My humble apologies to the former Canadian folk music band Tanglefoot who came from the general area I live in. I borrowed (and edited) one of their song titles for this post.
Just when I thought I had the theory and practice of ground radials down pat, along comes another surprise. After several months of deploying my portable center-loaded vertical antenna, with just four 10ft ground-mounted radial wires, I decided to give my old, bruised and battered MFJ-1979 17ft telescoping whip another try. And in so doing I learned a new trick.
My portable center-loaded vertical antenna was always mounted on a short pole, raising it about 3 feet off the ground. The four 10ft radials were arranged around the base of the pole. SWR was good – and if it wasn’t, an adjustment of the coil tap would kick it into shape, no problem.
But when I raised the MFJ-1979, it was on a spike only inches above the ground. I spread out the same set of radials that had got me through many successful POTA activations and was surprised to see the SWR was 2.2:1. Que pasa?

I shortened the whip a little but the SWR worsened. I measured the length of the whip when fully extended – 17ft 4inches. That’s too long for even the CW portion of the 20m band, so shortening it should have fixed the SWR. Maybe my bruised and battered MFJ-1979 is about to reach its expiry date.
I had just taken delivery of an extra long (18ft 5inches) telescoping whip from AliExpress, so I replaced the MFJ-1979 with the new whip and fully extended it. Bingo, the SWR was now down to 1.7:1. Revelation: whip length and radial density are interactive. When I added more radials the SWR dropped to 1.2:1 even with a shortened MFJ-1979.
Actually, it wasn’t even quite that straightforward. Test #1 involved replacing the 10ft radials with 23ft wires. The SWR was 1.7:1. Not bad, but could be better. I found that folding the ends of the 23ft wires back improved the SWR even more. So the secret, as I surreptitiously slipped into the last paragraph, is radial density. How much real estate do the radials cover? Oh yes, I could add lots more radials, but this is a rapidly deployable field antenna. I have to be able to set it up and pack it away in as little time as possible.
Test #2 involved eight 13ft radials. Same result. SWR around 1.6:1. Changing the layout of the radials had no impact. Hmmm, time for some chin-stroking deep thought.
Test #3 involved raising the same whip about 2 feet above the ground and laying out just four of those 13ft radials. Result? SWR was now down to 1.2:1 and it didn’t matter whether the radials were laid out in nice straight lines or just tossed haphazardly on the ground.

Ah, but always remember that low SWR is not necessarily an indicator of how well an antenna will perform – a dummy load has a 1:1 SWR. A quick POTA QSO with an activator somewhere down in the United States confirmed that the antenna was good enough for field portable ops.
Conclusion
The online armchair antenna experts are correct when they advise installing a lot of radials with a vertical antenna. But what they may be missing – as my winter backyard experiments have shown – is that it is much easier to raise the base of the whip so that the feedpoint is away from the ground – even by as little as a couple of feet. Same result, less effort!