D&D General As of 1998, 4,007,685 people played AD&D in the US, as estimated by Ben Riggs.

[John Lloyd1]

I had a player who xeroxed an entire Darksun book and the RC.

As a high school student in the 80s, I didn't have ready access to a photocopier. Mind you, I did manage to get my Dad to photocopy an entire 1st edition book at work (Players, I think). The books were pretty expensive in Australia at the time.

 

[John Lloyd1]

It’s called a Fermi Estimation and it’s a surprisingly reliable way to get within an order of magnitude of the actual number.

Thanks. I was trying to remember what it was called.

 

[Charlaquin]

Shannon reposted his response

Ben starts a great discussion on D&D players then & now.

Here was my own response, which of course is highly speculative too:

==
I think it's possible the classic D&D players could be an order of magnitude higher.

To start with, 4 wasn't the standard group size then. It maybe averaged 6-8 based on the player numbers on modules, but anecdotally people often talked about 10-20 member groups.
And then you have to start looking at not just parallel players, but sequential players (e.g., how many players did a single GM lead over time).

That's of course lessened by the number of players who also bought DMGs, but that was also lower, particularly in the '80s. But I think 4 million could easily become 10 million or 20 million or even 40 million.

(But a lovely discussion point whatever the answer is; thank you!)
==

And that’s exactly how Fermi Estimates work. Even with no data, you can make broad assumptions based on comparisons to values you do know, and the over-estimates will tend to “cancel out” the under-estimates to result in a guess that, while not precise, is within quite a reasonable margin of error given the large numbers involved.

 

[Charlaquin]

Thanks. I was trying to remember what it was called.

I had to use some quick Google Fu to find the name myself

 

[a.everett1287]

It’s called a Fermi Estimation and it’s a surprisingly reliable way to get within an order of magnitude of the actual number.

It's almost like Ben Riggs isn't just someone pulling data out of nowhere, and uses actual established methodologies.

Gee, who would have thought.

 

[Parmandur]

But I thought it was such a terrible assumption...

The assumptions in the original post seem sketchy, but the final number seems to be within the right ballpark, though seriously low-balling still.

 

[Charlaquin]

The assumptions in the original post seem sketchy, but the final number seems to be within the right ballpark, though seriously low-balling still.

Enrico Fermi powers, activate!

 

[a.everett1287]

The assumptions in the original post seem sketchy, but the final number seems to be within the right ballpark, though seriously low-balling still.

So then there's a distinct possibility that the 'assumptions' Ben Riggs made weren't actually that sketchy or terrible.

 

[Parmandur]

So then there's a distinct possibility that the 'assumptions' Ben Riggs made weren't actually that sketchy or terrible.

Hmm, still less useful than the actual market research data that WotC published from the same period of time.

 

[Charlaquin]

So then there's a distinct possibility that the 'assumptions' Ben Riggs made weren't actually that sketchy or terrible.

Hmm, still less useful than the actual market research data that WotC published from the same period of time.

Well, the beauty of the Fermi Estimation is that your assumptions don’t necessarily need to be accurate, as long as you’re “in the right ballpark.” And as long as you understand that what you’re getting is not an accurate answer, but a reasonable range. Some of your estimations will be too high, and some will be too low, but as you multiply these factors, the over-estimates and the under-estimates tend to cancel out and you end up with a result that’s “close enough” (that is to say, within about an order of magnitude), which makes it a useful tool for checking your calculations that are meant to produce an accurate results against. If the calculations you use to try and find an accurate answer produce one that’s off from your Fermi Estimate by more than about an order of magnitude, there’s a good chance that there’s a flaw in your methodology.