Me too. Surprisingly, 6842Z has the lowest predicted score of these by far. (around 21, where the other three are between 30 and 40)
So, as I said earlier, I had stopped development. That was before I was stuck in quarantine 
. Two things changed which encouraged me to pick this back up:
- I got an email from discord about the bot last night, informing me it was in far more servers than I’d realised, and,
- I have nothing else to do because I’m stuck indoors
If all went to plan this morning, the bot should be back up and running with the latest data (@Sylvie if it’s not, lmk and I’ll see what’s up) and with the sudden amount of spare time I now have I’m open to suggestions for all sorts of weird and wonderful things I could add to kill time to make the bot cooler. The bot invite link should still be the same, but I’m gonna edit the OP and fix up the now-broken formatting.
Has VEX contacted you asking to use it for Virtual Worlds?
Nope 
. As I said, it’s just because I now have a sudden amount of additional free-time I wasn’t expecting to have, and the fact it’s in far more servers than I’d realised made me think “gee I probably should fix this up”
So, you haven’t a clue why DRow unlocked this?
In that case
@DRow please don’t revive old threads
/s pls don’t ban me
The reason he unlocked it was because I asked him if it were better to make a new thread or unlock this one and he said unlock this one
I see… we’ve been fooled!!!
So how do I get the match simulation to run. Want to do some matches with teams in my organization.
Currently, you will have to run each match one at a time using =predict 6969A,6969B 420C,420D, to give an example, but I’m working on a mode that allows you to pass a whole list of teams (or an SKU) and the bot will work out the matches that need simulated, and simulate them all properly. I say “currently”, but I mean “in about 5 minutes” because currently the bot is rebuilding its internal cache of all match data, and during that time it locks out predictions because, well, it can’t really predict results without actually having data to predict with. (If you run a command, you should get a message with a eta)
Didn’t this bot install spyware on servers that tracked messages being sent and servers, resulting in many photos and videos of turning point robots getting leaked? I remember there being discussion of that from last year.
Haha nothing of the sort. For a short period of time (on request) it counted messages set in the UK vex server, but that was disabled ages ago. I don’t remember any such discussions, so I suspect you might be confusing this bot with a different one. If you’re curious, the source is all public at https://github.com/Bottersnike/VexBot
Robotevents: Allow me to just… creates discord bot that messages this discord bot and uses the responses for their “fantasy robotics”
Somehow the bot has said that my team is the best in my organization.
Also, I’m very surprised that 2616J is #2 in the global leaderboard. We’re from the same state (NJ) and they’ve been to all of the competitions I have. They’re a very good team, but they’re certainly not the second best in the world. They might not even be best in our state.
Maybe our state is just bad.
So I’ve just ran some numbers, and the algorithm that last year was getting ~90% accuracy, this year was sitting at a toasty 44% accuracy (that’s right, worse than random). With that in mind, I’ve rewritten the entire predictions algorithm to one with the highest accuracy I could get, which also means it’s now able to guess the score outcomes, too. Vague details are in =faq, but as ever the source is the best place to check. The leaderboard should hopefully be a tad more sensical now as well.
Just make it say the opposite of what it thinks and your at 56, above average baby
Yeah I see why now.
3217M is consistently putting up incredibly high scores… but keep in mind that 8349E helped in about half these matches (including one of the qual matches), and their opponents were consistantly putting up low scores… and low scores can only mean A) facing clawbots/jankbots every match or B) opponents dropping stacks. So no doubt 3217M is an amazing team, but there is also some “strength of schedule” involved if you know what I’m saying.
Their partners every match:
3217A
323V
32160D
8349F
3259A
8349E(x5) (A little context, 8349E is 19th in the world for skills)
I don’t know about a couple of the teams, but as far as I’m aware all of these were ideal partners for 3217M.
323V is pretty good, and would do well with a good team. I don’t know about other teams, but getting more than 80 points in one game is REALLY good. Not many teams in my state get that, and if they do, they are 6842Z/similar class team with a really good partner against a decent alliance. I can see why the bot chose it.
Edit: This was completely off topic. I am basically repeating what you said but more enthusiastically.
Guess this didn’t age well hmm?? 
Not at all
Of course it’s worth mentioning that 2616J has been to god knows how many tournaments and therefore has better top 15 scores
I’m curious as to who you rank above them in the state?