PGA FORUM

Comments

  • PuddinCheeks

    • Blogger of the Month

    Quicken Loans National

    • Congressional CC (Blue)(Par71) • Bethesda, MD
    • June 23rd – June 26th

    Event Websites
    Quicken Loans National
    PGA Tour Quicken Loans

    Past Quicken Loans National champions and sites

    • 2015 – Troy Merritt(-18), Cut(-2); Robert Trent Jones Golf Club, Gainesville, Va.(Par 71)
    • 2014 – Justin Rose(-4), Cut(+3); Congressional Country Club, Bethesda, Md.(Par 71)
    • 2013 – Bill Haas(-12), Cut(+3); Congressional Country Club, Bethesda, Md.(Par 71)
    • 2012 – Tiger Woods(-8), Cut(+6); Congressional Country Club, Bethesda, Md.(Par 71)
    • 2011 – Nick Watney(-13), Cut(+3); Aronimink Golf Club, Newtown Square, Pa.(Par 70)
    • 2010 – Justin Rose(-10), Cut(+3); Aronimink Golf Club, Newtown Square, Pa.(Par 70)
    • 2009 – Tiger Woods(-13), Cut(+2); Congressional Country Club, Bethesda, Md.(Par 70)
    • 2008 – Anthony Kim(-12), Cut(+3); Congressional Country Club, Bethesda, Md.(Par 70)
    • 2007 – K.J. Choi(-9), Cut(+4); Congressional Country Club, Bethesda, Md.(Par 70)

    You can see that Congressional was changed from Par 70 to Par 71 in 2012 so keep that in mind when doing history research.

    Links:
    Ownership Reports
    JimKronlund $10 50/50

  • BIF

    Thanks for opening the thread – I probably won’t be that energetic to do that for a few weeks – I have lots going on…..

    I posted the following in the US Open thread but copied it over to here as well:

    I’m going to be pretty busy over next 2 weeks so my activity on here will likely be a bunch of posts separated by hours of inactivity – we sold our house and bought a new place 13 hours away – moving day July 4th – now packing and getting things ready for the move plus I have a big 3 day Mid-Am tournament next week that is a qualifier for the national Mid-Am so I need to get some reps in for that.

    I’m pondering taking a week off from DFS but since that is difficult to do, I may just play a few H2H’s and some satellite/qualifiers

  • PuddinCheeks

    • Blogger of the Month

    Good luck with the Mid-Am. I’m finally working on my game again now that the kids are a little older. I’ve lost almost all touch around the greens but still hit straight shots getting there.

  • BIF

    @PuddinCheeks said...

    Good luck with the Mid-Am. I’m finally working on my game again now that the kids are a little older. I’ve lost almost all touch around the greens but still hit straight shots getting there.

    Thanks – my short game is normally great (well pitching/chipping that is – the putter and I have our differences at times but still much better than recent years, Thank You to Scotty Cameron and my big/ugly looking Futura X putter) but I have not played much the past month since we sold our house so I’m hoping to dedicate 1 hour a day on the range the next 6 days to tighten things up. I’m pretty familiar with the course the tournament is on but I have never putted real well there so getting out for a practice round this afternoon to try to change that. I honestly do not ever remember making a birdie putt there outside of 4 feet – greens are very quick and mostly flat with very subtle breaks so tough to read (for me anyways); I’d rather play crazy greens with big swales/mounds/tiers as at least you know you have a breaking putt and which way. The golf course also was built as a 9 hole course in the 40’s or 50’s then a new nine was added with modern architecture in the late 80’s – the two sets of holes are intertwined so you play a couple old then a couple new, etc…so the greens are a little inconsistent as well from one to the next.

  • DomTwan

    The US Open Hangover Invitational

  • badtrip

    I hate when they make the $3 tourny the 100k first place. Might just play a couple doglegs. Not going to get to watch any of it anyway. Spending the weekend at Mystic Lake Resort for my cousins daughters graduation party. Heard their course is real nice and playing it Sunday.

    Had a golf meltdown yesterday. Shot 96 on a course I once shot 66 on. I actually hit it pretty good for the first five holes. Took one puff off a peace pipe a guy in the group had on six tee and carnage followed.

    Hope the OT stuff is OK until the pricing comes out.

  • smallANDflaccid

    Last week I went back through and collected stats on the 2014 and 2015 FedEx St Jude and looked to see which ones were most highly correlated to DK score.
    I then used that to see how well that could predict 2016.

    (Technically I cared less about prediction, and more about finding the top 3, and then building a 3D scatter plot showing the 2016 players on those stats, and where the previous winners were in that cloud – still working on that. As for the prediction part, wasn’t great – at least compared to other algos.)

    Now looking at previous Quicken – last year was a different course, and I don’t have 2013 data to calc the correct stats – so with sample size of one tournament, here are some key correlations.
    (huge grain of salt here, as the data could be noisy, and the way it is gathered could be questionable)

    Quicken:
    PctCutsMade: 0.386
    3PuttFreq: -0.364
    BirdiesPerRound: 0.363
    BirdieRate: -0.326 (why these aren’t the same, or even closer, I am not sure – they aren’t the exact same stat, but I would still expect a closer corr to each other)
    AvgP4Score: -0.322
    PctBogeys: -0.271
    AvgMadePuttLength: 0.240

    Whether that level of correlation is useful is already something you could argue over (those are pretty weak numbers), and the rest of the stats (I track ~17 in this case) are lower correlations (0.20 or lower for the most part it seems)

    To put in perspective, St. Judes was on the same course in 2014 and 2015, yet they massively didn’t agree with each other on many stats between years.
    And if you would think that all courses will likely have the same stats at the top, it doesn’t seem to be.
    For example 2014 St Jude was in agreement with an avg correlation of -0.42ish on Par4Scoring – but a huge disagreement on AvtMadePuttLength – in 2014 it was 0.408 and in 2015 it was 0.059.

    That said – when you are looking at DK scores, I think made cut pct is likely always going to be way up there.

    If it isn’t obvious, going through those stats, they would read as the larger the absolute value of the number is, the more important it is in contributing to a higher DK final score. If the number is positive, then it means as the stat goes up, so does the DK score. If the number is negative, then it means as the stat value goes down, the DK score goes up.

    So you would want the lowest AvgPar4Score possible, and the highest Pct Cut Made possible to maximize the DK score.
    (not exactly earth shattering there)

    Once they release the list of names, I will see if I can show you who most correlates.
    (Keeping in mind that doesn’t seem to help terribly much in predicting what actually happens from what I have seen.)

  • smallANDflaccid

    Another idea from ^^ there is that since you don’t have enough data with just one sample, you can create a synthetic/derivative by finding courses that closely correlate on the same stats. That builds a larger sample size, and then if a player did well in that sample space, they should do better here.

    Extrapolating that out, you then do that at the hole level.

    That is where I have seen the most success.

    I may be switching jobs soon and won’t have time to play, so I may start posting how diff algo ideas work.
    (keeping in mind I’ve gotten destroyed the past few weeks, so if anything, do the opposite of anything I say)

  • JimKronlund

    Love this stuff ty for sharing.

    “BirdiesPerRound: 0.363
    BirdieRate: -0.326 (why these aren’t the same, or even closer, I am not sure – they aren’t the exact same stat, but I would still expect a closer corr to each other”

    Perhaps the birdie rate is lower because the same few players had most of the birdies. Thus making the Birdie rate lower which still allows for a larger variance of Birdies per round to be higher or lower.

  • smallANDflaccid

    @JimKronlund said...

    Love this stuff ty for sharing.

    “BirdiesPerRound: 0.363
    BirdieRate: -0.326 (why these aren’t the same, or even closer, I am not sure – they aren’t the exact same stat, but I would still expect a closer corr to each other”

    Perhaps the birdie rate is lower because the same few players had most of the birdies. Thus making the Birdie rate lower which still allows for a larger variance of Birdies per round to be higher or lower.

    Could be – but the stats are calculated excluding future knowledge and excluding this course.
    So it is saying, if you were to look at that course back when they were doing it, with what you knew then, what stats on those guys would be most correlated to what they were about to do.

    vs what things they did do that most contributed (slight difference and I think the way you described it was more the latter)

    Although from what I have seen here, data leaks like that don’t impact it as much as I would have expected (and as it does in some other predictive areas) – perhaps b/c the stats don’t change a whole lot
    (I’ve tried it using total knowledge and using no future knowledge and the output is close enough that I am not sure how much it really matters)

    Could also just be different sample sizes. Also what I call stats may not be the same standard as I may be misnaming things.

    Birdie Rate is how many holes before they see their next birdie.
    Birdies Per Round literally looks at each round they ever played and gets their avg birdies per round (which isn’t quite the same as adding up their total birdies and their total rounds due to the order things are done).

    Those slight differences may contribute as well.

  • PuddinCheeks

    • Blogger of the Month

    I think the St. Jude numbers are showing differences because they switched from Bentgrass to Bermuda greens this last year, right?

  • JimKronlund

    @smallANDflaccid said...

    “BirdiesPerRound: 0.363
    BirdieRate: -0.326

    In reality those numbers are so close it would be hard to use them separately as a predictor IMO
    Still it is fascinating to see the actual strokes played broken down into these segments.

  • cowbell204

    Prices are up. Kizzire and Kaufman finally back to earth.

  • PuddinCheeks

    • Blogger of the Month

    @cowbell204 said...

    Prices are up. Kizzire and Kaufman finally back to earth.

    Alright time to start the real job of research now that prices are out.

  • cowbell204

    Hubbard for 5,400? Uh, sure.

  • smallANDflaccid

    @PuddinCheeks said...

    I think the St. Jude numbers are showing differences because they switched from Bentgrass to Bermuda greens this last year, right?

    If so, then the 2016 numbers should track to the 2015 numbers more than the 2014 and more than 2015 to 2014.
    I don’t have them up in front of me, and am too lazy to check right now as it takes a bit of time for me to run all of the numbers.

    But could be it – there were also weather things going on from what people responded when I asked on here when I was looking at it on Friday.

    Last time I tried to look into how much the grass matters I couldn’t prove it to myself – certain players, when you just do raw dumps of figures, do better/worse on diff grass types it looks. But that doesn’t take into account other differences and when I started doing that, the grass seemed to matter less.
    But that was months ago and I don’t really recall – so it is feasible I just did it wrong

  • thetitanlb

    Pricing is out and wow i love the value. There are some legit golfers sub 6k

  • cowbell204

    @thetitanlb said...

    Pricing is out and wow i love the value. There are some legit golfers sub 6k

    Fathauer & Michael Kim

  • rjgboom

    Pricing seems very soft.

  • Brian6035

    WOW this pricing is FLAT! Some amazing value picks here. Might leave 4 figures on the table in a few.

  • whodat2

    But who are you going to pay up for? Thought I was going to be sneaky and own Fowler, but at that price?

  • mschu1213

    Sigh. The complete shit payout in the birdie from DK the week after a major again. Will they be running Milly maker sats this week? With only a few weeks before they Open I was hoping they would start running them. Thats assuming they are even having one.

  • nevergymless

    @rjgboom said...

    Pricing seems very soft.

    It always feels soft in a 100 K 3 dollar tourney for some reason . I was only going to throw a couple bullets because the winning lineups always make me want to throw up in theses gpps but fuck I’m already at 20 line ups .

  • smallANDflaccid

    @smallANDflaccid said...

    Once they release the list of names, I will see if I can show you who most correlates.
    (Keeping in mind that doesn’t seem to help terribly much in predicting what actually happens from what I have seen.)

    I’ve been putting them into a 3D plot using the top 3 correlated stats (PctCutsMade, 3PuttFreq, and BirdiesPerRound), but I cannot find an angle where they are easy to read.
    Meaning they don’t really distribute out into a cloud where you easily see obvious clusters (at least with these stats).

    I normalized the data including the guys from the 2014 top 10 and then the new Salaries list.
    (Meaning that regardless of whether the underlying stat goes up or down to improve, and regardless over what range of numbers the stats cover, I make each column normalize to 1 is the best and 0 is the worst.)

    I’ll try to get some better figures to rank them, but some general points…

    Three Putt Freq:
    These guys are all better in that (normalized stat) than the guys who finished top 10 in 2014:
    Bryson DeChambeau
    Anirban Lahiri
    Hiroshi Iwata
    Daniel Summerhays
    Erik Compton
    Patton Kizzire

    Pct Cutts Made:
    These guys are all better than the guys who placed top 10 in 2014:
    Jim Furyk
    Bill Haas
    Ryan Palmer
    Patrick Reed
    Daniel Summerhays
    Webb Simpson
    Charles Howell III
    Gary Woodland
    Francesco Molinari

    Birdies Per Round
    On the one hand, only one guy is on the list before you hit one of the guys from 2014 top 10:
    Byeong-Hun An

    BUT, he’s way ahead of everyone else. Even names like Rickie Fowler.
    (this may be due to different sample sizes – haven’t looked at that – but I did cut it off to sample sizes where things should normalize out – vs total outliers)

    If you want someone in the Top 10 of each of the normalized stats… there isn’t anyone.
    If you push that to Top 20 for each, then you get three players on the list:
    Justin Rose (isn’t playing, but he won the 2014)
    Patton Kizzire
    Patrick Reed

    From experience, I wouldn’t expect either Kizzire or Reed to win, as wins tend to be outliers – but this would indicate top 20.
    How useful that is – probably not very useful.

    NOTE: these stats are spanning multiple years and not necessarily the same as PGA Tour’s stats looking at just this year

    I’ll try to get actual correlation distances from the 2014 top 10 now if things stay slow at work.

  • PuddinCheeks

    • Blogger of the Month

    Here is a link to the hole by hole stats for the 2014 Tournament. There is a bug on the site to posts the wrong tournament name in the top bar but you can ignore that.

    Quicken Loans, Congressional

  • thetitanlb

    Pay up for Reed, Chappel, Woodland. Just off the top of the head i like it though. Especially for cash games. Chances are most of the top guys make the cut.

  • X Unread Thread
  • X Thread with New Replies*
  • *Jumps to your first unread reply

Subforum Index

RotoGrinders.com is the home of the daily fantasy sports community. Our content, rankings, member blogs, promotions and forum discussion all cater to the players that like to create a new fantasy team every day of the week. Our goal is to help all of our members make more money playing daily fantasy sports!

Bet with your head, not over it!
Gambling Problem? Call 1-800-Gambler