They haven't cited a source or their reasoning, I would guess they mean "seems to be an exception by observation"
Emphasis on "observation".
Currently we infer planets that orbit stars that are extremely distant by starlight dipping in intensity .. this is a method that works better with larger planets and that skews the results of observing distant planets.
We have observed some small rocky earth like planets IIRC but these are rare, observations perhaps due more to luck in looking than frequency of existence.
This is the crux of the problem; does the distribution of planets we observe match the distribution of planets that exist?
Emphasis on "observation".
Currently we infer planets that orbit stars that are extremely distant by starlight dipping in intensity .. this is a method that works better with larger planets and that skews the results of observing distant planets.
We have observed some small rocky earth like planets IIRC but these are rare, observations perhaps due more to luck in looking than frequency of existence.
This is the crux of the problem; does the distribution of planets we observe match the distribution of planets that exist?