The Cumulative -1 Rule, Or Khadim tries to do Mathematics Options Track this topic Email this topic Print this topic Download this topic Subscribe to this forum Display Modes Outline Switch to: Standard Switch to: Linear+

[knasser]

I was going to post this on the Chemistry thread, but there have been a few threads recently about the cumulative -1 rule and this is a general query on all of them, hence a new thread.

In SR4 pre-errata, there was an optional rule stating that you could only roll as many attempts on an Extended Test as you had dice in your pool. This was to stop PCs being able to accomplish anything given enough time and was invoked according to GM judgement. Post-errata (SR4A), this has been amended to say that your pool reduces by 1 on each attempt until you obviously stop at 0.

There have been some issues with whether this actually makes some things impossible where they should not be. One query was on the Chemistry skill for making explosives. Another (a bit more robust) was on the Data Search rules.

I worked out the following chances of success in Shadowrun (both extended and non-extended tests). I think they'll be useful for GM's to judge exactly what they're asking of their players.

http://knasser.me.uk/content/shadowrun/sr_..._by_knasser.pdf

I would like someone who knows their mathematics to confirm my numbers are correct, please.

If the numbers are correct, then I think the above shows a bit of a problem. Basically, with either the older 'maximum number of rolls = size of pool' or the new 'pool reduces by 1 each roll', you have a quite narrow band in the middle of the range of dice pools where there is uncertainty of outcome (chance of success in the range of 30-70%) and with dice pools above or below that middle band, the chance rapidly becomes either near certain failure or near certain success.

For example: A person with a dice pool of 8 trying for a threshold 12 test has a 56% chance of success. Take them to dice pool 9 and it jumps to 87%. Drop them down to dice pool 7 and it plummets to 19%.

That's awkward from both a playability and a realism point of view. Essentially, the more times you roll a dice, the more your results are going to descend on an average. Roll three dice and you might get 3 hits or you might get 0 or anything in between and no result will be surprising. But roll a hundred dice (which is how many you can roll with a dice pool of 14 reducing by 1 each roll) and you're very likely to come close to a third of your results being hits.

So two three questions:

One: Can anyone see anything wrong with my maths?

Two: Does anyone else find this a problem?

Three: What ways can we amend this to work better?

K.