z2newton
z2newton
  • Threads: 7
  • Posts: 62
Joined: Mar 22, 2014
September 27th, 2017 at 7:10:24 PM permalink
Hopefully someone can answer this for me. Wizard?? The Wizards table uses data on games through 2012. How could I recreate this table if I had more up to date data? Example: The probability of winning after buying 1 point on a spread of -7 is 55 %. How is the 55 % calculated?
EdCollins
EdCollins
  • Threads: 20
  • Posts: 1739
Joined: Oct 21, 2011
September 27th, 2017 at 7:36:08 PM permalink
The percentage listed is most likely calculated by simple division.

Take all of the games in which the spread was exactly -7, during the time span you'd like to look at.

Add up the number of times in which the underdog won the game, or lost the game by five points or less.

(Also keep track of the number of games in which the underdog lost by exactly 6 points, as these bets now would have been considered pushes.)

And finally, divide the number of games in which the underdog won the game or lost by 5 points or less by the total number of games with this -7 pointspread and there's your "buying 1 point on a spread of 7" percentage.


And yes, you can recreate (and or update) the table yourself. All you need is the final score of each team for each game, and the initial pointspread of each game. Putting everything in a spreadsheet would be ideal.
Wizard
Administrator
Wizard
  • Threads: 1518
  • Posts: 27037
Joined: Oct 14, 2009
September 28th, 2017 at 12:08:43 PM permalink
As I recall, I looked at how often the half point helped if moving to or off of 7. So:

+6.5 to +7
+7 to +7.5
-7 to -6.5
-7.5 to -7

I should update my NFL stuff. Maybe after this season.
"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)
SM777
SM777
  • Threads: 6
  • Posts: 762
Joined: Apr 8, 2016
September 28th, 2017 at 1:35:53 PM permalink
Quote: z2newton

Hopefully someone can answer this for me. Wizard?? The Wizards table uses data on games through 2012. How could I recreate this table if I had more up to date data? Example: The probability of winning after buying 1 point on a spread of -7 is 55 %. How is the 55 % calculated?



That will not be a profitable venture long term. Laying -130 or higher (assuming each half point is 10¢) on an NFL game is a bad idea.
z2newton
z2newton
  • Threads: 7
  • Posts: 62
Joined: Mar 22, 2014
September 28th, 2017 at 3:32:53 PM permalink
Quote: Wizard

I should update my NFL stuff. Maybe after this season.



Yes please and a big thank-you.

I only picked -7 and 1 point as an example on how the probability is calculated.

What I would find interesting and the reason I was thinking of making a table myself is separating regular season and the playoffs. Also a table that's only the previous 10 years.
cyberbabble
cyberbabble
  • Threads: 2
  • Posts: 89
Joined: Mar 24, 2013
September 28th, 2017 at 5:12:09 PM permalink
I don't have any connection with these guys. I downloaded some of their data 3 - 4 years ago. I never had any indications of problems or shadiness. They seem to keep the data archives up to date. No guarantees though.

push chart or half point chart - this is what you are asking about

http://www.sportsbookreviewsonline.com/halfpointchart.htm

data files

http://www.sportsbookreviewsonline.com/scoresoddsarchives/scoresoddsarchives.htm
z2newton
z2newton
  • Threads: 7
  • Posts: 62
Joined: Mar 22, 2014
September 29th, 2017 at 11:15:31 AM permalink
Thanks cyberbabble. I think this what I'm looking for.
  • Jump to: