(EDIT: Since the site seems to cut off the titles of threads, just to clarify I'm asking about the EOB trilogy)

Title pretty much says it all. The manual mentions several 2E mechanics that should be in the game, but many of them have been explicitly debunked (racial weapon bonuses, EXP bonus for high primary attributes) or the evidence strongly suggests that they aren't actually implemented (dual-wielding bonus for high DEX, WIS bonuses to saving throws).

The Protection vs. Evil 10' spell seems like it works in EOB2 as a useful defensive buff, even with a Paladin in the party, in which case the spell shouldn't do anything because it shouldn't stack with the Pally's aura. So that leads me to assume the Paladin's passive buff is one of the things that they weren't able to implement in time. Does anyone have any hard evidence for or against, or even anecdotal evidence?

Solved! Will explain for the benefit of anyone else who might wander into this forum wondering the same thing. It occurred to me I could just let an enemy wail away at me with a Paladin in my party, and track the results, but I didn't remember nearly enough from my college stats course to know how comfortable to be with the results, what kind of sample size I would need, etc. Then I realized that there were probably tons of binomial hypothesis test calculators online.

So I set up a little experiment between a Displacer Beast and my Paladin-led party on Level 9. Front-liners both had -4 AC. The Displacer Beast's THAC0 is 15. Thus if the Paladin's aura was NOT implemented, I knew the Displacer Beast would hit on an attack roll of 19 or 20 (15 + 4 = 19 to hit), or 10% of the time. This was the null hypothesis (H0). The alternative hypothesis (H1) was that the Paladin's Protection aura WAS implemented, meaning the Displacer Beast would only hit on a natural 20, or 5% of the time.

Google said a sample size of 102 would be needed to test H0 and H1 with these parameters, with 112 trials recommended to account for continuity—which I won't pretend to understand—so I felt 120 trials was very conservative. Fortunately, the Displacer Beast attacks very quickly! I thus allowed it to attack my front row 120 times & recorded the number of hits: 14. Obviously in line with the null hypothesis's expected total of 12 (120 * 0.10), so the question was, was 14 an inordinately high number of successes if H1 were true? I found an online binomial test calculator and input the data:

n=120 (# of trials)
K=14 (# of successes)
P=0.1 (expected probability of success, given null hypothesis—no innate Protection from Evil aura)

The results were as follows. Given a 10% chance to hit (no Protection aura):

The probability of exactly 14 hits in 120 trials is 9.45%.
The probability of exactly, or fewer than, 14 hits in 120 trials is 78.18%.
The probability of exactly, or more than, 14 hits out of 120 trials is 31.27%.

A couple more hits, or a couple fewer hits, would still comfortably be within a standard deviation of the expected total, and the results are obviously in line with a 10% chance to be hit by a Displacer Beast with both front-liners at -4 AC. It's not looking good for the Paladin's innate Protection aura.

In the calculator, I left the n and K values as they were, but ran it with P=0.05 to test how reasonable an expectation of 14 hits in 120 attacks would be, if the aura WERE implemented and the Displacer Beast only had a 5% chance to hit (would have to roll a 20). The results are the nail in the coffin of the Protection aura. Given a 5% chance to hit:

The probability of exactly 14 hits in 120 trials is 0.18%.
The probability of exactly, or fewer than, 14 hits in 120 trials is 99.90%. (!)
The probability of exactly, or more than, 14 hits out of 120 trials is 0.28%.

If the Paladin's innate Protection from Evil 10' aura were actually implemented in the game, my results are an extraordinary outlier, quite literally a one-in-a-thousand fluke. If the aura is NOT implemented, my results are pretty much at the crest of the bell curve of expected results.

So it seems like we can say conclusively that the Paladin gets no innate Protection aura, much like the Ranger gets no dual-wielding bonus and Elves get no bonuses with swords or bows. Lay on Hands is still nice, as are the supplemental low-level Cleric spells late in EOB2, so those are the reasons folks will want to roll a Paladin—not for the innate defensive buff, which unfortunately doesn't exist.

You know, we always used to do these kind of tests on the FRUA forums to figure out if stuff had been implemented, but I never saw anyone do formal hypothesis testing before. ;) Kudos!

Displacer beasts are not evil in 1st edition AD&D.

Edit: Nevermind, the manual reveals the game is 2nd edition.