Popular Features

    The Top Machine Learning WR Prospect Will Surprise You

    The Top Machine Learning WR Prospect Will Surprise You
    Anthony Amico March 16, 2020 1:12PM EST

    What Can Machine Learning Tell Us About WR Prospects?

    One of my favorite parts of draft season is trying to model the incoming prospects. This year, I wanted to try something new, so I dove into the world of machine learning models. Using machine learning to detail the value of a WR prospect is very useful for dynasty fantasy football.

    Machine learning leverages artificial intelligence to identify patterns (learn) from the data, and build an appropriate model. I took over 60 different variables and 366 receiving prospects between the 2004 and 2016 NFL Drafts, and let the machine do its thing. As with any machine, some human intervention is necessary, and I fine-tuned everything down to a 24-model ensemble built upon different logistic regressions.

    Just like before, the model presents the likelihood of a WR hitting 200 or more PPR points in at least one of his first three seasons. Here are the nine different components featured, in order of significance:

    1. Adjusted Age
    2. Adjusted Receiving Yards Per Pass Attempt
    3. ESPN Rank
    4. ESPN Grade
    5. Career Yards Per Carry
    6. Receiving Touchdowns Per Pass Attempt
    7. Career Punt Return Touchdowns Per Game
    8. School SRS Rating
    9. Career Kick Return Touchdowns Per Game

    This obviously represents a massive change from the original model, proving once again that machines are smarter than humans. I decided to move over to ESPN grades and ranks instead of NFL Draft Scout for a few reasons:

    • NFL DS has undergone a lot of site changes, and it is unclear just who is creating the rankings
    • Mel Kiper has been a staple of ESPN’s draft coverage for decades, making the ranks and grades more reliable in testing
    • The ESPN inputs are relative to an entire class, whereas NFL DS only provided position ranks

    Those changes alone made strong improvements to the model, and it should be noted that the ESPN overall ranks have been very closely tied to actual NFL Draft position.

    The Top Machine Learning WR Prospect Will Surprise You

    Having an idea of draft position will always help a model since draft position usually begets a bunch of opportunity at the NFL level.

    WR Prospect Model History

    Since the model is built on drafts up until 2016, I figured perhaps you’d want to see the results from the last three drafts before seeing the 2020 outputs.

    corey davisWestern Michigan201771.9%
    d.j. mooreMaryland201856.1%
    christian kirkTexas A&M201838.9%
    n'keal harryArizona State201938.0%
    john rossWashington201735.8%
    andy isabellaMassachusetts201932.8%
    taywan taylorWestern Kentucky201729.1%
    juju smith-schusterUSC201728.4%
    d.j. charkLSU201826.8%
    jj arcega-whitesideStanford201926.4%
    marquise brownOklahoma201926.2%
    dante pettisWashington201824.6%
    deebo samuelSouth Carolina201924.5%
    chris godwinPenn State201721.8%
    calvin ridleyAlabama201820.2%
    carlos hendersonLouisiana Tech201719.4%
    james washingtonOklahoma State201819.3%
    curtis samuelOhio State201718.1%
    greg dortchWake Forest201916.9%
    tre'quan smithCentral Florida201816.4%
    a.j. brownMississippi201914.2%
    zay jonesEast Carolina201713.2%
    isaiah mckenzieGeorgia201713.0%
    malachi dupreLSU201711.2%
    dede westbrookOklahoma201710.7%
    courtland suttonSMU20189.5%
    quadree hendersonPittsburgh20189.4%
    ryan switzerNorth Carolina20178.9%
    josh maloneTennessee20178.8%
    keevan lucasTulsa20178.4%
    mecole hardmanGeorgia20197.6%
    dk metcalfMississippi20197.3%
    jordan lasleyUCLA20186.7%
    michael gallupColorado State20186.7%
    anthony millerMemphis20186.7%
    riley ridleyGeorgia20196.5%
    hakeem butlerIowa State20196.2%
    keesean johnsonFresno State20195.6%
    keke couteeTexas Tech20185.2%
    jerome laneAkron20175.1%
    mike williamsClemson20175.0%
    shelton gibsonWest Virginia20174.8%
    byron pringleKansas State20184.7%
    penny hartGeorgia State20194.6%
    isaiah fordVirginia Tech20174.3%
    kd cannonBaylor20174.1%
    ardarius stewartAlabama20174.1%
    fred rossMississippi State20174.0%
    preston williamsColorado State20193.9%
    deontay burnettUSC20183.7%
    allen lazardIowa State20183.7%
    diontae johnsonToledo20193.5%
    noah brownOhio State20173.5%
    emanuel hallMissouri20193.2%
    anthony ratliff-williamsNorth Carolina20193.1%
    zach pascalOld Dominion20173.0%
    victor boldenOregon State20173.0%
    parris campbellOhio State20193.0%
    equanimeous st. brownNotre Dame20182.9%
    braxton berriosMiami (FL)20182.7%
    olabisi johnsonColorado State20192.3%
    lil'jordan humphreyTexas20192.2%
    rodney adamsSouth Florida20172.1%
    jaylen smithLouisville20192.1%
    miles boykinNotre Dame20192.0%
    deon cainClemson20181.9%
    dillon mitchellOregon20191.8%
    keon hatcherArkansas20171.7%
    artavis scottClemson20171.6%
    auden tateFlorida State20181.6%
    trent taylorLouisiana Tech20171.6%
    travis fulghamOld Dominion20191.6%
    kelvin harmonNorth Carolina State20191.5%
    kenny golladayNorthern Illinois20171.5%
    robert davisGeorgia State20171.4%
    antoine wesleyTexas Tech20191.4%
    tyre bradyMarshall20191.4%
    stacy coleyMiami (FL)20171.4%
    josh reynoldsTexas A&M20171.3%
    ray-ray mccloudClemson20181.3%
    daesean hamiltonPenn State20181.3%
    stanley morgan jr.Nebraska20191.3%
    terry mclaurinOhio State20191.3%
    jehu chessonMichigan20171.2%
    cam phillipsVirginia Tech20181.2%
    travin duralLSU20171.1%
    trey quinnSMU20181.1%
    anthony johnsonBuffalo20191.1%
    chad hansenCalifornia20171.1%
    amba etta-tawoSyracuse20171.1%
    amara darbohMichigan20171.1%
    darius slaytonAuburn20191.0%
    j'mon mooreMissouri20180.9%
    davon graysonEast Carolina20180.9%
    travis rudolphFlorida State20170.9%
    javon wimsGeorgia20180.9%
    damarkus lodgeMississippi20190.8%
    nyqwan murrayFlorida State20190.8%
    deangelo yanceyPurdue20170.7%
    ryan davisAuburn20190.7%
    jeff smithBoston College20190.7%
    terry godwinGeorgia20190.6%
    cody thompsonToledo20190.6%
    hunter renfrowClemson20190.6%
    drew morganArkansas20170.6%
    russell gageLSU20180.5%
    marquez valdes-scantlingSouth Florida20180.5%
    jalen hurdBaylor20190.5%
    jaleel scottNew Mexico State20180.5%
    jester weahPittsburgh20180.5%
    james quickLouisville20170.5%
    gabe marksWashington State20170.5%
    quincy adeboyejoMississippi20170.5%
    bug howardNorth Carolina20170.5%
    jakobi meyersNorth Carolina State20190.5%
    michael rectorStanford20170.5%
    kermit whitfieldFlorida State20170.5%
    jamal custisSyracuse20190.4%
    ricky seals-jonesTexas A&M20170.4%
    jeff badetOklahoma20180.4%
    mack hollinsNorth Carolina20170.4%
    jamari staplesLouisville20170.3%
    tyron johnsonOklahoma State20190.3%
    marcell atemanOklahoma State20180.3%
    damore'ea stringfellowMississippi20170.3%
    johnnie dixonOhio State20190.3%
    dylan cantrellTexas Tech20180.2%

    It is encouraging to see some hits towards the top of the model, but there are obviously some misses as well. Your biggest takeaway here should be just how difficult it is to hit that 200 point threshold. Only two prospects the last three years have even a 40% chance of success. The model is telling us not to be over-confident, and that is a good thing.

    WR Prospect Model 2020 Results

    Now that you’ve already seen some results, here are the 2020 model outputs.

    Tee HigginsClemson53.5%
    Henry Ruggs IIIAlabama48.9%
    CeeDee LambOklahoma48.6%
    Justin JeffersonLSU46.4%
    Tyler JohnsonMinnesota37.8%
    Geraud SandersAir Force31.3%
    Jerry JeudyAlabama26.0%
    KJ HamlerPenn State22.8%
    Brandon AiyukArizona State19.1%
    Jalen ReagorTexas Christian16.1%
    Laviska Shenault Jr.Colorado15.5%
    Quintez CephusWisconsin15.1%
    Michael Pittman Jr.USC11.5%
    Donovan Peoples-JonesMichigan10.7%
    Joe ReedVirginia9.3%
    Denzel MimsBaylor8.8%
    Isaiah HodginsOregon State7.8%
    Marquez CallawayTennessee7.7%
    Antonio Gandy-GoldenLiberty5.7%
    Kalija LipscombVanderbilt4.8%
    Gabriel DavisCentral Florida4.7%
    Chase ClaypoolNotre Dame3.8%
    Bryan EdwardsSouth Carolina3.7%
    Devin DuvernayTexas3.4%
    Binjimen VictorOhio State2.6%
    Trishton JacksonSyracuse2.3%
    Cody WhiteMichigan State2.3%
    Lynn Bowden Jr.Kentucky2.3%
    Omar BaylessArkansas State2.2%
    K.J. HillOhio State2.1%
    Nick WestbrookIndiana1.6%
    Jonathan NanceArkansas1.4%
    Austin MackOhio State1.4%
    Diondre OvertonClemson1.4%
    Bailey GaitherSan Jose State1.3%
    Quartney DavisTexas A&M1.2%
    Collin JohnsonTexas1.1%
    James ProcheSMU1.0%
    Jauan JenningsTennessee0.8%
    Van JeffersonFlorida0.8%
    JaMarcus BradleyLouisiana0.7%
    Seth DawkinsLouisville0.7%
    LaMichael PettwayArkansas0.7%
    Lawrence CagerGeorgia0.6%
    Kendall HintonWake Forest0.5%
    Deshaunte JonesIowa State0.5%
    Derrick DillonLSU0.5%
    Tyrie ClevelandFlorida0.4%
    Darrell Stewart Jr.Michigan State0.3%
    Scotty WashingtonWake Forest0.3%
    George CampbellWest Virginia0.3%
    Juwan JohnsonOregon0.3%

    Tee Higgins as the top WR is likely surprising for a lot of people, but it shouldn’t be. Higgins had a fantastic career at Clemson, arguably the best school in the country over the course of his career. He is a proven touchdown scorer, and is just over 21 years old with a prototypical body-type.

    Nobody is surprised that the second WR on this list is from Alabama, but they are likely shocked to see that a data-based model has Henry Ruggs over Jerry Jeudy. The pair is honestly a lot closer that many people think in a lot of the peripheral statistics. The major edge for Ruggs comes on the ground. He had a 75 yard rushing touchdown, which really underlines his special athleticism and play-making ability.

    The name that likely stands out the most is Geraud Sanders, who comes in ahead of Jerry Jeudy despite being a relative unknown out of Air Force. You can mentally bump him down a good bit. The academy schools are a bit of a glitch in the system, as their offensive approach usually yields some outrageous efficiency. Since 2015, 12 of the top 15 seasons in adjusted receiving yards per pass attempt came from either an academy school or Georgia Tech’s triple-option attack. Sanders isn’t a total zero, his profile looks very impressive, but I would have him closer to a 10% chance of success given his likely Day 3 or undrafted outcome in the NFL Draft.

    Popular Features