Yeah, although the term "bounded accuracy" hadn't been coined at that point it certainly seems to apply to OD&D. That's one thing I like about it.
Marv / Finarvyn DCC playtester (2011) S&W WhiteBox author (2009) C&C playtester (2003) Builder of the TrollBridge for T&T; Amber Diceless player since 1993 OD&D Player since 1975; Metamorphosis Alpha since 1976
"Don't ask me what you need to hit. Just roll the die and I will let you know!" - Dave Arneson
I don't remember how BA relates to 5e/ODD design--anyone care to summarize?
In AD&D and 3e, to hit bonuses increased by up to +1 for every level (for fighters, less for everyone else). What 5e did was make it so every bonus increases at +1 for every four levels for all rolls and all classes.
In 3e, because the bonuses were all based on a mathematical formula that continued forever, it would be possible at high levels for some die rolls to be such that anyone that got +1/level would auto succeed whereas everyone with +1/2 levels would auto fail. The Epic level handbook fixed this by making all bonuses raise at +1 / 2 levels after 20th. This way bonuses would increase while keeping the gap in ability (about +10 at 20th level between character that were good vs bad at a particular thing). [This is sort of how hit dice in OD&D work; you gain fewer per level after name level]
The term "bounded accuracy" isn't really correct as their aren't any real bounds. You'll still have the same problem it'll just take you 4x as long to actually see it.
Personally, I see this as a symptom of the fact that the 5e designers just couldn't think of a good way to make the game fun when you had wildly different skill or attack bonuses. So they just sort of hid the problem. Kicking it down the road so to speak. It's a shame that the 5e designers didn't look at games like Hero Quest to see how large skill gaps can still work. (something like that once a fighter hits 20th level, his attack bonus drops back to +1 but he gains one automatic hit per round.)
Post by howandwhy99 on Dec 3, 2015 16:49:48 GMT -6
5e edition is based off a 30 point number scale so it still slides at high and low levels. 1st level PCs vs anything over 19th will find the odds flatten at 5%.
OD&D bases its own on the 20 point number line with attack saving throws matrices only spanning within that range. So taking on the most powerful creatures, magic, traps, etc. in the game will always be accurate in both odds and dimension.
Post by howandwhy99 on Dec 5, 2015 11:30:55 GMT -6
It's been a year or two since I've looked at it again. I thought the game had 30 levels.
My understanding is from 3e which had a floating d20 roll on a natural number line, which topped and bottomed out at 5% probability (usually). It had a problem though, the advancement each time was too steep for the number of class levels in the game: 20 levels at a 1:1 advancement.
The problem was, rolls against creatures too high or too low became the same odds: 5% (whether to succeed or fail depending).
5th edition tried to fix this by flattening out the progression rates to closer to early D&D. But 5e still had 20 levels, not 10. The difference of a few levels between characters isn't as steep anymore, but the 5% problem still arises at the edges.
This means you can have a more unbalanced group group in 5e without resulting in an overshadowed character. 5e can functionally have a wider spread of character levels in a PC party as well as a wider spread of viable challenges in the world for them. However, it is still not OD&D where all the odds follow the same probability progression throughout. There is no collapse into 5% in OD&D. It maps its probabilities to the die roll.