United States
Member #68,992
January 3, 2009
2,879 Posts
Offline
Crow,
Congrats on your mid-day win.
WTG on getting the NCEL moolah.
I have a question, if you don't mind. Some players post a host of pairs, but what is perplexing to me, is how are these pairs derived? Is it based on past matches and affiliated followers based on the historical record or trends?
United States
Member #69,528
January 11, 2009
7,803 Posts
Offline
Quote: Originally posted by sonnet on Sep 15, 2011
Crow,
Congrats on your mid-day win.
WTG on getting the NCEL moolah.
I have a question, if you don't mind. Some players post a host of pairs, but what is perplexing to me, is how are these pairs derived? Is it based on past matches and affiliated followers based on the historical record or trends?
/s
Are you saying that you want me to give you a lengthy binomial/hypergeometric distribution modeling algorithm as a means of accuracy?
I discovered some time back that those algorithms were very difficult to model based on missing data....which is why i complain a lot about "missing data".
If you look back at some of my posts regarding quantum entanglement/modeling you'll see i caught a lot of helll for some of my ideas which is why i decided to keep my mouth shut and instead work on fitting in with my attempts at being a comic...
in a way i think its just a lot easier and i got tired of arguing all the time about theory anyway.
sorry for my sarcasm....but i just cant seem to kick the comic relief symptom/syndrome...
United States
Member #68,992
January 3, 2009
2,879 Posts
Offline
Oh hell no! I understand the diffculties of the alogrithms and the consquent programming issues. Not possible because of missing data and imputing is a waste ot time. Was just wondering whether it is a gut relationship based on past, following drawn #s.
For example, for tonite, I have the following deducted plays:
335, 535, 538
outliers=492, 890, 708
long shot=328, if not 380 in the mix.
chances are 50/50 no clear absolute chance in the prediction---just a guest-estimate
United States
Member #69,528
January 11, 2009
7,803 Posts
Offline
Quote: Originally posted by sonnet on Sep 15, 2011
Crow,
Congrats on your mid-day win.
WTG on getting the NCEL moolah.
I have a question, if you don't mind. Some players post a host of pairs, but what is perplexing to me, is how are these pairs derived? Is it based on past matches and affiliated followers based on the historical record or trends?
/s
Sorry for my last post... i'll try to answer your question more directly.
Pairs are very hard to predict for especially when you try to narrow pairs down to as little as 2 or 3 pairs...
Because there is missing data in pre-draws this makes things even more difficult given that no one really knows what is missing and what is not.
Based on just pure statistics given pairs have to hit so the real trick is in trying to figure out what structure is "actually" missing. It seems that we are lacking 4/5's of the data needed to achieve a "somewhat" uniform understanding of the game ....it is this reason why it is hard to compile the correct algorithm based on any known percentage value.
Here are the odd and even pairs that hit in all combination configurations in pick 3 unmatched:
13x-15x-17x-19x-35x-37x-39x-57x-59x-79x
02x-04x-06x-08x-24x-26x-28x-46x-48x-68x
Let say for example that the digits 5-7 haven't been seen together in 80 draws then the 4/5's absentee rule can apply.
combinations played would be ---> 570-571-572-573-574-575-576-577-578-579
United States
Member #68,992
January 3, 2009
2,879 Posts
Offline
Quote: Originally posted by crow on Sep 15, 2011
Sorry for my last post... i'll try to answer your question more directly.
Pairs are very hard to predict for especially when you try to narrow pairs down to as little as 2 or 3 pairs...
Because there is missing data in pre-draws this makes things even more difficult given that no one really knows what is missing and what is not.
Based on just pure statistics given pairs have to hit so the real trick is in trying to figure out what structure is "actually" missing. It seems that we are lacking 4/5's of the data needed to achieve a "somewhat" uniform understanding of the game ....it is this reason why it is hard to compile the correct algorithm based on any known percentage value.
Here are the odd and even pairs that hit in all combination configurations in pick 3 unmatched:
13x-15x-17x-19x-35x-37x-39x-57x-59x-79x
02x-04x-06x-08x-24x-26x-28x-46x-48x-68x
Let say for example that the digits 5-7 haven't been seen together in 80 draws then the 4/5's absentee rule can apply.
combinations played would be ---> 570-571-572-573-574-575-576-577-578-579
Crow,
No need to be sorry. Discussion is good, because it helps to clarify some of the cobwebs, I may have had. Your explanation helps me to understand how the pairs are derived, and also will in the future help me to sharpen my focus and understand what are the missing keys.
For example, I suspected that the "62" was coming in, based on gut instinct, but did not play the entire set (0, 1, 2, 3, 4, 5, 6, 7, 8, 9) and therefore lost. Therefore, I was just inquiring, on how one can deduce the probablepairs with a "clinical/statistical eye", instead of gut feeling.