multiplicative (to provide a more natural curve and allow both armor and resistances to matter at low levels). As is the 500th point of armor provides 3x as much of an effect as the 100th point. 20% multiplicative resistance per 100 points looks good per my math. Similarly I believe the inbalance in magic could be fixed by providing a flat 10% magic regeneration bonus with the primary perk line of each spell tree and changing the % reduction in spell cost from enchantments from additive to multiplicative, for a similar end result of making spells (especially low level ones) easier to rely on at low levels, increasing the importance of skills, and reducing the game breaking ability of gear. The game could also use a lot more variation in enemy magic resistances, to encourage using the full breadth of spells instead of just optimizing and spamming one of them. Lockpicking and pickpocketing are a bit harder to fix I think. As is they’re both a bit useless. Any lock can be picked without skill or perks given a handful of picks and some patience, and between combat and crafting there’s already an abundance of treasure. The best fixes I can come up with are to provide small dagger attack bonuses along the pickpocket perk line (also keeping daggers valid into the endgame and leaving thieves less useless against prepared enemies), to increase found treasure along the lockpick perk line, and to make the in game economy significantly more vicious so that resorting to thievery makes more sense. Hopefully a modder will come out with something along these lines before long. Though unfortunately doing so is likely to make the game even more addictive than it already is.
EDIT: On latter thought I’d like to see the super-smithing issue solved by just removing the potions for enchanting and smithing, as well as the enchantments for smithing and potions. They diminish the value of skills, take time, and happen to be boring. Also probably best to change the bonuses of smithing from cumulative to multiplicative. Generally that seems to be the biggest problem with Skyrim balance: not knowing when cumulative increases provide a cleaner power curve or when multiplicative increases do so. Any balancing should ultimately act to reduce the amount of time that needs to be spent on boring parts of the game and to increase the ability for fighter, mage, and thief builds to play effectively in a variety of dungeons at both low levels and high levels given sufficient skill.
