From: lotteryprediction@gmail.com   
      
   在 2022年1月7日星期五 UTC+8 20:36:36， 写道：   
   > 王乾懿 Wang Qianyi (Nan) wrote:    
   > >   
   > > Note: Given the second round of Nigel's doubts about the revised    
   > > data' fairness, it would be preferable if everyone posted their own    
   > > numbers as late as possible, or tried to verify that they will not be    
   > > modified at random.   
   > You've borrowed from human selections previously and nobody seemed too    
   > cut up about it. I'd like to see you keep one of your sets of selections    
   > (say Nan) as personal choice and be open about how your AI system    
   > selections are generated if they borrow from other entrants.    
   >    
   > Good luck,    
   >    
   > Evil Nigel   
   This year, I did not use other forecasters' numbers and make my predictions,   
   as evidenced by the fact that I always posted my predictions ahead of everyone   
   else.   
      
   Although I have previously scored 100 points by combining the core numbers of   
   other contestants, and that score is nearly unbreakable. However, I don't   
   believe this improves the performance of my AI. The main reason for this is   
   not out of sportsmanship,    
   but rather because I believe that human predictions are not the best screening   
   numbers.   
      
   My AI predictions were derived by filtering random data from RANDOM.ORG, and   
   anyone could have easily copied my method and excelled. I'd have the   
   random.org site produce 100 sets of 24 numbers, and if I found any   
   same numbers in the 15th through 17th    
   digits from the the last drawing's winning numbers, that set of 24 numbers   
   would be included in the selection.   
      
   Try it for yourself if you don't trust me.   
      
   Nan   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)