IMO, this is a misrepresentation because those links go to Windows 7, a 9 year old OS where mainstream support ended over 3 years ago. It does not tell the whole story.
Why would the OS with, arguably, the more dominant market share be in any way a misrepresentation ? Netmarketshare shows Win 7 having 50% more (44.40 % to 29.88 %) active installs than Win 10 ...Stat counter has them just about even (last I checked (February IIRC).
https://tinyurl.com/ycx8jftg
If you are a Win 10 user, you just have to click on the Win 10 and get the other results, certainly not beyond the capability of most forum users
And what changes ?
Kaspersky again scored 6.0 / 6.0 / 6.0
BitDefender again didn't quite do as well with 6.0 / 6.0 / 5.0
Microsoft did worse with 6.0 / 5.4 / 5.0
Avast Free - Same 6.0 / 5.5 / 5.5
AVG - Same 6.0 / 5.5 / 5.5
Norton - did worse at 6.0 / 5.0 / 5.5
Trend Micro - dropped a bit to 6.0 / 5.5 / 6.0
And let's not forget, that's January / February results. Not everyone "hits it outta the park" so to speak in every test.
However, if you don't use the internet and don't install files from media you are also safe. I visted a friend recently and I was to arrive before he got home... he said
"just go inside, door's open... I haven't locked my door in 30 years unless I'm outta town for cupla days". So what is safe / unsafe... door locks, door locks and alarms.... locks, alarms and cameras. Each involves a level of better protection at increased costs.
The user has to make that judgement. So how much is your data worth ? How much are you willing to pay to have your system cleaned. I spent 9 hours cleaning my S-I-L's machine it had over 1200 infections with a "free solution". As for protection being the only consideration, I reject that notion entirely. When BD changed the management model, I went from a single download and network installs to having to download and enter information from each machine on the network, greatly increasing the time and cost of ownership. When a AV impacts the speed of our systems, this can affect one's ability to earn their living. Software installs can take up to 3 times longer with free solutions versus paid. Frequently used web sites can take 10 times as long to open than paid solutions. When AVs pick up false positives, that can mean hours of time trying to figure out where the problem is. Like insurance, it's never really important till you have a problem.
So the questions each user should answer for themselves are ...
A. Is my data worth paying $4 - $20 a seat for the "insurance" over a free solution ?
B. Is my time and reduced performance impacts worth paying $4 - $20 a seat for over a free solution ?
C. If I am going to pay for an AV, is there any reason to choose a lesser product if they cost the same ?
Regarding C, there is ... for example, AV vendors have this habit whereby you might purchase the product at a discount for say 5 seats for $19.95 ... then at renewal time, it's now $59.95. Several options ...
a. Change vendors
b. Buy new instead of renewing
c. Wait a bit.... when ya don't renew, they tend to send out 'special' offers' to get yu back.
No different from any other PC related purchases.... if two GFX cards, cost the same and one performs better than the other in performance, power usage, temps whatever, is there a logical basis for not getting the one that performs better ... sometimes there is ... I have had users pick cards based upon color or brand loyalty, but most of the time, folks make a logical choice which provides the best performance / cost ratio for their particular usage. If you are an avid gamer, that's all that is on ya PC and time is not a consideration, one could argue that using any AV is not justified since the system could be wiped reloaded in a few hours. I have 30 years of financial and project records on my PC which is all backed up. But if my system was fudged from an infection... restoring all that is a several hours of my time. And that time is worth the cost of a top notch AV utility more than 50 times over.