• RanchWest
    503
    In racing literature and messaging, bias is a common topic.

    Bias is usually focused on a few winning efforts over a few days or even a few races.

    But in studies of topics other than bias, we usually talk in terms of a very minimum of 30 wins before we consider there to be any trend. So, why do we look at such a small sample of bias when that is contrary to our normal habits?

    Now, certainly, inside horses run a shorter distance than outside horses, so there is going to always be an advantage to the inside. Just look at the starting positions of human sprinters... starts are staggered to avoid lane advantage. Does the positional bias in horse racing shift?

    And, studies clearly do show that front runners win more often than closers.

    In winter, there is an explanation for the combination of inside and early speed bias. The horse that gets the early lead can take a firm inside ground for a shorter distance and pelt the trailing horses with ice and other solid particles that sting (the jockeys and probably the horses, too) in the cold weather. So, there is that.

    Do you trust an inside/outside bias or an early/late speed bias off of a small sample? Or is it really mostly based on the random chance of which horses are in assigned positions (gates) in the races? What do you think? How much of your handicapping relies on bias?
  • RanchWest
    503
    And, do you take into consideration track circumference and/or the run-up to the (first or only) turn?

    What about turf racing? Many people point to a bias against the outside horses. Is that because they are running on grass? Or is it because turf courses have tighter turns? And, do you adjust for odd configurations, such as the downhill races at Santa Anita? What about the "half turns" and chutes configurations?
  • RanchWest
    503
    When considering speed bias, do you only look at individual horses? Or do you in any way look at the field as a whole?
  • Tony Kofalt
    397
    Ranch- seriously well thought out group of questions. Thank you!! Here are a few thoughts from my perspective.

    But in studies of topics other than bias, we usually talk in terms of a very minimum of 30 wins before we consider there to be any trend. So, why do we look at such a small sample of bias when that is contrary to our normal habits?
    By definition, bias is a disproportionate weight in favor of or against something. I suppose that if a bias held up over a larger sample size it would become the norm and lose much of the value it may offer. That being said, I view horse racing as a cyclical game and I'm always looking for 'short term' trends that may offer value. For me a handful of races may be more than enough.
    I utilize a similar mindset when betting 'hot' trainers. I'm looking for trainers that win 2 or 3 of their last 4 or 5 races, not 12 of their last 30. My belief is that value is more prevalent in short term trends
  • RanchWest
    503
    Thanks, Tony. I appreciate your response.

    By the way, I was not anticipating any right or wrong answers. Just curious as to the thought processes to get to the pros or cons of bias.
  • Rich Val
    7

    I definitely believe in biases but learned a long time ago that the make up of the race has a lot to do with who wins.

    A race with 4 front runners does not have the same likelihood of a front runner winning as an race with no front runners. Therefore, you need multiple models for each distance and pace pressure.

    Back in the 1990's I used The Handicapper's Notebook software. At had this feature that looked at post position with early speed points. Took about a cool bias!

    It showed that an inside post could be a big positive for a closer and a big negative for a front runner!

    Running style means so much in building a model.
  • Conley
    424
    I am a firm believer in track bias and so much so that I have been keeping a notebook full of notes for keeping up with track bias at (almost) every track in North America for TBs

    What I do is look at the charts (preferably Equibase) and dissect them so that I get the full understanding of how the track was playing out and also what disadvantages I need to jot down so next time I see a shipper shipping into a track I am playing with a troubled line I can look back to see if I have that tracks notes and date to see if the horse was disadvantaged that specific day which does help so you don't throw away that horse just because he/she had a poor start without a lot of knowledge of how that horse REALLY ran

    Right now a big bias I am noticing in the following tracks is:
    Keeneland for Dirt Sprints favouring Front Runners with Stalkers at a disadvantage
    Tampa Bay favouring heavy Front Runners on Dirt Sprints with Stalkers/Trailers/Outside disadvantage
    Parx favouring Front on DS/DR with posts 4-6 13 wins last 3 race days and also Stalkers/Outside bad
    Turf Paradise DS/DR favouring Front Runners with Stalkers and Outside disadvantage


    Hope this helps in some sort of way but I am always willing to add something to this project if someone would like me to Thanks
  • RanchWest
    503
    Running style means so much in building a model.Rich Val

    Thanks, Rich. Do you use the published running style (E,E/P,P, etc.)? Do you model any similar information? I know some people model the %M off of their pace line.
  • Rich Val
    7
    Thanks, Rich. Do you use the published running style (E,E/P,P, etc.)? Do you model any similar information? I know some people model the %M off of their pace line.RanchWest

    Sorry, I've been gone for a couple of weeks,

    No. I use a slight variation on Dave's ESx object plus his approach from Understanding Early Speed.
  • RanchWest
    503
    Thanks, Rich!
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment

Please register to see more

Forum Members always see the latest updates and news first. Sign up today.