1. #1
    antonyp22
    antonyp22's Avatar Become A Pro!
    Join Date: 01-12-14
    Posts: 78
    Betpoints: 2528

    Opening lines or closing lines to test a model?

    I've tested a model over a season and it has gone 74.5% ATS on opening lines and 56% ATS on closing lines. Which is the more reliable indicator?

  2. #2
    a4u2fear
    TEASE IT
    a4u2fear's Avatar SBR PRO
    Join Date: 01-29-10
    Posts: 8,147
    Betpoints: 35447

    I've had similar results and not received a clear answer from anyone. Rule of thumb is, if you are beating the closing line you will be profitable, and if your system is the real deal, it will "move the line" and you will beat the closing line anyway.

    i have a local and only get lines 1 hr pregame so my model testing well vs closers is good for me
    Nomination(s):
    This post was nominated 1 time . To view the nominated thread please click here. People who nominated: antonyp22

  3. #3
    EXhoosier10
    EXhoosier10's Avatar Become A Pro!
    Join Date: 07-06-09
    Posts: 3,122
    Betpoints: 4390

    Quote Originally Posted by antonyp22 View Post
    I've tested a model over a season and it has gone 74.5% ATS on opening lines and 56% ATS on closing lines. Which is the more reliable indicator?
    With closing lines, there is more information available than at the opener. Unless you think Vegas is just fvcking with bettors and moving the lines at will, the closer is more reliable.
    That being said, your goal is to get the best odds available for the side you want to bet; if that means betting the opener, so be it.
    Nomination(s):
    This post was nominated 1 time . To view the nominated thread please click here. People who nominated: antonyp22

  4. #4
    HeeeHAWWWW
    HeeeHAWWWW's Avatar Become A Pro!
    Join Date: 06-13-08
    Posts: 5,487
    Betpoints: 578

    There are practical problems with using openers in a backtest:

    1) reduced limits. How often is the stake you want to play above them?

    2) the time of day they are released. Is it consistent/predictable? If you can't actually place the bet at that price, it's a misleading backtest to use these.

    3) how long after release before the opener line/price typically changes? If it's very quick, again, rather misleading cos you won't get that price.


    Pretty much any decently liquid market you will do better against the openers.

    If you can get the data, try using the 3rd or 4th price, or the one that's available 2 hours after the opener- these are probably more attainable with real betting. If not, just average the opener and the closer to guesstimate some sort of middle market point.
    Last edited by HeeeHAWWWW; 02-20-14 at 08:35 AM.

  5. #5
    Sawyer
    Sawyer's Avatar Become A Pro!
    Join Date: 06-01-09
    Posts: 7,592
    Betpoints: 6650

    Closing line of course.

    74.5% ATS on opening lines and 56% ATS on closer is a good sign. Make sure you bet early.
    Last edited by Sawyer; 02-20-14 at 09:39 AM.

  6. #6
    gui_m_p
    gui_m_p's Avatar Become A Pro!
    Join Date: 09-18-13
    Posts: 123
    Betpoints: 1537

    Quote Originally Posted by antonyp22 View Post
    I've tested a model over a season and it has gone 74.5% ATS on opening lines and 56% ATS on closing lines. Which is the more reliable indicator?
    The fact that you did better against the opening than closing is a good indicator that your model got value, because market moved with you.

    However, what Heehawww said is true, it's very dificult to bet against the opening. I suggest you track your bets, starting from here, comparing (i) performance at time you could place the bet and (i) performance against closing.

  7. #7
    a4u2fear
    TEASE IT
    a4u2fear's Avatar SBR PRO
    Join Date: 01-29-10
    Posts: 8,147
    Betpoints: 35447

    what's more important that you didn't mention: how many plays and years is the 74% ATS from?

  8. #8
    Sawyer
    Sawyer's Avatar Become A Pro!
    Join Date: 06-01-09
    Posts: 7,592
    Betpoints: 6650

    Quote Originally Posted by a4u2fear View Post
    what's more important that you didn't mention: how many plays and years is the 74% ATS from?
    probably not much hehe. if it's %74, must be a small sample size like 100-200 plays.

  9. #9
    Sawyer
    Sawyer's Avatar Become A Pro!
    Join Date: 06-01-09
    Posts: 7,592
    Betpoints: 6650

    Quote Originally Posted by gui_m_p View Post
    The fact that you did better against the opening than closing is a good indicator that your model got value, because market moved with you.

    However, what Heehawww said is true, it's very dificult to bet against the opening. I suggest you track your bets, starting from here, comparing (i) performance at time you could place the bet and (i) performance against closing.
    You must always bet the opener. Because if you have an edge, you must bet early to get best price. And if line will move against you, then probably you don't have an edge.

    It's a good value if you picked Dnipro at 2,80, Bayern at 1,85. Bayern Closing was 1,63-1,65 and Dnipro closer was 2,25.

  10. #10
    statnerds
    Put me in coach
    statnerds's Avatar Become A Pro!
    Join Date: 09-23-09
    Posts: 4,047
    Betpoints: 103

    Quote Originally Posted by Sawyer View Post
    It's a good value if you picked...Bayern at 1,85. Bayern Closing was 1,63-1,65
    But what if your own numbers had Bayern at 1,55?

  11. #11
    antonyp22
    antonyp22's Avatar Become A Pro!
    Join Date: 01-12-14
    Posts: 78
    Betpoints: 2528

    74.5% across two seasons in the AFL (Australian Football League) where there are 18 teams and 23 rounds in a season. It was a total of 168 plays for 125 successful bets on openers.

  12. #12
    gui_m_p
    gui_m_p's Avatar Become A Pro!
    Join Date: 09-18-13
    Posts: 123
    Betpoints: 1537

    Quote Originally Posted by antonyp22 View Post
    74.5% across two seasons in the AFL (Australian Football League) where there are 18 teams and 23 rounds in a season. It was a total of 168 plays for 125 successful bets on openers.
    That's good, I don't think that closing line value is much suscetible to variance like a game outcome. So your sample is not that small. Good luck!

    And help us, post your plays lol!

  13. #13
    antonyp22
    antonyp22's Avatar Become A Pro!
    Join Date: 01-12-14
    Posts: 78
    Betpoints: 2528

    Quote Originally Posted by gui_m_p View Post
    That's good, I don't think that closing line value is much suscetible to variance like a game outcome. So your sample is not that small. Good luck!

    And help us, post your plays lol!
    Lol I might have to keep my secrets to myself this season

  14. #14
    banger
    banger's Avatar Become A Pro!
    Join Date: 10-03-13
    Posts: 19
    Betpoints: 810

    Quote Originally Posted by antonyp22 View Post
    I've tested a model over a season and it has gone 74.5% ATS on opening lines and 56% ATS on closing lines. Which is the more reliable indicator?
    i read another forum post elsewhere from you that you got the lines from ausportsbetting which has last years, you say its 74.5% on openers, then say its for two years? where are you getting the openers from 2012?

    btw the openers from aussportsbetting are from when pinnacle releases them on Weds/Thurs, Australian bookmakers have had lines out from Monday

  15. #15
    antonyp22
    antonyp22's Avatar Become A Pro!
    Join Date: 01-12-14
    Posts: 78
    Betpoints: 2528

    Quote Originally Posted by banger View Post
    i read another forum post elsewhere from you that you got the lines from ausportsbetting which has last years, you say its 74.5% on openers, then say its for two years? where are you getting the openers from 2012?

    btw the openers from aussportsbetting are from when pinnacle releases them on Weds/Thurs, Australian bookmakers have had lines out from Monday
    Got the lines for 2012 from another source. Unfortunately they weren't from Pinnacle but from Aussie bookmakers so comparing them across two different sets of odds (I'm assuming that because the Aussie bookmakers release odds on Monday they'll be less efficient than Pinnacle odds released on a Thursday).

Top