What is the statistical probability of two User-Agent values ​​matching?

Question

Browsers

Interests actually a subject. Maybe someone conducted "research" in this matter, or found such data somewhere. Google says it doesn't know about this... Or maybe I didn't ask correctly...
UPD. No one imitates anything, they just go to the site of 100 unique organisms, for example, everyone has a conscience like a diamond. What is the statistical probability / ratio (1:1000 for example or how to explain correctly) that two unfamiliar and unrelated visitors will have the same user-agent, which is generated by the client browser.
About fakes, etc. don't forget - conscience is like a diamond! :)
UPD #2. Look at my profile, do I look like someone who would confuse user-agent with cookies and session?))
I know perfectly well what data the user-agent contains, I know that it can be emulated even in the form of S. Marshak's work "They are looking for firefighters, the police are looking for ...")))
I just want to know the probability of matching user-agents for two site visitors under conditions when everything is in default , who, having a conscience like a diamond, still spend all their free time from Internet surfing in church and pray to all the gods in a row.
UPD #3. But what happened to people))) The browser and its version is 1/fuck_know_what from all the information that is contained in the user-agent.
For example mine: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.10
From which you can see, in addition to the browser version, a hell of a cloud of information!
So I need statistical data on the probability that this line in italics will be the same for two visitors, which means that all factors must match: OS, OS version, OS bit depth, and browser version, engine and something else .... And on the phones, the brand and model are still in addition!
UPD #4. Here is a look at this data https://developers.whatismybrowser.com/useragents/...and this is only a global grouping by one of the factors, and now imagine that each of this is still divided by the same number, and then the result that remains in the final group (let it be the arithm mean.) is also presented as a possible probability of meeting on 1 one site with attendance, let's say ~ 500 people / day. At the same time, the site is regional (Kyiv), and we also cut off a damn cloud of those who are from Guadeloupe, etc. As a result, the probability of a coincidence is reduced to a minuscule, even in principle to zero ... Well, this is purely in the mind, not counting))
Am I right in thinking, or did I make a mistake somewhere?
UPD #5. I will analyze this data myself. This will be the easiest way and the reliability of the data will be just in my segment, so to speak.
The only thing, tell me how to "tag" a visitor better, so as not to re-fix his ua in the database? Cookies or localStorage? Maybe something else to take into account ... Xs.

Answer 1

Please clarify what you mean.
That two different users will go to the same browser?
That two different browsers will send the same User-Agent?
That someone's wget will deliberately imitate another browser's User-Agent?
Update:
User-Agent is a text string that is part of the HTTP request, starting with User-agent: or User-Agent:, and usually including information such as the name and version of the application, the computer's operating system, and the language.
That is, it's just the browser version, OS version, and current language. The probability of matching for popular platforms is very high, for unpopular ones it is lower.
Maybe you are confusing cookies and session?
In general, take a look here:
https://developers.whatismybrowser.com/useragents/

Answer 2

User-Agent не уникален. Большое количество людей может быть с одинаковым User-Agent.

Answer 3

1/5-1/20 примерно.
Браузеры сейчас как правило оперативно обновляются - версия одна.
Распространенных ОС тоже не много.