What to Do When Two of Your Own Rules Disagree
The conflict arrives quietly. One rule describes a valid entry and the conditions are met. Another rule describes a circumstance in which you do not trade, and that condition is also met. Both were written carefully, neither is wrong, and together they produce no answer at all. The plan has run out of instructions at precisely the moment it was supposed to be most useful.
Conflicts Are a Property of Short Sentences
A rule set is a compression. It takes a complicated set of judgements about when an approach works and squeezes them into a handful of sentences you can read quickly. Compression loses information, and the lost information shows up as cases the sentences do not jointly cover. Adding more rules does not fix this. It creates more pairs that can disagree, and the number of pairs grows faster than the number of rules.
So the goal is not a conflict free rule set. That is not achievable with a list short enough to be read at the open. The goal is a rule set that already contains its own tie breaker.
Precedence Is the Missing Piece
Most plans list rules as though they were equals. They are not. Some describe when an opportunity exists and some describe when you must not act regardless of opportunity. Those two categories are not symmetric, and writing them in one flat list hides the asymmetry.
The cleanest arrangement puts the prohibitions above the permissions. A rule that says do not trade under a stated condition outranks any rule that says take this entry, without needing to be argued about. This is not because avoidance is inherently wiser. It is because a fixed order produces the same answer every time, and a fixed answer you can live with beats a better answer you cannot reach consistently.
Resolve It Before It Happens
The moment a conflict surfaces during a live session, you have already lost the argument, because the resolution will be made by someone who can see the chart and knows which way they would like it to go. The rule that gets bent will be whichever one is currently inconvenient, and afterwards you will be able to explain why that was reasonable.
A conflict noticed during a session should therefore be recorded and skipped rather than resolved. No trade is the correct output of a plan that has produced no answer. That is unsatisfying, particularly if the trade you skipped would have worked, but the alternative is a plan whose contradictions are settled by preference and then justified after the fact.
Repairing the Pair Afterwards
Once the session is over, a conflict is a small and tractable piece of work. Usually one of three things is true. The two rules are describing overlapping territory and one of them should be narrowed. One of them is not really a rule and belongs in the preferences list. Or both are genuine and the plan needs an explicit statement about which wins, added as its own line rather than left implied.
The third case is the most common and the most often skipped, because writing an ordering feels bureaucratic when there are only a few rules. It stops feeling bureaucratic the second time the same pair collides and you find yourself having the same argument with a live position pending.
The List Should Get Simpler, Not Longer
Every conflict is evidence about the design of the rule set, not just about the two rules involved. A plan that produces frequent contradictions is usually a plan built by accumulation, where each new rule was added in response to a specific painful session without anyone checking how it interacted with what was already there.
The repair for that is subtraction. Rules that have never once changed a decision can go. Rules that fire on the same conditions as another rule can be merged. A shorter list has fewer pairs, produces fewer contradictions, and is more likely to be read in full rather than remembered approximately. Remembering a rule set approximately is how you end up with conflicts you never even notice, resolved silently in favour of the trade you wanted.