ProvablyFair.org

We conducted an independent technical audit of Winna​.com's Plinko game following their March 10 incident report. Our analysis shows that for nearly 3 months, players were playing Plinko on worse odds than the site publicly stated. If you played Winna Plinko between Dec 17 and March 10, your bets were placed under those probabilities. Here’s what the code shows. 🧵

Good question. The RNG didn't change, only the probability table it mapped to. So the same random number that gave you 0.2x under the flawed table might have given you 1000x under fair math, but that same logic applies to every other bet you placed during that period too. Some bets that won under the broken table might have lost under fair odds, and vice versa. To know your actual net impact, you'd need to replay all your bets against the correct probability distribution and compare the totals. Complicated process if you have thousands of bets with multiple seeds, but worthwhile if you played a lot or high stakes. The tool provided allows you to check bulk nonces (bets) per seed pair: provablyfair-org.github.io/winna-plinko/

@provablyfairorg Hi - according to your tool I missed a 1000x on december 30th, it paid 0,2X. Do I understand correctly if they didn't change their formula on december 17 I would have actually hit that 1000x instead of a 0,2x? Or is my logic flawwed?

We conducted an independent technical audit of Winna​.com's Plinko game following their March 10 incident report. Our analysis shows that for nearly 3 months, players were playing Plinko on worse odds than the site publicly stated. If you played Winna Plinko between Dec 17 and March 10, your bets were placed under those probabilities. Here’s what the code shows. 🧵

A player reached out to us in November with concerns. Because the client seed was casino chosen at the time (not real PF) we thought it was worth investigating, we ran 67.5M simulations across 25 real seed pairs and all 27 configurations converged on ~99% RTP - see post 3. It was clean then. The change came with the Dec 17 update. We didn't pick this up until Feb when we retested it again.

@provablyfairorg @ProfaneLiving still curious though who actually first flagged this publicly?

@ninobyny7 @ProfaneLiving Thats what Im saying. We investigated in November it was fine, we then tested it again in Feb and found the code had been changed. You can check our GitHub to see the the dataset collected in feb.

@provablyfairorg @ProfaneLiving It only started Dec 17, you said November was fine with expected RTP. Why skip that part? And where's the credit for the guy who actually found it first?

Good question. Our analysis focused specifically on Plinko. We first looked at it in November and confirmed the math was sound at that time - the client seed issue was noted and it wouldn't have passed what we consider PF but the seeds were producing expected RTP over 67.5M simulations. The probability table change came later with the Dec 17 update. We haven't done a full audit of any of their other games so can't answer that specifically.

@provablyfairorg excellent work, a more broad question tho.. did you find their originals to be provably fair, 100% ? until this last week, a player had no input over the client seed... which i have never seen before and always made me suspect something fishy

@explorer_q72756 We spoke with several affected players during our research and appreciate everyone who raised concerns early. Some preferred to stay anonymous, but happy to credit anyone publicly. Please tag them here

@provablyfairorg why you haven't give the credit to the guy who did all the research about it you dont even mention even once who did the research and how you get to know about the issue ...its good you enquire tooo but give the real guy credit who did all the study

@rpcrevo You're right, several players helped with information early on. Happy to give credit if they want to be known publicly - can you tag them here?

@provablyfairorg @provablyfairorg great info, but why did you guys choose not to credit the original author's work? A lot of effort and persistence went into it, and he deserves some recognition for exposing such blatant avarice and shameful behavior in an industry that's already rife with both

MARCH 14 We extracted the live code again. Low, Medium and High modes now use fair binomial coin-flip logic, restoring ~99% RTP. If you played Plinko between Dec 17 and March 10, we recommend contacting support to understand how your play was reviewed. This case is also a good example of why independent third-party verification matters when casinos build and deploy their own games. Github: github.com/ProvablyFair-o…

VERIFICATION TOOL We’re publishing a tool where affected players can input their seed pairs and see which bets would have produced different outcomes under fair binomial probabilities vs the extracted tables. If you played Winna Plinko during the affected period, you can check your results here: 🔗 provablyfair-org.github.io/winna-plinko/

Common Provably Fair marketing claims that most players will see and just believe because it sounds technical.