The single most common mistake with Storage Hunters mutations is treating stacked modifiers as additive. Two independent 5% mutations do not give you a 10% shot at both — they give you 0.25%. Additive intuition overestimates double-mutation odds by a factor of forty at those rates, which is why "I've opened hundreds and never seen a double" is a completely ordinary experience rather than evidence of a bug.
Each reroll is independent. The unit you burned twenty rerolls on is exactly as likely to mutate on reroll twenty-one as it was on reroll one. This sounds obvious written down and is nearly impossible to feel while playing, which is the gambler's fallacy doing its usual work. The useful reframing: rerolls buy you additional independent draws, not accumulated progress toward a guarantee — unless the game implements an explicit counter, in which case it will say so.
The decision is an expected-value question with two inputs people usually skip: the cost of the reroll in the currency you are actually short of, and the marginal value of the mutation over what you already have. A mutation that adds 8% to a stat you are not bottlenecked on is worth close to nothing regardless of how rare it is. Rarity and usefulness are independent axes, and the trading market prices them independently too.
If you want to put real numbers on a specific stack rather than reason in the abstract, themutation odds calculator for Storage Huntershandles the compounding across multiple modifiers, which is the step where hand arithmetic tends to drift.
Community mutation tables are usually built from player-reported samples, not from datamined values. Player-reported data is biased toward rare outcomes — people post about the mutation they hit, not the four hundred openings that produced nothing. Expect community-sourced rates to read optimistic unless the source explicitly says it corrected for reporting bias.