To my extreme shock, I had a f2p account that botted mining for three straight days and lived to tell the tale. I assumed it would be banned and so I completely abandoned it, but I checked today and it's still around... Keep in mind that this is a week later or so, meaning the account is probably in the clear.

This was done using my own home-made script. The script mined tin at Barbarian Village and deposited it in the Varrock west bank. I believe the longevity of the account comes from a mistake in the script. Tin rocks have two separate ID's for some reason. I was aware of this, as all ores have 2+ ID's, but I forgot to add the second ID to the script, resulting in very quirky behavior compared to other mining scripts. 

This leads me to believe it is uniqueness, not human-like behavior, that prevents bans.

My theory is Jagex collects behavior profiles over time. Profiles can then be analyzed and compared to other profiles to determine whether or not they come from a bot. However, if a unique profile appears from nowhere (new script), then Jagex/Jagex's system has nothing to compare it to, and it goes undetected. 

Of course, most common methods (woodcutting, mining, fishing) don't leave much room for deviation. All bots are going to behave in basically the exact same way, making private scripts less useful. However, if you intentionally (or unintentionally like in my case) make the bot make a mistake or behave in a quirky, inefficient way, you get a lower ban rate. Had I not made this mistake with my script, I don't think the account would have survived.

Edited March 6, 20187 yr by SqueezeTheorem

7 minutes ago, SqueezeTheorem said:

This leads me to believe it is uniqueness, not human-like behavior, that prevents bans.

My theory is Jagex collects behavior profiles over time. Profiles can then be analyzed and compared to other profiles to determine whether or not they come from a bot. However, if a unique profile appears from nowhere (new script), then Jagex/Jagex's system has nothing to compare it to, and it goes undetected. 

Of course, most common methods (woodcutting, mining, fishing) don't leave much room for deviation. All bots are going to behave in basically the exact same way, making private scripts less useful. However, if you intentionally (or unintentionally like in my case) make the bot make a mistake or behave in a quirky, inefficient way, you get a lower ban rate. Had I not made this mistake with my script, I don't think the account would have survived.

This is an interesting theory and would account for why private scripts tend to have lower ban rates. 

You mentioned in your first post that anti-ban is "just a marketing scheme" and, for the most part, I would agree. However I do think it can be used as you've described above. When I used to make simple bank standing scripts a few years ago I would include unneeded behaviour in the script just to try and mimicked some of my "ticks" from normal gameplay. One example I can think of are mouse movements towards an object before the need to click on that object; sometimes I find myself pre-hovering the next tree I would cut, sometimes I do not. I am unsure how much data Jagex collects on in-game behaviour, although in this world of big data I am sure it is a lot, but including these quirks in the form of a unique anti-ban does seem likely to help. The problem arises when this unique anti-ban is used by a mass of people; suddenly everyone is acting in a particular way and detection, based on the analysis you described above, skyrockets.

22 minutes ago, Tw3nty said:

This is an interesting theory and would account for why private scripts tend to have lower ban rates. 

You mentioned in your first post that anti-ban is "just a marketing scheme" and, for the most part, I would agree. However I do think it can be used as you've described above. When I used to make simple bank standing scripts a few years ago I would include unneeded behaviour in the script just to try and mimicked some of my "ticks" from normal gameplay. One example I can think of are mouse movements towards an object before the need to click on that object; sometimes I find myself pre-hovering the next tree I would cut, sometimes I do not. I am unsure how much data Jagex collects on in-game behaviour, although in this world of big data I am sure it is a lot, but including these quirks in the form of a unique anti-ban does seem likely to help. The problem arises when this unique anti-ban is used by a mass of people; suddenly everyone is acting in a particular way and detection, based on the analysis you described above, skyrockets.

Bank standing scripts are another interesting example. I suicide botted an account (that I still have access to) to ~80 fletching going 12 hours per day. I can't remember exactly how long this took, but botting pretty much anything else would have probably resulted in a ban. 

Just the other day, I saw an account at the GE that had literally 0 xp in all skills, except fletching, where he had ~59m xp.

So why fletching? Why does it have a low ban rate, and why are fletching bots somewhat notorious for not getting banned? Under my theory, there really isn't a lot of deviation between human and bot behavior when it comes to fletching. There's some, but not much. You bank, click the knife, click the logs, and repeat. And so comparing behavior profiles in this case really wouldn't tell you much since they all basically look the same. 

Compare this to something like mining or woodcutting which is very click intensive and can be performed in a number of different ways (all the possible pathing possibilities, afk time, etc) and I'm sure bot profiles stick out like a sore thumb. 
 

EDIT:
I found this old thread on another botting website with these images attached:

This shows the walking path of one of its most popular RC scripts at the time and the ban heat map as released by Jagex. There are very few bans outside of that path. This could be because less people used other scripts at the time, or it could be because the other scripts that were used were less popular, and therefore their behavior couldn't be profiled. Just food for thought

Edited March 6, 20187 yr by SqueezeTheorem

I love threads like this, best of luck to you!

Good luck man, Nice thread!

Good luck bro! If you need fresh accounts, let me know!

Got 10 bots running right now, wish me luck lol

19 minutes ago, SqueezeTheorem said:

Got 10 bots running right now, wish me luck lol

Flax spinners? Good luck!

30 minutes ago, dazeldo said:

Flax spinners? Good luck!

I wish haha. Still f2p lobsters

Close to a bond so I'm just gonna finish up the cash I need for one

 

Edited March 10, 20187 yr by SqueezeTheorem

As mentioned in the thread previously, i do think they've cut down on suiciding weekends massively

5 hours ago, SqueezeTheorem said:

I wish haha. Still f2p lobsters

Close to a bond so I'm just gonna finish up the cash I need for one

 

Can i ask what GUI u use for linux? i liked it tbh

1 hour ago, progamerz said:

Can i ask what GUI u use for linux? i liked it tbh

I'm using Gnome. Think it's the one that comes with Ubuntu 16 now that they've moved away from Unity. If not, I'm using Ubuntu Gnome. Been a while since I installed so I can't remember exactly.

Edited March 10, 20187 yr by SqueezeTheorem

Made it to the 18 hour mark. Muled 1.5M. Total cash now 4m. Considering doing fletching instead of flax for my first p2p bot since the banrate is much, much lower. Gonna let the remaining 10 bots go until they're banned. 

10 more hours and I can afford 2 bonds instead of 1

Edited March 11, 20187 yr by SqueezeTheorem

2 hours ago, SqueezeTheorem said:

Made it to the 18 hour mark. Muled 1.5M. Total cash now 4m. Considering doing fletching instead of flax for my first p2p bot since the banrate is much, much lower. Gonna let the remaining 10 bots go until they're banned. 

10 more hours and I can afford 2 bonds instead of 1

Good luck on fetching hope it works out for you.

R u using private/self written fishing script? Gl

Edited March 11, 20187 yr by z10n