I went back as far as I could. I couldn't find the 2e MM on Amazon. For 1e, I just found a PHB, and it had a single (5 star) review, which is too low a sample even by my standards.
Anyway, here's a table of what I found.
4E, on the other hand, was seen by many as coming out too soon after 3E. The majority of the people who are vocal in their dislike for 4E also include tirades about it being a money-grubbing venture (not all of them, just a lot of them). These types are very likely to run out and slam reviews and scores hard everywhere because they literally want the edition to fail. After all, if it succeeds, it validates the "money-grubbing" strategy.
That'd be my take.
That'd be my take.
And in product marketing that'd lead to the ruin of a company. That's why every time someone asks WotC if there will be a 4.5, the company trips all over itself to make sure people know that it won't happen...
Now the problem with Amazon's rating system is simple, people who feel slighted are far more likely to complain than people who feel elated are to praise. For instance I like 4th ed better than 3rd (as does most of my group) but I haven't gone to Amazon and posted a review.
Basically it is actually one of the very first lessons you learn in a statistics class, you have to go out and get the answers yourself, you can't let them come to you.
My assumption would be probably simpler. When 2ed stuff came out no Amazon. Now the people buying up 2ed stuff are already big fans thus they feel strongly about this stuff. Hence the good reviews.
Now the problem with Amazon's rating system is simple, people who feel slighted are far more likely to complain than people who feel elated are to praise. For instance I like 4th ed better than 3rd (as does most of my group) but I haven't gone to Amazon and posted a review.
Basically it is actually one of the very first lessons you learn in a statistics class, you have to go out and get the answers yourself, you can't let them come to you.
The general "people who feel slighted are more likely to complain than people who feel elated are to praise" principle applies to all editions, though, so I don't really see how it serves as an explanation for the decline.
Admittedly, the results for the 2e DMG can certainly be discarded on the grounds of small sample size. The results of the 2e PHB have a larger sample size, but since it it was out in 1989, while Amazon wasn't really started until 1995, its results might also be considered dubious since its review scores weren't impacted by the people who first bought the 2e PHB the day it came out, read it over, and then immediately wrote a review on Amazon.
However, when the 3e and 3.5e cores were released, Amazon was already up, running, and a fairly prominent bookseller, its status not changed significantly from its status today. When the 3e PHB came it, people could buy it on the first day of its availability, read it, and review it on Amazon, so that's essentially the same as 3.5e and 4e.
The general "people who feel slighted are more likely to complain than people who feel elated are to praise" principle applies to all editions, though, so I don't really see how it serves as an explanation for the decline.
Admittedly, the results for the 2e DMG can certainly be discarded on the grounds of small sample size. The results of the 2e PHB have a larger sample size, but since it it was out in 1989, while Amazon wasn't really started until 1995, its results might also be considered dubious since its review scores weren't impacted by the people who first bought the 2e PHB the day it came out, read it over, and then immediately wrote a review on Amazon.
However, when the 3e and 3.5e cores were released, Amazon was already up, running, and a fairly prominent bookseller, its status not changed significantly from its status today. When the 3e PHB came it, people could buy it on the first day of its availability, read it, and review it on Amazon, so that's essentially the same as 3.5e and 4e.
Actually it won't matter for the 2nd ed books but you pretty much get the reason why so I won't go into more detail. The 3rd ed books have had more time to be out over 4th thus they have had more time to even out. I bet if you tracked the initial release of them you would find a similar event. When they were first released a lot of detractors would of voted poorly (or not another factor to consider is that the internet was not as prevalent in 2000 as it is now.) 3.5 would also be similar except would probably be a better measure since it was current in the scheme of things yet it still had 3 years to even out compared to 4th's 5 months or so.
Either way the current results do not tell us anything.
I bet if you tracked the initial release of them you would find a similar event. When they were first released a lot of detractors would of voted poorly (or not another factor to consider is that the internet was not as prevalent in 2000 as it is now.) 3.5 would also be similar except would probably be a better measure since it was current in the scheme of things yet it still had 3 years to even out compared to 4th's 5 months or so.
Yes, that was my idea too ("shortly after its release, the 3.5 PHB's ratings on Amazon looked a lot more like the 4e PHB's do now"). One could actually verify this, since each customer review has its date recorded. One could look at the reviews of the 3.0 PHB in 2000, in 2001, 2002, etc, and the reviews of 3.5 immediately after its release compared to reviews in 2006, say.
But... I'm not willing to do that much work. Just glancing at the star summary tables was enough for me
.Namely, in the case of each PHB except the 2e AD&D one, organize the reviews in "Newest First" format (rather than "most helpful first"), go to the very back, then check the average score of the first thirty reviews. Thus, reviews of the 3.0 PHB from 2003 and reviews of the 3.5 PHB from 2007 are excluded: you only really get the ones from shortly after the release. Naturally, this reduces sample size, but also might compare a more relevant sample.
My prediction: the averages of the first thirty reviews will be much closer together than the averages of the total reviews.
Alright: results!
An idea occured to adjust for time availability without too much extra effort.
Namely, in the case of each PHB except the 2e AD&D one, organize the reviews in "Newest First" format (rather than "most helpful first"), go to the very back, then check the average score of the first thirty reviews. Thus, reviews of the 3.0 PHB from 2003 and reviews of the 3.5 PHB from 2007 are excluded: you only really get the ones from shortly after the release. Naturally, this reduces sample size, but also might compare a more relevant sample.
My prediction: the averages of the first thirty reviews will be much closer together than the averages of the total reviews.
Alright: results!
That is interesting and a lot more helpful in terms of data. I think the case for the high reviews of 3rd ed early on was due to the fact that many people were waiting for a new edition at that point.
So, that would also apply to 4.0
So, that would also apply to 4.0
Possibly, like I said it is only a guess.
That data supports my earlier commentary.
You can see from the distribution - note the curve for the DMG and MM, and compare it to the PHB, as well as the huge population of 1s relative to 2s and 3s.
Its not disliked by people. Its a small number of trolls - about 60 of them.
This is why Amazon reviews are oftentimes worthless. Without at least a few thousand reviews, you will see erroneous data like that.
If you throw out the trolls, the curve looks very similar to the 3.x curves. You can see some countertrolling on the 3.5 PHB.
So really, its pretty much exactly the same in terms of rating, realistically speaking.