Good morning to everyone,
(Sheet follows).
First of all, to say that I disagree with all the skirmishes in the forum, no reason. I don't know what has finally happened with the band or not (I'm talking about value and error), but I'm not looking for it much, I don't have the luxury of time, lately.
Starting with what was written there, I wanted to dig a little further into the 0% rake theme. Something didn't stick with all of this and I started looking for it in theory and also re-checking the code I had written and corrected 1-2 conditions and also rounded up more accurately the performances. The result was different and I will list below what happened.
I'll start with a question that nobody ever asked themselves, looking at the performance of the companies:
"Why do companies cut their pack to find the RIGHT returns?" Or at least as accurate as possible? Why hire staffs, statisticians, mathematicians, psychologists, economists. Why not give an odds of 5 up or 5 down and get the job done? And a group of players to gather, I think will give numbers close to the odds. Why;
Let's go a little bit at a simple but extreme example in mathematics to understand some things. The following yields are without rake.
Suppose we play singles in a bucket. Book A gives us 2,0-2,0 performance, mastery and good in math. Buccane B, who owns a rake but does a lot of math, gives us 1,50-3,00. Question: Which book will I play? And 2 without a rake.
In the second block I have 0,5 * 1,50 + 0,5 * 3,00 = 2,25 / 2 = 1,125, so + 12,5% average profit if I play at luck.
Here's another example: Buick A, the right one, gives us real 1,22-5,56 chances. That is, 82% -18%.
B comes again and tells us "I give 1,30-4,35". Irrelevant again? Yes, but to our detriment now:
0,82 * 1,30 + 0,18 * 4,35 = 1,85 / 2 = approx. 0,925. That is, the average gap -7,5%.
What do we see? That the buck, whenever he strays away from the real chances, he is very likely to be defeated or taken away by us.
Another example: Real odds 1,85-2,17, Book B gives 1,69-2,44. If we do the calculations, average profit + 1,7%.
If one looks at several examples, which I did, one would see that the closer the chances are to the 50-50, the lower the gain or the chuckle.
1 Conclusion: Buck wants a match that is close to 50-50 because it eliminates the possibility of a rash due to an error in estimation.
2 Conclusion According to 1, the player wants a long distance match in the odds because there he can take advantage of odds. In short, if I know that the XBook is underestimating or overestimating results, I choose 1,10-10 and not 1,50-3,00.
Someone will say, nice boy, but how do I know what to do? Should I play 10.000 bets on luck? Does it suit?
After fixing some things in the code, I ran it again and the result is in favor of the player and I'll explain the data.
I assume I'm playing under-over, I get a random value from 6-94 for over and the rest from 100 is under. Why from 6-94? To be able to underestimate or overestimate returns at random.
Then I raise or lower the actual 5 units, that is, if I have over 40%, then the company over will become 45%, which means that in percentages it would be 2,50 and 2,22. After I draw, random player selection and result: + 6% for player.
As I reduce the rate of overstatement or depreciation, the profit falls and ends up at 0, when the company's returns match the actual returns.
3 conclusion, companies want rake in to protect themselves from mistakes, obviously.
I was talking yesterday to someone who owns the sport as a mathematician and after I told him the whole experiment, he told me that yes I can win, but again this is not the right strategy. So he told me that the right strategy is that once you win in the big gaps, the strategy should be "I bet the highest odds" and not in the blind.
In fact I tried it and the profits almost doubled: + 11%.
Then I remembered an old man on the forum who said over and over again "how do they calculate bf and give a yield of 500? Odds of 1000? Why not 400 or 900 respectively?".
He was probably right then, because the great returns are the great values and the big mistakes.
I would like to say that these are my conclusions, in no way applicable since you cannot find thousands of games with 0% rake. Also the basic assumption that is made is that the stock is underestimated or overvalued at random with a probability of 50-50. If he does one of the two permanently, we will either become rich or go to prison.
I hope we all become a little wiser, get food for thought. I also don't want guns and am available to give code that I wrote for testing, maybe making mistakes. O @PyBet can read and write I guess, maybe other kids.
(Sheet follows).
First of all, to say that I disagree with all the skirmishes in the forum, no reason. I don't know what has finally happened with the band or not (I'm talking about value and error), but I'm not looking for it much, I don't have the luxury of time, lately.
Starting with what was written there, I wanted to dig a little further into the 0% rake theme. Something didn't stick with all of this and I started looking for it in theory and also re-checking the code I had written and corrected 1-2 conditions and also rounded up more accurately the performances. The result was different and I will list below what happened.
I'll start with a question that nobody ever asked themselves, looking at the performance of the companies:
"Why do companies cut their pack to find the RIGHT returns?" Or at least as accurate as possible? Why hire staffs, statisticians, mathematicians, psychologists, economists. Why not give an odds of 5 up or 5 down and get the job done? And a group of players to gather, I think will give numbers close to the odds. Why;
Let's go a little bit at a simple but extreme example in mathematics to understand some things. The following yields are without rake.
Suppose we play singles in a bucket. Book A gives us 2,0-2,0 performance, mastery and good in math. Buccane B, who owns a rake but does a lot of math, gives us 1,50-3,00. Question: Which book will I play? And 2 without a rake.
In the second block I have 0,5 * 1,50 + 0,5 * 3,00 = 2,25 / 2 = 1,125, so + 12,5% average profit if I play at luck.
Here's another example: Buick A, the right one, gives us real 1,22-5,56 chances. That is, 82% -18%.
B comes again and tells us "I give 1,30-4,35". Irrelevant again? Yes, but to our detriment now:
0,82 * 1,30 + 0,18 * 4,35 = 1,85 / 2 = approx. 0,925. That is, the average gap -7,5%.
What do we see? That the buck, whenever he strays away from the real chances, he is very likely to be defeated or taken away by us.
Another example: Real odds 1,85-2,17, Book B gives 1,69-2,44. If we do the calculations, average profit + 1,7%.
If one looks at several examples, which I did, one would see that the closer the chances are to the 50-50, the lower the gain or the chuckle.
1 Conclusion: Buck wants a match that is close to 50-50 because it eliminates the possibility of a rash due to an error in estimation.
2 Conclusion According to 1, the player wants a long distance match in the odds because there he can take advantage of odds. In short, if I know that the XBook is underestimating or overestimating results, I choose 1,10-10 and not 1,50-3,00.
Someone will say, nice boy, but how do I know what to do? Should I play 10.000 bets on luck? Does it suit?
After fixing some things in the code, I ran it again and the result is in favor of the player and I'll explain the data.
I assume I'm playing under-over, I get a random value from 6-94 for over and the rest from 100 is under. Why from 6-94? To be able to underestimate or overestimate returns at random.
Then I raise or lower the actual 5 units, that is, if I have over 40%, then the company over will become 45%, which means that in percentages it would be 2,50 and 2,22. After I draw, random player selection and result: + 6% for player.
As I reduce the rate of overstatement or depreciation, the profit falls and ends up at 0, when the company's returns match the actual returns.
3 conclusion, companies want rake in to protect themselves from mistakes, obviously.
I was talking yesterday to someone who owns the sport as a mathematician and after I told him the whole experiment, he told me that yes I can win, but again this is not the right strategy. So he told me that the right strategy is that once you win in the big gaps, the strategy should be "I bet the highest odds" and not in the blind.
In fact I tried it and the profits almost doubled: + 11%.
Then I remembered an old man on the forum who said over and over again "how do they calculate bf and give a yield of 500? Odds of 1000? Why not 400 or 900 respectively?".
He was probably right then, because the great returns are the great values and the big mistakes.
I would like to say that these are my conclusions, in no way applicable since you cannot find thousands of games with 0% rake. Also the basic assumption that is made is that the stock is underestimated or overvalued at random with a probability of 50-50. If he does one of the two permanently, we will either become rich or go to prison.
I hope we all become a little wiser, get food for thought. I also don't want guns and am available to give code that I wrote for testing, maybe making mistakes. O @PyBet can read and write I guess, maybe other kids.
