What do people usually use to detect proxies? Or which proxies are more accurate
Do you mean proxy software or the IPS themselves?
Proxies are usually detected by 2 methods:
One is your proxy software itself leaking the fact that it's a proxy.
By default, most proxy servers out there send a
X-Forwarded-To header to the server you're making the request to.
So imagine you bought a proxy ip from a provider who's proxy server is not configured to prevent this, and you try to access Instagram using this proxy. Instagram can know that you are using a proxy, because your request contains that header and they can easily check for that.
Another method is through analysing traffic from an IP address. Most proxy providers will not give you a fixed IP address, rather you get an IP from a pool of addresses which is shared among other users of the service when you are inactive.
So say you get the IP address
a.b.c.d from a proxy provider. It doesn't leak anything, and you access instagram from your mac all well and good.
After 2 hours you disconnect from the proxy. Now this same address will be used by another user on windows, with a different TLS cipher, timezone and other such request fingerprints.
Instagram logs this information. And figures out that the same IP can't have such wildly different fingerprints and concludes it's a proxy.
Tho this method is complicated and to my knowledge, most sites don't use this method. Not even Instagram, atleast they don't block a page when trying to access though a proxy. This was just a hypothetical example.