Trends and performance analysis

[Ron]

Hello Alex

Thank you for a great piece of software.

I have a couple of questions:

1. In the guide you write:
"Some cells may have trend hints. They are coloured in a violet (blue on LCD monitors), and indicate the odds that are not recommended to bet on due to either unfavourable trends or poor statistics."
Could you please elaborate a bit on this? Is there any way for us to know if the blue color is either because of bad trends or poor statistics? Is it a bad idea to bet on a team with no color when the oposing team has a blue color?

2. When there is a lot of matches in a specific league (for example "ATP-Singles") I am not able to scroll down to see the matches at the buttom of the list.

3. Performance Analysis: I have read the guide several times, but still have a hard time fully understanding it and how to use the information given. Could you please explain the "success rate" and "entropy of successful probs" in a simple and easy to understand kind of way.

I hope you will help me with these questions, I believe the answers would benefit other newbies as well.

Thank you
Ron

[Alex]

1. In case of poor statistics the whole string of cells (1,X,2,1X,12,2X,etc.) is colored in violet (blue).
It is not bad idea to bet on a team whose opposite team is colored in violet (blue). Moreover, it is a good idea to bet if the odds meet criterion Fair < Line.

2. Use a splitter control to scroll over the odds table: along with paging over the long odds lists (> 24) using Fixtures > Page next fixtures.

3. Both Success rate and the entropy have certain definitions described in the manual. Both these values are intended to be used as an indicators of the league predictability. Otherwise they can be used to compare different prediction algorithms. These values have been published in order to allow users to see criteria being used by developers in their efforts to improve prediction strength. Also users may manipulate by depreciation period values (default = 900 days in most leagues) and see how success rates / entropy change when fine-tuning custom (user built) leagues. And finally we've invited all our competitors to a permanent contest among algorithms which naturally requires certain values to be used as criteria of prediction strength.

[Ron]

Thank you Alex, it was very helpful.

[StevenDinh]

Dear Alex,

Sorry for inconvenience because of oldest topic. But I have just seen the sample video about stake wizard 4 and you constantly pick the team that have colored cells for choosing materials. Contrary to your advice in this topic.

Thank you for your explanation.

[Alex]

That's right. I have picked the cells colored in green that meet criterion Fair < Line odds.
You were advised not to pick the cells colored in violet (blue). There is no contradiction.

[StevenDinh]

Thank you for your great explanation, Alex.