I'll bet you haven't looked at these records, but are instead relying on what sounds likely to you. In the same vein, if one looks at the records of successfully completed circumnavigations…the preponderance is for celestial navigation. In other words, both statements assume the past.
The reason I don't think your statement is true (nor mine) is that if you look at the World ARC and other types of rallies that have sent many, many boats RTW, you will find that most of them are not full keel, or even long keel, boats.
For every Pardey boat in the past, there are many new boats today.
You'll lose that bet if the present time was near the years 2000-2005 when I did my topical research when looking for a well built and well designed long distance cruising boat.
At that time there were several organizations, including Latitude38 (as a statistical sample), who published lists of circumnavigation by cruisers on a regional basis that showed names/dates/boat model, etc. These lists, at that time, were mostly in quasi-excel spreadsheet format or could easily be loaded into spreadsheet format for easy scrutiny and comparison of the data. When when that data was restricted ~1975-2005 that data showed the preponderance of full/long keeled boats; and, for 'american designed' boats - designed by Bob Perry, Bill Creighlock, Robt. Harris, etc.
I believe the Valiant40 (long fin) is still touted as one the most successful circumnavigating sailboats; and, quickly followed in 'numbers' by other variants of the essentially the same basic design.
These lists have somewhat disappeared from the internet; although all I can find now is 'more compact', yet less spreadsheet-friendly: Latitude 38 - West Coast Circumnavigators' List
Of course, the evolution to more efficient and stronger composite etc. construction has advanced the trend away from full/long keeled boats ... but that data doesn't seem to show even in the current (quite sparse) listings that are available.
Latitude 38 - West Coast Circumnavigators' List
and one can use even this small amount of data as a statistical correlation. And too, one can look at that Yachtworld article posted previously and gain some preliminary statistical insight/trend of where 72 fin keels have fallen off in essential the same time period and where ~24 souls were lost.
As regards 'safety' and the current/recent historical record, I don't think you or anyone else can find a comparable data set over the same time frame of a proportional amount of full/long keelers having the same issues.
Ed./Add: the same argument can be carried by a simple rationalization of the loss of spade rudders vs. 'barn door'/skeg mounted rudders.