fw
Member
- Joined
- Aug 1, 2007
- Messages
- 120
Hey guys,
at Aachen Open 2011, we used the stackmat timer to generate random numbers between 0 and 99 (last two digits) to decide who competes in our mystery event. Someone pointed out, that this might not be a uniform random distribution.
I tried to empirically test that thesis by having a look at results in the WCA database. For now, I only used 3x3 results (all, not just best singles or averages), removed all but the last two digits and counted the occurences. I got the following:
[[0, 5061], [1, 1070], [2, 4543], [3, 4672], [4, 485], [5, 4525], [6, 4735], [7, 446], [8, 4515], [9, 4648], [10, 435], [11, 4554], [12, 588], [13, 4473], [14, 423], [15, 4602], [16, 4446], [17, 424], [18, 4872], [19, 4538], [20, 429], [21, 4655], [22, 4444], [23, 410], [24, 453], [25, 4449], [26, 422], [27, 4351], [28, 4741], [29, 440], [30, 4480], [31, 4660], [32, 443], [33, 4615], [34, 4783], [35, 502], [36, 4639], [37, 685], [38, 4395], [39, 565], [40, 4900], [41, 4445], [42, 500], [43, 4700], [44, 4515], [45, 483], [46, 4701], [47, 4485], [48, 459], [49, 564], [50, 4483], [51, 447], [52, 4456], [53, 4841], [54, 515], [55, 4445], [56, 4544], [57, 429], [58, 4483], [59, 4662], [60, 510], [61, 4335], [62, 632], [63, 4423], [64, 455], [65, 4643], [66, 4531], [67, 406], [68, 4799], [69, 4465], [70, 414], [71, 4700], [72, 4530], [73, 442], [74, 489], [75, 4525], [76, 407], [77, 4373], [78, 4648], [79, 418], [80, 4591], [81, 4754], [82, 415], [83, 4526], [84, 4582], [85, 425], [86, 4582], [87, 627], [88, 4510], [89, 391], [90, 4796], [91, 4492], [92, 402], [93, 4772], [94, 4528], [95, 428], [96, 4675], [97, 4446], [98, 441], [99, 518]]
It appears that every third or fourth number is significantly less frequent than the others (which seem to be more or less equally frequent). Does anybody have an explanation for this?
Thanks
Flo
at Aachen Open 2011, we used the stackmat timer to generate random numbers between 0 and 99 (last two digits) to decide who competes in our mystery event. Someone pointed out, that this might not be a uniform random distribution.
I tried to empirically test that thesis by having a look at results in the WCA database. For now, I only used 3x3 results (all, not just best singles or averages), removed all but the last two digits and counted the occurences. I got the following:
[[0, 5061], [1, 1070], [2, 4543], [3, 4672], [4, 485], [5, 4525], [6, 4735], [7, 446], [8, 4515], [9, 4648], [10, 435], [11, 4554], [12, 588], [13, 4473], [14, 423], [15, 4602], [16, 4446], [17, 424], [18, 4872], [19, 4538], [20, 429], [21, 4655], [22, 4444], [23, 410], [24, 453], [25, 4449], [26, 422], [27, 4351], [28, 4741], [29, 440], [30, 4480], [31, 4660], [32, 443], [33, 4615], [34, 4783], [35, 502], [36, 4639], [37, 685], [38, 4395], [39, 565], [40, 4900], [41, 4445], [42, 500], [43, 4700], [44, 4515], [45, 483], [46, 4701], [47, 4485], [48, 459], [49, 564], [50, 4483], [51, 447], [52, 4456], [53, 4841], [54, 515], [55, 4445], [56, 4544], [57, 429], [58, 4483], [59, 4662], [60, 510], [61, 4335], [62, 632], [63, 4423], [64, 455], [65, 4643], [66, 4531], [67, 406], [68, 4799], [69, 4465], [70, 414], [71, 4700], [72, 4530], [73, 442], [74, 489], [75, 4525], [76, 407], [77, 4373], [78, 4648], [79, 418], [80, 4591], [81, 4754], [82, 415], [83, 4526], [84, 4582], [85, 425], [86, 4582], [87, 627], [88, 4510], [89, 391], [90, 4796], [91, 4492], [92, 402], [93, 4772], [94, 4528], [95, 428], [96, 4675], [97, 4446], [98, 441], [99, 518]]
It appears that every third or fourth number is significantly less frequent than the others (which seem to be more or less equally frequent). Does anybody have an explanation for this?
Thanks
Flo
Last edited:
