On-off cycling peaks at about 50% duty. On the coldest day the unit runs nearly continuously and barely cycles at all.
Both ends collapse: at very low load the space barely cools, so the off-time runs away; at very high load the unit can barely keep up, so the on-time runs away. The maximum sits in between, at ordinary shoulder-season weather for a correctly sized system. An oversized unit reaches that point on a much milder day and spends more of the year there, which is why oversizing — bought for reassurance — is the commonest cause of short-cycling.
Real controllers deliberately break the relationship. Minimum on and off timers, anticipators that switch off early, and multi-stage plant all exist to stop the equipment doing what the simple physics says it would. If your system cycles differently from the model, that is usually protection working rather than the model being wrong.
Estimate with the rule, then check it against the calculator that models it properly.
Open On-Off Control & Hysteresis →On-off cycling peaks at about 50% duty. On the coldest day the unit runs nearly continuously and barely cycles at all. Both ends collapse: at very low load the space barely cools, so the off-time runs away; at very high load the unit can barely keep up, so the on-time runs away.