One of the challenges of game design is occasionally you come across statistical oddities that make things either challenging or interesting.. Well that happened for me tonight. I was working up a list of artifacts for a particular cleric which according to the random rolls the treasure table indicated should have three written works. For the first one I rolled d% on the content table and got a result greater than 66. The table indicates that for rolls of 66-100 that you should roll twice more. (Many tables in Harnmaster have this kind of structure, but I don't think any others have such a high probability.)

I rolled a second time and got another result above 66 indicating rolling twice more. If N is the number of rolls remaining then at this point N=3. So I rolled again for a third consecutive time got a result above 66 (N=4). Perhaps an anomaly; I rolled several more times and after generating 21 different types of content for this one written work, and N had grown to over 25. Clearly something was odd.

Once your result is "roll twice more" on this table the expected value of the subsequent two rolls is that you increase N by 4/9. Almost a 50% chance that N will increase.

Of course a GM can always ignore this type of result, but my preference would be that the table be fixed so that "roll twice more" occurs infrequently enough that this sort of event should not be common...

_________________
There is no truth to the vile rumor that I am a fireball throwing Peonian Priest!

Well; reading is an addictive habit

It does promote; not an entirely inaccurate image imho, of clerics drowning in paperwork..

_________________
Plots and schemes are the same thing..

The thing I've always found odd about that table is the relative lack of religious texts. If it were a table for Medieval Terra, it would be something like "01-90 Religious Work" but for Harn it's only a few percent. So, if you think all the various monks and priests of the 10 deity pantheon ought to be producing a few more pages, you could amend that last result to take up some of the slack.

A little more math and it's clear that if the odds of "roll twice more" on a table exceeds 41% then the expected value of the number of rolls you will make will increase with each roll. Which is the case with this table. Aside from the image of clerics toiling away in the scriptorium. (Which strangely, or perhaps not so strangely as the case may be, brings to mind images of the men of the Crimson Permanent Assurance toiling away at their ledgers). I like the idea of reducing the "roll twice more" number to around 17% or so and moving 50% of the results to "religious text"...

_________________
There is no truth to the vile rumor that I am a fireball throwing Peonian Priest!

Your dice are broken, so you need to go out and buy new ones. Thats all.