The BECMI rules are really good where they are really good, and when they are bad tend to fall into the categories of missed opportunities, kludgey, or the same things that all D&D/AD&D at the time did. More discussion about the good at the end.
The first issue is the kludge -- while it is clear (from what we know of the company history, and just that they already had a basic-expert version of the game with Moldvay-Cook) that TSR intended to produce some or all of C, M, and I from the beginning, but from the actual products sometimes it is hard to tell that. Demihumans cap out at the same levels as in B/X, but the game keeps going up in level (upon which an unnecessarily awkward form of non-levelled-advancing is grafted on to them). Weapons like tridents are introduced in the Companion set. Weapon Mastery and some other combat options are included in the Master Set. Domain rules are in the Companion Set despite rulership starting at name level. Most notably the first printing of the B and E books used the same thief ability progression as Moldvay-Cook BX, but then subsequent printings changed them to fit a trajectory where thief abilities could approach 100% at 36th level*
*My personal experience was first playing with BX, getting first printing BE as the first books I owned, and then collecting C, M, and I as they came out. We were all mightily confused at what exactly was going on.
Second is things all of D&D had in the TSR era. These pop up in discussion about BX and AD&D as well. These are the things that people sometimes dislike the implementation of the replacement in WotC-era D&D, but probably understand why the change was made. Such as: I don't think level limits to balance demihuman abilities* ever worked well, but if you are going to do so, placing the hardest cap on halflings and the least stringent on elves never made sense. Everyone loved thieves and roguish characters as a concept, but no one ever wanted to play them as-is** because they were barely less fragile than magic users in exchange for the grand luxury of... some dungeoneering skills, many of which subject you to peril if you fail, and which other classes may or may not be able to replicate with spells or actions-anyone-can-take. Lots of 'be perfectly prepared for anything at all times' deathtrap monsters. And so on.
*or to make it a humanocentric game, as the purported real reason was
**Everyone I knew house-ruled something. I think we gave them a fighter chasis, and just traded being able to wear chain and plate for thief abilities.
And then there are missed opportunity. Firstly, Ralif is right that
"the progression kinda goes like this: dungeons (B) > hexcrawls (E) > tourneys/armies/kingdoms (C) > rulership & potentially questing for godhood (M) > godhood (I)". Second, Yora is spot on that the game's strength is
"A game with clear procedures, a goal, and an objective measure of success and progress. Search for a dungeon, explore the dungeon, collect as much treasure as you can without getting yourself killed, return with the treasure to a town." The missed opportunity, I feel, is in not fleshing out and making enjoyable the systematic components of the rest of what Ralif mentions not encompassed by what Yora mentions. The game does a spot-on job of setting up mechanisms for a procedurally rigorous and challenging (and enjoyable, if you enjoy the specific challenges upon which it focuses) dungeon-crawling experience. And then the expert set sets up the procedural wilderness mechanics that... do a really good job of determining how long it takes and how many encounters you go through to get to the next dungeon you can go back to procedurally dungeon-crawling. It doesn't really do a great job of making the wilderness hexcrawling interesting or fun, but more an obstacle or resource drain you seek to minimize so you can hit the next dungeon at full HP/spells/arrows. I bet you (whomever's reading) had a great time wandering the wilderness in BECMI back when you played the game BitD, and I also bet a good and imaginative DM was involved. I think that's why so much OSR ink is spilled on hexcrawl creation -- recapturing and proceduralizing what everyone made houserules/solid DM fiat to do. The Companion set rules for tourneys/armies/kingdoms is also a set of procedural rules for which I think a lot of house rules and DM inventiveness supplemented. They're not bad, they just, seem like they're there because someone thought they had to be (in case someone asked 'So what do I do with this keep and followers? How much revenue do I bring in? What happens if I declare war on my neighbor?'), more than built to be actually fun to play for extended periods. The Master Set quest for godhood, on the other hand, is really exciting and engaging and I remember plenty of people getting really excited about that and getting back into their high-level characters when it came out. In this case, things have flipped -- it seems fun as it's own thing and not just support structure for more dungeon-crawling (or planar adventure, as high-level often switches over to) and the like, but in this case the procedural structures are rather thin and it becomes very freeform/DM adjudication (again, to which I think lots of people had fun doing, because they had DMs who could do it well). The Immortal set -- I gotta be honest, I remember loving questing for immortal status, and loving the bizarro reality of the BECMI planes and laws governing the universe (which seemed so much more imaginative than AD&D's wheel of alignment-planes), but no one I knew really spent much time playing
as immortals. I think there was some confusion about what exactly you were supposed to
do or even
want at that power scale.
After all those caveats, I do think that BECMI is a good game. If for no other reason than because it does well the part that I think most people actually want out of their D&D-like games rules structures. Sure, I would love a hugely rigorous and fun for its' own sake wilderness adventure rules and I'll bet someone else wants a great domain management game, but most people (I think) want a game where you go through dangerous environments and use your awareness, cleverness, and appropriate use of limited resources to evade traps, defeat nasties, and acquire treasure. Compared to WotC-era D&D, it has the (arguable, I know) advantage of keeping you hungry and scared. Your expendable resources are pretty scarce, you're fairly vulnerable to a fair fight or un-found trap, and a lot of your power coming from last adventure's loot. That makes mere survival or some potions or a +1 spear you don't need but your henchman can sure use still mean something six or eight dungeons into your career. The biggest advantage, I feel, is gp=xp -- although I think some OSR types overplay the value, and you definitely want to modify something if you aren't going to play dungeon-crawler--in that it gives a solid explanation for why you are down in these dungeons where the game rules want you to be and also solidly defines success (having acquired treasure, and lived to take it back to town).
Definitely starting with the B and E parts. As for how it's run, the BECMI progression kinda goes like this: dungeons (B) > hexcrawls (E) > tourneys/armies/kingdoms (C) > rulership & potentially questing for godhood (M) > godhood (I).
But what the system does really well, which I think I've not really seen anywhere else, is that it's laid out and structured as a game. A game with clear procedures, a goal, and an objective measure of success and progress. Search for a dungeon, explore the dungeon, collect as much treasure as you can without getting yourself killed, return with the treasure to a town. And then count the points you made. That's a simple procedure that players can follow when they know nothing about the world and have no ideas what kinds of plans would be possible or interesting for their characters to pursue. And most importantly, it's something that the players can do on their own, without the GM telling them where they should go and what they are supposed to do when they get there.
Personally, I think these are best used as a grab bag of ideas to be picked through and not implemented whole hog.
I think that's almost necessary. There's a lot of stuff that either isn't necessary, isn't everyone's cup of tea, doesn't work in a given context, or just don't come together to something meaningful
unless it's a central game focus.
With that in mind, I am going to advocate for Moldvay/Cook (B/X). It is, in my opinion, the simplest and best D&D ruleset out there. If nothing else, what happened to the Thief in Mentzer was a crime.
I don't disagree. That said, the B/X thief was also... an interesting design choice to which I do not agree. Honestly, pretty much all TSR thieves (mechanically-speaking. Obviously the character concept of the thief is pretty engaging) seem like also-rans. From their introduction in oD&D Supplement I, the thief was set up as giving up just about everything (decent HP, AC, saves, attack value) in exchange for their abilities (and a more favorable xp chart which usually only has them a level or two ahead of others). Exactly why was never clear to me, and I sure think that the game would have worked very well with a thief class which was kind of a mirror of the cleric -- fighter-like attack instead of fighter-like defense, d6 hd, reasonable saves, and some late-game abilities (like cleric's higher level spells) the accounted for the games' shifting focus. I know in all versions of TSR-era A/D&D (including both BX and BECMI) we houseruled the heck out of thieves unless you played multiple characters (in which case if the thief was a swiss army knife who did nothing until a trap check was needed, it didn't matter as much).
*in retrospect I think Gygax may have preferred the 'anyone can climb and hide, and you find traps by saying you are searching the right spot the right way' method of pre-Greyhawk, and just included thieves for anyone who didn't.
