In our <i>Behind the Screens</i> thread, John A1nut asked:
Why is it a lower chance of success for something that I've already done? That's like saying it's easier to knit the first time you do it, and it gets harder the next time.Well, from a strictly technical perspective, the answer is that it is sometimes easier (and sometimes harder) to learn something than to do it again because the odds are calculated differently.
It's entirely possible that you could design a skill of a low enough bias relative to your bias that the bias difference would make it relatively simple for you to learn it the first time, but the penalties on it would be so great you would never do it again. It's never happened that I recall, but there have been some strong discrepancies in such things, and there are situations in which players decide to design a new skill instead of using an old one precisely to get the advantage of bias (although in the long term this is not as good a strategy, since increasing skill ability level will increase the power of many skills).
In the rules, we say that this can be written off to beginner's luck.
You, however, want a better justification of it than that. It turns out that I have one.
The fact is, your skill learning roll is about whether you teach yourself to work the new skill; if you make it part of the learning process that you would actually do the skill, then successfully learning it means successfully doing it. On the other hand, it's entirely possible to teach yourself a new skill without doing it.
Let's take for an example learning to make brownies. Most people who want to learn to make brownies get all the materials together, read the recipe, follow it step by step, and in the end have a pan of brownies, and now they are amateurs at making brownies. However, it is entirely possible for someone to get hold of the recipe, have none of the materials available, but run through the process several times carefully in their own minds until they know how to do it. Now they know how to make brownies, and few would doubt that such a person has as good a chance of making a pan of brownies as the other person who made one once. (We could even include in this that the one person has someone standing there instructing them, and the other person is standing there watching someone who is instructing them--even watching Rachel Ray on television--to get the information.)
So whether you successfully do what you are teaching yourself to do depends on whether you have designed the learning process "hands on" such that you are learning by doing.
That still doesn't explain why it might be easier for you to do something while you're learning it than it would be the next time; but I'm getting there.
You are of course very much focused on thinking through the process the first time, deciding what has to be done. It's not entirely unreasonable to suggest that this level of focus is going to make a difference. Ah, but you will object that you can apply that same level of focus the next time, and I can't really argue with that.
I can, though, argue with the very premise.
You are asking why it is more likely that you would succeed when you are learning it than after you have learned it. That, though, is an artifact of the mechanics--something your character cannot possibly know. When he is learning it, he knows that he will succeed or he will not succeed. He might have some feel for how easy or difficult it would be to learn this, or to do it once he knows it. In fact, though, he never really knows how likely--or unlikely--it is that he will succeed. He only knows whether he succeeds or not. It could be that he had a one percent chance of success, and he rolled the one; it could be that he had a hundred fifty percent chance of success and it didn't matter what he rolled. He knows that he succeeded.
Further, if he succeeded, he never has that particular probability of success again.
Thus I am looking at "how likely is it that you can learn how to do this," and giving you successfully doing it once for free.
In designing the rules, we gave some consideration to making such situations a double roll: that you had to roll once to see if you could learn the skill (without actually doing it) and again to see whether you could use the new skill successfully. After some discussion and debate, we agreed that if you designed your skill learning process such that you would learn by doing, successfully learning would be successfully doing it exactly once, but if you designed the skill learning process such that you would learn the skill and then do it, you had to face the double roll.
It doesn't usually come up; most people attempt to teach themselves new skills by trying to do them, most of the time.
--M. J. Young
