When one radio amateur sends a signal to another, there’s a chance that the signal will be received above the threshold of reception. If the signal received is indeed above the threshold of reception, an exchange of voice or data messages may be possible.
The threshold of reception is different for each transmission mode, frequency band, local noise level and transmitting and receiving equipment. I describe elsewhere the notion of system value – the maximum loss permissible between transmitter and receiver. The threshold is a description of the receiver limit in the system value calculation.
The propagation loss between the two stations is not fixed. The environment between the two is constantly changing. The result is that sometimes the threshold is exceeded, and sometimes it’s not.
There are two principal dependent variables in success – time and locations.
For every point on the Earth’s surface attempting communication with every other point, there is a chance, expressed in percentage time and locations for each path, for which a viable exchange might take place.
This state is fundamental to understanding propagation, particularly where propagation is via a sporadic medium like an aurora or via sporadic ionisation of the ionosphere’s E region.
To elaborate, consider this. The chance of an exchange between two radio amateurs with a pair of VHF handhelds each anywhere on a football pitch approaches 100% of locations for 100% of time. That’s to say that if the two users wandered at random around the pitch, they’d almost always be in contact – and if they stood still at any two points, they would be able to be in touch almost for ever (or until the batteries expired!).
The further apart the two stations move, the lower the percentage time and locations.
On the other hand, if one station is in Scotland and the other is in Spain, each wandering at random in their country, there would be some locations where they can communicate, and others where the received signal will be lower than threshold.
If the stations remained at two chosen locations and attempted to communicate over a long period of time, communications would be possible for some of the time. The rest of the time, the signal would fade below the threshold.
So, whatever the scenario, communications are defined in terms of ‘threshold exceeded (or communications possible) for x% of time over y% of locations’.
This applies across all frequencies and propagation modes.
Often propagation prediction software fixes the locations and reports results in terms of path or circuit reliability as a percentage of time.
To complicate things slightly, the time variate element has two parts: short term and long term. Path loss varies second by second, and minute by minute. Communications can be established, and then fade out minutes after. And in the long term, signals vary hour by hour, week by week, month by month, and year by year.
This leads neatly into the idea of a fade margin. If the average received signal at a receiver is just on threshold, the signal will, in the short term, be usable for half the time. To be more confident about communicating between two stations at given locations, the received signal needs to be well above the threshold – by increasing the transmitter power or changing some other equipment characteristic at either end to yield a fade margin.
The same is true in the long term. And this is seen when stations make use of data modes like FT8, which has a huge effective gain over SSB. A path that allows SSB communication when conditions are good can fade well below threshold months and years later. If the technology changes to FT8, the system value, the maximum loss between transmitter port and receiver port, improves hugely and overcomes the fade, effectively re-establishing a fade margin, albeit for a lower information rate.
So, what can we conclude?
Simply that chance governs communication between amateur stations.
The idea is shown graphically below.

fixing time for which comms are available1
Every path, between pairs of radio amateurs across the globe, can be described by a path loss for a given propagation mode and frequency. For a given system value (technology, power, gains), the path loss will allow communications for some percentage of time. In the example shown, communications will be available for 10% of time. For 90% of time, the path loss is just too high. The path may improve, and path loss reduce, later in the day, tomorrow, next month or next year – or indeed it may never improve enough.
Many amateurs embrace this tail of the path loss distribution. Low chance gives the well-established concept of ‘DX’.
DX is:
Establishing a communication, when propagation conditions suggest that there’s a small chance (in percentage time and locations). Many radio amateurs chase low probability events like meteor trails simply because the low chance excites them.
DX is also:
Establishing a communication over a path that’s a first for the hobby, or perhaps working a path that is a first for you. You are the lucky one that happened to be there at the right time and place to exploit the chance.
Good luck and enjoy the QSO when and wherever it happens!
[1] Path loss is typically normally distributed about a median. Hence the use of a normal or Gaussian distribution.