I touched on the idea of Drift in an earlier Blog post (HERE) but I think it is worth discussing on its own. This is something I've observed in my own play tests (I'm sure others have noticed this as well and I'm sure the term, or variations of it, is floating around the net). It occurs when the rules of the game state X but over time you start to do Y instead without realizing the error. This simply happens over time and feels natural so you don't notice until you re-read the rule in the manuscript or book. In particular I'm talking about Drift that arises when you are developing a game, but it definitely is something you see with existing systems.
Drift is something we are always looking for. Most recently it occurred with our Restrain rules in Wandering Heroes of Ogre Gate. We always wanted to keep this simple, not get into more involved mechanics. Somehow it shifted at the table away from what the text said and toward something else.
I see this all the time and it is usually a sign of one of two things: the rule cuts against natural expectations or the rule doesn't adequately handle what it is meant to represent (it can also just mean people are being forgetful, thinking of another system or weren't paying attention to the rules document). When it cuts against expectations, that can be okay. Sometimes you are just used to mechanics working a certain way and need to get used to change. However if it continues to be a problem, I think it is a sign that the mechanic doesn't fit well with how people want to run the game and changing it becomes a strong option.
If the mechanic simply doesn't do the job it is supposed to do, then you have to rework it or get rid of it. Drift can be a product of fixing rules live during play without deliberate effort. You are just unconsciously smoothing it over, and that becomes an unofficial rule that replaces what the text says.
Sometimes Drift happens because people didn't review the rules document enough. I generally make a point of referencing rules we are working on every time they come up initially, so this is rare, but it still crops up if I let myself get lazy.
The problem with Drift is it is one of the more insidious things that can creep into a game without you realizing it before it hits the shelves. If you don't catch drift then you are effectively playing a different game then you are releasing to the public and that is a huge issue. That is a significant disconnect and it makes it difficult to address ongoing issues with the system.
Preventing Drift and catching it when it does arise, is something I've learned to make a priority. The last thing you want to be doing is fixing Drift right before edits or (even worse) right before layout.
Here is what I do to prevent Drift:
1) Read my own rules: This sounds simple, it sounds easy, but when you've read your own writing 100 times, it can be easy to miss content and project content onto it. So rather than re-reading the manuscript in one sitting each week, I break it into chunks, slowly going over each section individually and reviewing key sections before and after playtest.
2) Print my own rules: I always print one new copy of the manuscript before each game session (I print several in the early stages, but it eventually gets too expensive to print multiple copies every game). I also have two players who have the most current version of the rules on their laptops.
3) Look-up rules during play: I usually assign one player to be the official rules reviewer/lawyer for the session. I also keep my copy of the rules handy and look them up with they arise. As an example when playing Wandering Heroes of Ogre Gate, we review the text of each technique when it comes up so we know we are applying it correctly.
4) I playtest RAW: Outside of Playtest, I don't worry about RAW as much. But in a playtest for a system I'm developing I play the rules as written and am pretty rigid about it. I just want to make sure the text on the page reflects what is actually going on at the table.
All of this is quite helpful. The biggest things really though are the regular reviews of the text and looking things up during play. The latter definitely slows things down. It can be a pain to lose momentum just to look up a minor rule, but Drift happens if you don't.