NASCAR FORUM

  • gje627

    New Hampshire Motor Speedway

    Schedule — all times are Eastern Time (ET)

    — Practice 1: Friday, 12:00 PM (NBCSN)
    — Qualifying: Friday, 4:45 PM (NBCSN)
    — Practice 2: Saturday, 10:05 AM (CNBC)
    — Race: Foxwoods Resort Casino 301, Sunday, 2:00 PM (NBCSN)

    Race Details

    — 37 Drivers Entered; 301 Laps, 318.46 Miles
    — Stage 1: Lap 1 – 75; Stage 2: Lap 76 – 150; Stage 3: Lap 151 – 301
    — Expected Fuel/Pit Window – 85-90 Laps

    Note: Fuel Cell capacity is an estimate as of 7/19/2018.

    New Hampshire Motor Speedway – Fantasy Relevant Track Details/Rankings 2005 to 2018

    — 1.058 mile oval; 12-degree banking in all four (4) turns; 2-degree banking frontstretch/backstretch; frontstretch/backstretch 1,500 feet
    — Winner Average Margin of Victory: 1.888 seconds; Low – 0.068 seconds (2007, race #17); High – 8.941 seconds (2015, race #28)
    — Rank Most Green Flag Passes: 12 of 23 Tracks (11.48 Green Flag Passes per 100 Miles)
    — Rank Most Quality Passes (Top 15 Passes): 12 of 23 Tracks (3.90 Quality Passes per 100 Miles)
    — Rank Most Lead Changes: 13 of 23 Tracks (4.01 Lead Changes per 100 Miles)
    — Rank Most Cautions: 7 of 23 Tracks (2.06 Cautions per 100 Miles)
    — Rank Highest % Drivers Finishing on Lead Lap: 8 of 23 Tracks (52.72% per Drivers/Race)
    — Rank Highest % DNF Drivers – Mechanical: 11 of 23 Tracks (8.51% per Drivers/Race)
    — Rank Highest % DNF Drivers – Crash: 18 of 23 Tracks (4.53% per Drivers/Race)
    — Rank Highest % DNF Drivers – Overall: 18 of 23 Tracks (13.04% per Drivers/Race)

    Note: Data and Rankings are Cup series only. Green Flag Passes, Quality Passes, Lead Changes, and Cautions standardized to 100 miles per driver to account for differences in miles per race and number of drivers in field. Lead Lap and all DNF categories standardized to drivers per race. All statistics updated through most recent race (Kentucky Speedway, 7/14/2018).

    Comparable Tracks

    Indianapolis Motor Speedway, Martinsville Speedway, ISM (Phoenix International) Raceway, Pocono Raceway, Richmond International Raceway

    Last Six (6) Race Winners

    — 7/19/2015: Kyle Busch
    — 9/27/2015: Matt Kenseth
    — 7/17/2016: Matt Kenseth
    — 9/25/2016: Kevin Harvick
    — 7/16/2017: Denny Hamlin
    — 9/24/2017: Kyle Busch

    Last Six (6) DK and FD Fantasy Points Leaders

    7/19/2015: Brad Keselowski (106.00 DK FPts) – Kyle Busch (84.2 FD FPts)
    9/27/2015: Kevin Harvick (104.50 DK FPts) – Matt Kenseth (81.7 FD FPts)
    7/17/2016: Matt Kenseth (100.50 DK FPts) – Matt Kenseth (85.4 FD FPts)
    9/25/2016: Matt Kenseth (97.75 DK FPts) – Matt Kenseth (83.5 FD FPts)
    7/16/2017: Martin Truex, Jr. (111.25 DK FPts) – Kyle Larson (88.7 FD FPts)
    9/24/2017: Kyle Busch (137.25 DK FPts) – Kyle Busch (91.7 FD FPts)

    Race Data, News, and Information

    Rotogrinders NASCAR Free tools
    ESPN Jayski
    Motor Racing Network
    Racing Reference

    Race Day Breaking News and Weather

    Stephen Young – stevietpfl
    Bob Pockrass
    NASCAR Weather

    Note: As needed information above will be updated.

  • gje627

    Update to a Previous Post: Almost two (2) weeks ago in the Daytona forum, very last post in the thread that can be found here Daytona Forum, I stated I was doing some research on a number of measures of predictability in fantasy NASCAR and identifying the best lineup-type, etc. per track.

    While I haven’t written anything-up just yet, I am finished with all of the analysis I plan to perform, so if anyone is interested in the results, or peer-reviewing, critiquing what I will eventually write, please let me know.

    While I know this may not be of interest to a lot of people here, the only request is that if you wish to see the results, you be a regular, helpful, courteous, and respectful contributor to this thread this year. Been hammered by trolls and other ne’er do wells recently (my own fault, as I know sometimes I can be a dumbass here), but since I’m not paid for the headache I only want to share with people I trust and who will find the analysis useful.

    As a FYI, this week I will likely only have information for New Hampshire ready/available, as I don’t plan on writing everything up until the next NASCAR off-week after the August 18 Bristol race.

    Thanks !!!

  • jt77316

    Thanks for posting the thread. I would be more than happy to take a look at anything you’re willing to share.

  • Toddalan2286

    @jt77316 said...

    Thanks for posting the thread. I would be more than happy to take a look at anything you’re willing to share.

    I would be happy to see it as well!

  • jt77316

    Harvick scored 104.5 in the 2015 Fall race and had the highest driver rating (130.8) on the day. For those like me wondering how he wound up finishing 21st, it was because he ran out of gas. He had finished 42nd the week before, and was chasing the win to ensure he would advance to the next round of the chase. His crew estimated he would run out a lap early, and they were correct. Harvick went on to win the next week at Dover, securing his position to the next round. Off topic, but the next round ended with the infamous Talledega race where Harvick knew the motor was blowing, and he wrecked Trevor Bayne to end the race under caution so that he could advance yet again.

    This proved extremely controversial because it also robbed Dale Jr. of the win at Talledega as NASCAR determined Joey Logano was leading at the time of the caution and credited him with his third straight win, sweeping the Contender round. Had Dale Jr. won, he would have advanced to the next round (round of 8). Dale Jr. subsequently won the race before Homestead (Phoenix) and had the Harvick drama not went down at Talledega, would’ve put Dale in the Championship 4 in Miami. Since he had already been eliminated, the win was nothing more than a nice payday, and fodder for us to now discuss what might have been.

  • nickfromcwe

    • 849

      RG Overall Ranking

    @gje627 said...

    While I haven’t written anything-up just yet, I am finished with all of the analysis I plan to perform, so if anyone is interested in the results, or peer-reviewing, critiquing what I will eventually write, please let me know.

    I have not been a regular, but plan to be. I would love to look at whatever you have put together.

  • gje627

    That’s great…

    Thanks jt77316, Toddalan2286 and nickfromcwe.

    I’m going to post some summary correlations here in this thread sometime today. I’ll let you guys know when I’ll have the more advanced New Hampshire information formatted and available for this week. Guessing Thursday or Friday.

  • Cooper08

    • Blogger of the Month

    Hey GJE, I would be happy to see your info. Thanks!

  • gje627

    @Cooper08 said...

    Hey GJE, I would be happy to see your info. Thanks!

    Even before you asked, you were already on my list…. :)

    And like it or not, Stevie you are too…

  • Cooper08

    • Blogger of the Month

    @gje627 said...

    Even before you asked, you were already on my list…. :)

    And like it or not, Stevie you are too…

    Thanks!

    I tried to link from twitter but couldn’t get the video, Byron had a bad accident today at Charlotte on roval for practice…totally looks like he forgot to turn left and hit the tire wall hard..just saw going back to try to link video again he lost his brakes, he ok.

  • gje627

    @Cooper08 said...

    Thanks!

    I tried to link from twitter but couldn’t get the video, Byron had a bad accident today at Charlotte on roval for practice…totally looks like he forgot to turn left and hit the tire wall hard..just saw going back to try to link video again he lost his brakes, he ok.

    Try this: William Byron – ROVAL Crash

    The link includes two (2) videos.

  • gje627

    As promised, forthcoming are three (3) tables.

    If you recall, the task at hand is to try and determine which, if any variables known before a race can be used to accurately predict/project race results, DK fantasy point results (rank), and FD fantasy point results (rank)….

    Again, the impetus for doing this research was my contention that plate tracks were a “crap-shoot” and for fantasy purposes there is no reliable strategy (skill) to select drivers and construct lineups for races at Daytona and Talladega (yes, one of my “sour-grape” moments)….

    Thus, to begin to answer our questions one must first determine whether two (2) or more variables are correlated.

    Per a previous post there are two (2) primary measures of correlation, by far the most common is Pearson’s but for NASCAR variables the better measure is Spearman’s rank correlation.

    If you recall, in the Talladega race thread I used Spearman’s to compare the differences between fantasy scoring between Draftkings and FanDuel, just prior to the inaugural FD NASCAR contests. Thus, to avoid an even more long-winded post than necessary please go to that thread for my explanation of some of the reasons (not all, if you want more just ask) why Spearman’s is preferable to Pearson’s for NASCAR data. My post can be found here: Spearman’s Rank Correlation.

    As indicated by the subsequent tables, we are trying to predict three (3) things to apply our skill to identify drivers and identify groups of drivers in combinations of five (5) or six (6) depending on the site, that will win fantasy NASCAR contests. Again, what we are trying to predict are:

    1. Finish Order
    2. DK Fantasy Point Total (rank)
    3. FD Fantasy Point Total (rank)

    Additionally, during lineup construction there are several things we do know before a race starts. Important among the known information is:

    1. Start Position
    2. Qualifying Position (sometimes different than Start Position)
    3. Practice Speed (rank)
    4. Quality of Driver/Car/Team

    Thus, these are the predictor variables I have chosen for this research.

    While the first three are self-explanatory, for Practice Speed (rank) I have decided to use highest single lap speed rank across all practices. I did this for a number of reasons, primarily some races only have two (2) practice sessions, other races three (3), and some none, if weather forces cancellation for all practice sessions. Additionally, practices occur in different order for different races. For example, sometimes “Happy Hour”, the final practice session before a race occurs, takes place before qualifying, and sometimes it occurs after qualifying; hence the order of practice sessions relative to qualifying changes the goals and strategy for how drivers/teams approach certain sessions (i.e. race-trim, qualifying trim, work on corner entry-exit, etc.)….

    Next, to project Quality of Driver, three (3) separate measures are used:

    1. Average Driver Rating for the Year
    2. Average DK Fantasy Point Rank
    3. Average FD Fantasy Point Rank

    Finally, a few general observations about correlation coefficients. First and foremost, correlation measures the direction and strength of a relationship between two (2) or more variables (Pearson’s also shows linearity, Spearman’s does not). Essentially, answering the question “Does a relationship even exist between two (2) or more variables?” In this regard, it is extremely useful, but again only as an important first step because it does little to nothing to help explain the exact nature of this relationship (the latter analysis will be provided to list members).

    Additionally, ideally correlation measures will be reported along with p-values. An excellent definition for p-values from Investopedia is: p-values are “the level of marginal significance within a statistical hypothesis test representing the probability of the occurrence of a given event.” Here I am not reporting p-values, as given the length of this post trying to explain p-values would likely take another page or two, and while important p-values are less important than the underlying correlation.

    Finally, always remember, that measures of correlation range from -1.000000 to +1.0000 and a negative number does NOT mean “no correlation”, rather a negative number represents a direction of the relationship that is the opposite of a positive number. Thus, the important thing to remember when reading correlation tables is that the further away the correlation coefficient is from absolute zero (positive or negative) the stronger the correlation between two (2) variables is.

    Thus, to assist with correlation interpretation, please use the following table:

    –1.000000 – A perfect downhill (negative) relationship
    –0.700000 – A strong downhill (negative) relationship
    –0.500000 – A moderate downhill (negative) relationship
    –0.300000 – A weak downhill (negative) relationship
    0.000000 – No relationship
    +0.300000 – A weak uphill (positive) relationship
    +0.500000 – A moderate uphill (positive) relationship
    +0.700000 – A strong uphill (positive) relationship
    +1.000000 – A perfect uphill (positive) relationship

    And this is the easy stuff !!!

    But, without further adieu…..

    Three (3) tables within the next 30 75 minutes, or so….

  • gje627

    Per previous post….

    All errors and omissions are mine and mine only.

    Please advise if you see what you believe to be a likely error.

    Please advise with questions.

    EDIT: Hopefully it’s obvious, but correlations begin in column four (4)… Races indicate number of races at each track and drivers indicate number of drivers at each track…. Races are included because later analysis sample/population size is important as a measure of confidence; drivers are important for later analysis requiring a F-statistic.

    NASCAR 2005 to 2018 – Spearman’s Rank Correlation: Finish Position Correlated with Various Variables

    Track Races Drivers Start High Practice Qualify Driver Rating – Year Total DK FPts – Year Total FD FPts – Year
    ———————- ———— ————— ————- —————- ————- —————- ————— —————
    Atlanta 20 845 0.420746 0.454340 0.327445 0.604700 0.584566 0.588046
    Auto Club 20 846 0.454071 0.513618 0.393184 0.669525 0.648621 0.668057
    Bristol 27 1144 0.398171 0.374312 0.286793 0.571760 0.557955 0.578551
    Charlotte 27 1146 0.474771 0.482305 0.426661 0.591686 0.562105 0.586154
    Chicagoland 14 592 0.553137 0.571811 0.333050 0.727287 0.690293 0.709135
    Darlington 13 553 0.490823 0.425618 0.421222 0.689866 0.663236 0.672411
    Daytona 28 1186 0.121849 0.154582 0.027573 0.300020 0.335164 0.342484
    Dover 27 1143 0.514318 0.529583 0.379262 0.684822 0.656120 0.674387
    Homestead 13 552 0.503993 0.517891 0.495346 0.621126 0.600008 0.619555
    Indianapolis 13 553 0.473797 0.490991 0.457033 0.625708 0.599870 0.610507
    Kansas 21 885 0.471298 0.504397 0.467714 0.659087 0.650813 0.658022
    Kentucky 8 333 0.648459 0.596637 0.325454 0.739339 0.712954 0.719494
    Las Vegas 14 588 0.487456 0.566401 0.395848 0.657588 0.612802 0.639969
    Martinsville 27 1142 0.502287 0.454117 0.340143 0.633247 0.615443 0.623840
    Michigan 27 1141 0.489522 0.526089 0.448519 0.664348 0.630251 0.645221
    New Hampshire 26 1104 0.561336 0.578650 0.485966 0.709827 0.683634 0.703578
    Phoenix 27 1141 0.549930 0.581962 0.523417 0.685199 0.657809 0.667733
    Pocono 27 1141 0.566604 0.538954 0.442801 0.665609 0.622419 0.650408
    Richmond 27 1141 0.541934 0.518926 0.356683 0.702786 0.669216 0.685376
    Sonoma 14 589 0.414925 0.448752 0.415275 0.584075 0.552728 0.575539
    Talladega 27 1146 0.169980 0.118607 0.115627 0.349554 0.363460 0.378079
    Texas 27 1143 0.489567 0.515316 0.439264 0.659200 0.626726 0.647140
    Watkins Glen 13 550 0.503358 0.499862 0.363781 0.555203 0.536514 0.533989

    NASCAR 2005 to 2018 – Spearman’s Rank Correlation: DraftKings FPts Rank Correlated with Various Variables

    Track Races Drivers Start High Practice Qualify Driver Rating – Year Total DK FPts – Year Total FD FPts – Year
    ———————- ———— ————— ————- —————- ————- —————- ————— —————
    Atlanta 20 845 0.084053 0.263820 0.054467 0.439012 0.471506 0.453639
    Auto Club 20 846 0.110543 0.325359 0.106289 0.510254 0.555203 0.552966
    Bristol 27 1144 0.123322 0.231814 0.077403 0.464964 0.497435 0.496469
    Charlotte 27 1146 0.177513 0.306124 0.155283 0.455500 0.491146 0.491830
    Chicagoland 14 592 0.258980 0.394814 0.144045 0.580308 0.594606 0.594958
    Darlington 13 553 0.177228 0.254104 0.152884 0.524316 0.552415 0.543420
    Daytona 28 1186 -0.271567 0.044674 -0.109457 0.112931 0.208843 0.183137
    Dover 27 1143 0.229960 0.360491 0.173712 0.533319 0.546427 0.547041
    Homestead 13 552 0.165048 0.329560 0.161545 0.462333 0.503444 0.502450
    Indianapolis 13 553 0.123524 0.281567 0.113086 0.459527 0.506336 0.489730
    Kansas 21 885 0.166944 0.336626 0.173451 0.499770 0.546473 0.532025
    Kentucky 8 333 0.391901 0.382325 0.180111 0.561340 0.592108 0.571502
    Las Vegas 14 588 0.170976 0.364239 0.120030 0.505493 0.511689 0.520020
    Martinsville 27 1142 0.211123 0.303051 0.140007 0.476664 0.506175 0.498187
    Michigan 27 1141 0.154970 0.332444 0.139884 0.492358 0.528142 0.519337
    New Hampshire 26 1104 0.265586 0.383912 0.233770 0.541530 0.567810 0.569237
    Phoenix 27 1141 0.237945 0.393692 0.230848 0.522742 0.561315 0.555562
    Pocono 27 1141 0.244937 0.337136 0.177188 0.472614 0.488528 0.493343
    Richmond 27 1141 0.243817 0.349922 0.155911 0.552736 0.579815 0.571685
    Sonoma 14 589 0.044576 0.237140 0.048019 0.401473 0.439625 0.434191
    Talladega 27 1146 -0.237934 0.023427 -0.176726 0.184286 0.271795 0.251496
    Texas 27 1143 0.178796 0.335792 0.166193 0.511909 0.545453 0.543030
    Watkins Glen 13 550 0.147746 0.324597 0.123237 0.333939 0.382384 0.343771

    NASCAR 2005 to 2018 – Spearman’s Rank Correlation: FanDuel FPts Rank Correlated with Various Variables

    Track Races Drivers Start High Practice Qualify Driver Rating – Year Total DK FPts – Year Total FD FPts – Year
    ———————- ———— ————— ————- —————- ————- —————- ————— —————
    Atlanta 20 845 0.217343 0.346325 0.160095 0.513511 0.523648 0.517435
    Auto Club 20 846 0.232402 0.398416 0.209127 0.579227 0.600394 0.610009
    Bristol 27 1144 0.221791 0.284132 0.147584 0.508489 0.523065 0.533803
    Charlotte 27 1146 0.290249 0.380321 0.258033 0.512302 0.519059 0.532546
    Chicagoland 14 592 0.373324 0.475760 0.219789 0.651601 0.637254 0.649532
    Darlington 13 553 0.308138 0.334538 0.263839 0.611714 0.612469 0.615857
    Daytona 28 1186 -0.121713 0.092846 -0.058815 0.192839 0.266246 0.254563
    Dover 27 1143 0.348727 0.439338 0.259607 0.612068 0.604422 0.617294
    Homestead 13 552 0.307850 0.415363 0.302507 0.539057 0.557235 0.565735
    Indianapolis 13 553 0.264466 0.374928 0.250581 0.547250 0.565767 0.562860
    Kansas 21 885 0.283068 0.408121 0.285176 0.574252 0.596757 0.592510
    Kentucky 8 333 0.517904 0.489910 0.256122 0.664352 0.662916 0.656386
    Las Vegas 14 588 0.291976 0.447674 0.224259 0.571976 0.557092 0.575170
    Martinsville 27 1142 0.321742 0.358706 0.217109 0.550294 0.559932 0.561722
    Michigan 27 1141 0.284591 0.416864 0.261135 0.572798 0.577744 0.582341
    New Hampshire 26 1104 0.384945 0.471097 0.335776 0.627721 0.629837 0.644587
    Phoenix 27 1141 0.362781 0.483027 0.346998 0.605965 0.612422 0.615188
    Pocono 27 1141 0.384647 0.433274 0.290459 0.573636 0.560658 0.579044
    Richmond 27 1141 0.360606 0.421588 0.235892 0.620611 0.616225 0.622155
    Sonoma 14 589 0.181813 0.322756 0.183108 0.486547 0.497563 0.502983
    Talladega 27 1146 -0.071266 0.069146 -0.059719 0.265954 0.323144 0.318134
    Texas 27 1143 0.306074 0.421267 0.276765 0.584199 0.586128 0.596903
    Watkins Glen 13 550 0.291861 0.406237 0.224208 0.437134 0.452934 0.429965
  • gje627

    NOTE: Updated thread header to reflect that only 37 drivers are entered in the New Hampshire race this weekend.

    I should have a better idea of fuel mileage by tomorrow morning, and that will be updated then.

    Fuel mileage updated in header to 85-90 laps. Plenty of fuel to get through each of the first two (2) stages.

  • unageo09

    Per, DK, we are getting a .50 cent mini max this weekend. It’s not up in the lobby yet though.

  • wscooby

    Hey, gje, Thanks for posting the correlations!! I cannot comprehend how much work you have put into it. Maybe a correlation comparison b/t stage racing and pre-stage racing.should more emphasis be placed on the track correlations since stage racing began?

  • Cooper08

    • Blogger of the Month

    @wscooby said...

    Hey, gje, Thanks for posting the correlations!! I cannot comprehend how much work you have put into it. Maybe a correlation comparison b/t stage racing and pre-stage racing.should more emphasis be placed on the track correlations since stage racing began?

    Nice question! I am sure you realize the data is small on this but still interesting. Perhaps it could be broken down by track type so the data will have a slightly larger sample. Great question.

  • gje627

    @wscooby said...

    Hey, gje, Thanks for posting the correlations!! I cannot comprehend how much work you have put into it. Maybe a correlation comparison b/t stage racing and pre-stage racing.should more emphasis be placed on the track correlations since stage racing began?

    Don’t ask great questions, PLEASE !!! :)

    By nature, I’m lazy and even “minor-more-work” gives me a headache and requires a nap…. (:

    Just kidding.

    You do make an excellent point and there’s also a case for doing this by year too, and then cross-comparing the results to account for the different car generation (aero, etc.) ….Which I have done BTW….

    But, there’s only so much time in the day and you saw the effort I made in explaining correlations, and the reality that correlations are the the easy part…..

    Since I want to be sure that everyone is clear on the advantages and limitations of what I report, I feel I need to over-explain everything so nobody misinterprets any one (1) set of analysis (i.e. the correlations above) as being the “Holy Grail” of NASCAR predictive analytics….

    Ideally, I would have multiple people looking over my shoulder, peer reviewing my results, and decisions (i.e. “Why did gje627 re-code practice data for weather events this way, particularly because in this practice session not all drivers even had a chance to even take the track?” or “Where the hell are DNF results included in this analysis, given that since 2005, 15.71% of all drivers in all races combined have finished DNF… isn’t this significant?”…. etc., etc.,

    Eventually, confidence levels and confidence intervals will be reported to “list members” like yourself for the more important analysis :), so this will make my hypothetical questions above moot… more-or-less (though reduce the confidence levels/intervals in the results) but until then since I’m flying solo here… and I’m lazy…. This is the best I can do !!!

    So unless we have someone here who currently works for NASA, or some old-fart who once worked for Bell Labs… this is the best I can do,

    So moving forward….

    Please dont’ ask anymore EXCELLENT questions !!! :)

  • gje627

    @Cooper08 said...

    Nice question! I am sure you realize the data is small on this but still interesting. Perhaps it could be broken down by track type so the data will have a slightly larger sample. Great question.

    Another great question, uggh !!!

    Yes, I do have the analysis by track type….

    But since we don’t play fantasy each week by track type, but instead by individual track, this is the reason I’m reporting by track….

    Additionally, see my discussion of Spearman’s Rank Correlation and my statement that there are other advantages of Spearman’s over the more common Pearson’s that I didn’t mention in my previous Talladaga post……

    The Big Kahuna?

    It works well with small sample sizes !!!

    EDIT: Just to add a little more information…. I believe Spearman’s is sufficient for even our smallest sample size, Kentucky (8-races)…. Additionally, and importantly, confidence level/intervals will take care of any problems with sample size….

    So to reply to your comment that “I am sure you realize the data is small on this but still interesting. “

    All I can say is this….

    Yes, the sample may be lower than ideal, but it’s all we have with the limited number of NASCAR races we have to work with, BUT the overall analytical results will STILL BE statistically/mathematically significant/valid…. And, after all, what we are all looking for is accurate final conclusions, that also include any caveats for limitations in the data.

  • Numberoneoutlaw

    So from the results you posted New Hampshire seems to be one of the more predictable tracks. Thanks for sharing!

  • gje627

    @Numberoneoutlaw said...

    So from the results you posted New Hampshire seems to be one of the more predictable tracks. Thanks for sharing!

    You’re welcome.

    But not necessarily more predictable….

    This is the reason for all my caveats…. So please, before staking your house, read everything I’ve written above, and then some….

    The tables above only include approximately 1/5 of the entire research necessary to reach any solid, reliable conclusions….

    Yes, correlation is a necessary and valuable first-step.

    But, the good stuff comes later :)

  • gje627

    And now I’m thinking my posts have gone far beyond the typical fantasy contest/slate posts in these forums, regardless of sport….

    So, like the texts you receive on your mobile phone from [insert name of favorite international lewd dating site here], if anyone feels my posts need to stop, just reply with:

    STOP

  • gje627

    One last thought….

    For anybody not familiar with the science of rudimentary and advanced statistical analysis….

    The science is NOT concerned with proving what is true…..

    But proving (eliminating from consideration) what is NOT true….

    The Null Hypothesis

    So, if anyone is concerned with sample size, various aero packages, DNFs, rookie-drivers, etc., souring the results, please remember this….

    These things are taken care of in the analysis….

    The substantial majority of all true statistical/analytical studies (maybe up to 75%) focuses on making sure the obscure, outlier, disconnected, etc. events/data are accounted for, and dismissed, before reaching any real conclusions (or maybe more accurately estimation) on what in fact Is likely to be TRUE….

    And at it’s essence all statistics are based first and foremost in probability theory and taking the best information (data) we have and squeezing it for all its riches, while discarding all the chaff (see quote: “to separate wheat from chaff”)…. If my analogy is poor, my apologies….

    Anyways, keep in mind all the potential flaws with data, etc. are also recognized and accounted in the end….

    No more tonight.

    Good-Night….

  • jt77316

    Question for GJE or anyone else for that matter, is there anywhere to find driver salary info from previous years/races? I imagine that is something that anyone with a database may hang on to, but I would like to know if anyone or any site has tracked driver pricing since NASCAR was first offered, or at least since DK went to the six driver format (whenever that was). This is only my second full season. Thanks!

  • Toddalan2286

    The only thing I have found is LineStar DK (so If there are other places I’m all ears too) It lists all the historical optimal lineups only through the 2016 season. It will have driver and driver salary.

  • gje627

    @jt77316 said...

    is there anywhere to find driver salary info from previous years/races?

    I don’t have historical driver salaries and I’m not aware of any website that maintains it either. Sorry.

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

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.

If you or someone you know has a gambling problem, crisis counseling and referral services can be accessed by calling 1-800-GAMBLER (1-800-426-2537) (IL). Gambling problem? Call 1-800-GAMBLER (NJ/WV/PA/MI), 1-800-9-WITH-IT (IN), 1-800-522-4700 (CO), 1-800-BETS OFF (IA), 1-888-532-3500 (VA) or call/text TN REDLINE 1-800-889-9789 (TN).