• mikesal57
    61
    Yes...Pct - IV - BV falls in line ..Hi to Low
    But
    ROI is same through out ranks...

    Do you use it in objects?

    I would say no...if yes, Why?

    fwiardrbsra43ar9.jpg
  • Gino
    32
    using factors with a rank of one win rate of 20% and an ROI of a buck forty nine
    puts you in the same boat with name bettors and guys with lucky numbers...
    PIV (pioneered by DS hisself) wil serve you far better over the long haul than IV...

    your "no" should be a resounding "NO" !!!!!
    sorry to be a thread killer, LOL, but even here in the land of exclusive free thinking, the only
    guy who is gonna say yes is gonna have to put down his darts so he can type....

    gino
  • Dave Schwartz
    46
    I would also say "No."

    Just guessing here, but it appears that you're looking at a subset of data. That's the only thing I can think of that could produce such a completely flat $Net across all ranks.
  • mikesal57
    61
    Actually Dave this was done with ALL races from 2018..........
    Quite a few were like this.....
    Even PSR is like this...
    2v4ywl1m97mvg1x8.jpg
    With PSR Ranks 1-2-3 are about the same but the First Half being at 1.49 tells me ranks 4-5-6 are way lower...

    I will run it against 2017 to see if its similar........

    So say it is relatively flat......What does that do in terms of useability.....?
    This is what I dont get....
    PCT wise its fine it flows downward like it should
    But if the ROI's dont flow downward , then I believe this factor has no validity....

    I'm I correct?
  • mikesal57
    61
    What I'm trying to say....

    What makes a impact factor work?

    Downward flow in PCT....Downward flow in ROI....or Downward flow in BOTH ????
  • mikesal57
    61
    thank you Gino....I had been seeing the PIV is better for judging the factor...

    Even tho a factor looks valid the PIV shows it isn't..
    and the example above is a definite 'NO"

    PS-Gino, thanks for a post that I can understand :)
  • Gino
    32
    mr mikey-
    the science part of Handicapping Resurge is fairly simple, if you can figger out the overlap between
    Catholics and Episcopalians, you can make up your mind about Factor Analysis....
    1. Win %
    2. ROI
    3. PIV

    how you value these three metrics is up to you and your method....but anyway ya slice it or dice it,
    ya need some meat in the sammy, capische?
    Accentuate The Positive
    Cross reference any killer horse with your mother's birthday or Game 7 telecasts...that bet didna get
    made, and if it is what nudges ya into the black, be real....no fudging yer way to profits...

    make it yer own, be a non-prophet organization...

    if you are going down the break even, live off the rebate strategy, well boyo yer gettin into sunni/shiite
    territory...cross gino off yer list, we got nothin to talk about....

    rank o' one $1.49 is a stone loser and so is a buck sixty six....
    something is haywire in yer sample as far as a starting point....
    think Archie Bell and The Drells


    amor y paz,
    chino
  • Dave Schwartz
    46
    What makes a impact factor work?

    Downward flow in PCT....Downward flow in ROI....or Downward flow in BOTH ????
    mikesal57

    YES.

    What makes "flat" as a bad thing is that there is no advantage or disadvantage to be gained with the factor. IOW, it is a waste of time.
  • Dave Schwartz
    46
    PS-Gino, thanks for a post that I can understandmikesal57

    Welcome to the Gino Bennetti Fan Club and Chowder Society. (GBFCaCS)
  • mikesal57
    61
    He just ruined it...:(
  • mikesal57
    61
    In regression analysis , winners are looked at vs Rank #1 in each factor...
    So if the ROI's in ranks 2-3-4 are the same as rank 1 , then theres no difference between the rankings , and makes the factor useless.
    This is going to be a job that I need to look into more..
    Most likely I'll have to manually evaluate each factor
    My decisions are:

    Do I look at FH vs RH?
    Do I look at Ranks 1-2-3 and for get the rest
    Do I look at Rank1 vs RH for a big difference

    Theres going to be a few scenarios that I'll run across and then need to make a decision...

    Or am I making this a bigger thing than it really is?
  • mikesal57
    61
    Ok...here we got PSR for 2019 Jan - July

    cy92p8ldstbmsqdg.jpg

    Compare it to above's 2018, it almost identical
    Ranks 1-2-3 are about the same but after that it flows downward.....
    Does flowing downward after rank 3 help the situation?
    I dont believe so because theres no separation between 1-2-3
    I'll have to look at others to see the whole story...

    Just letting you know Dave, I'm not picking on the factors itself
    I'm trying to understand whats happening and to better utilize the findings.

    Did you know that Rank 3 for PSR was no different than Rank 1?
    its a new ballgame now and we gotta change it up

    How? Have no clue right now...
  • Gino
    32
    yo mike-
    well for my sake ya better hurry up....i just heerd thru the cyber grapevine that you'll be working this out back to Facebook as this here site is getting its water shut off...
    i don't do Facebook or PA/ FU, so vaya con dios, amigos....

    gino
    10-7
  • mikesal57
    61
    I know G.....

    Its a shame most of you old timers will be lost and missed because of FB
    i wasnt a big FB user till I got HSH.......its not a big deal
    Hope you guys come around...
    Also , I dont post much of this because its sometimes over their heads and I dont want to clutter Daves pages with tech stuff
    Mike
  • mikesal57
    61
    My decision on this will be to check , as Gino mentions, Daves PIV values.....compared to the others its the only one that tells me the value of the factor ....as long as there is a downward flow, it will validate it.
    With PSR...its good to go
  • mikesal57
    61
    All this will help me in my main project , where I can have a list of plays instead of going thru the whole HSH process for each race
    .....

    This excel format is set....just working on factors now
    In this race the #2 paid $20 :)

    aitzm3j2844fpju8.jpg
  • mikesal57
    61
    Dave is it possible & feasible to change the factor like this...
    Original Ranks
    yguf4a8kz4b81fb7.jpg

    My adjusted Ranks
    awobhl1nxr79lrq5.jpg


    ??
  • Dave Schwartz
    46
    Are you asking if you can put them together like this?
    1
    2-3
    4-5
    6-7
    8+

    Sure. Look at the IV Table structure.

    d28kigo42p2dunms.jpg
  • mikesal57
    61
    Got it...will give it a go...
    thxs
  • Gino
    32
    mr met-
    there are some factors that vs. certain filters may have horizontal value and not downward
    linearity and still be very powerful handicapping tools....
    downward linearity is very comforting and certainly is easier to comprehend, but i don't see it
    as the most important result found in the sample....you may be using a factor as a stepping stone
    or gateway to something bigger and better, and perhaps a solid foundation is very important
    to your process, if so, disregard my nosy advice....
    i had an 8/5 ML PSR 1 horse pay $12.60 today....a "to the moon, Alice" result that fluffs up
    your horizontal metrics in a most delightful way...
    keep up the work, focus on your process, and poco a poco the results start to make sense...

    until the plug is pulled,
    gino de dos centavos
  • Dave Schwartz
    46
    until the plug is pulled,
    gino de dos centavos
    Gino

    We're rethinking the plug-pulling.



    Those factors are not the same. The hit rates are far different.

    Try this... build a 2-dimensional IV table:
    PSR with Odds Rank.

    What does that look like?
  • Gino
    32
    i used the cackle-later in my head, methinks the hit rates are the same...the second table is
    combo ranks....
    go sal go
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.

More Discussions