Quote:
Originally Posted by tombos21
To be clear, there were 8 hands posted in the discord. I reran them through the upgraded Fair Play tool:
1 solid match.
7 false positives (different positions, stacks, format, etc).
This seems like a shitshow then. If this guy's not actually guilty, a lot of people have screwed up pretty badly here, including GTOW for releasing a half baked product that's driving a lot of FUD about RTA in the poker community.
Has GTOW done any testing to ballpark how likely false positives are? Seems like it should be pretty easy, just pick some representative timestamps (not random ones but ones that correlate to GTOW activity) and random boards and run them through the checker. Do this like 10k times and see how many hits you get.
In the other thread in the forum a couple of the hands had timestamps and they didn't seem anywhere close to matching the GTOw lookups. That should be step #1 for interpreting these results. Even the HH posted in this thread doesn't actually contain timestamps.
edit: another really important point for detecting false positives is being aware of how many hands you look up. If you check 10 hands and get 5 positives, that's very different from getting 5 positives out of 100 hands. Using a tool like this actually requires using your brain a bit and I hope the NJ regs have been doing that before publicly accusing this guy.