Anticheat for us is hard, like really hard
I'm fairly sure you've heard the overview of why anticheat development for us is more difficult than other networks:
- We support so many different versions: Sentinel runs across three networks: 1.8, 1.9+ and Bedrock. Each of these versions have their own set of behaviours and paramaters and their own "quirks". On 1.9 we are having to cross some fairly major mechanic switches especially with regards to swimming and flying as we support 1.9 through 1.15. This effectively triples how hard it is for us to get a really good anticheat across the network.
- There are cheat developers out there who focus on CubeCraft, it's a continuing and tiring game of cat and mouse. Ultimately constantly patching up bypasses as we see them out in the wild means we aren't spending time on more complex and complete detections and a balance has to be made
Ultimately we need more people working on Sentinel but the skill level required to be effective and the length of time before a developer fully understands the Sentinel codebase means that this is a costly process and one which we can't justify without a more solid strategy for anticheat.
I wish there was an easier answer here but I'm afraid I don't have one. Sentinel is a continuing topic of discussion within the management and moderation teams. For those keen eyes on our staff list you'll notice we have a new main developer for Sentinel come onboard recently (Mats). They've got some pretty cool stuff in prototype and hopefully we'll see a lot of that rolling across the network over the coming months.
I feel I should come to Sentinel's defense somewhat - It does catch and block a lot of cheaters (thousands every week) but I share your frustration that seemingly obvious cheats are obvious on the server and we don't do enough to show it's successes.