I never used the percent alarms because they only apply their trigger condition to some arbitrary fixed checkpoint price that's becoming more and more obsolete as time goes on. The way percent-change alarms would be really useful to me would be if they behaved like trailing-stop orders: percent-change always calculated vs. the latest "opposite-direction peak" price of the currency. So the checkpoint/reference price would always be the lowest price ever seen since the creation of the alarm in case of a +n% trigger condition, or the highest price ever seen in case of a -n% trigger. The point is for this alarm to only react if there's a large enough change to signal a trend reversal (or trend acceleration).
I never used the percent alarms because they only apply their trigger condition to some arbitrary fixed checkpoint price that's becoming more and more obsolete as time goes on. The way percent-change alarms would be really useful to me would be if they behaved like trailing-stop orders: percent-change always calculated vs. the latest "opposite-direction peak" price of the currency. So the checkpoint/reference price would always be the lowest price ever seen since the creation of the alarm in case of a +n% trigger condition, or the highest price ever seen in case of a -n% trigger. The point is for this alarm to only react if there's a large enough change to signal a trend reversal (or trend acceleration).