Unofficial work sample. Not affiliated with PrizePicks. Lines and results are real; players and entries are simulated.

Game Ops Console · NFL

Line Recap · Weeks 1–2

Every PrizePicks standard line from Weeks 1–2 (31 games, 1,001 lines), recorded three days, one day, and minutes before kickoff, then graded on the final box score. It shows how many lines were beatable and whether they were fixed in time.

Real dataLines: The Odds APIResults: nflverseHow this is computed →

A beatable line's week

  1. T-72hFirst snapshot, three days out

    Each PrizePicks line is recorded next to whatever sportsbook prices exist at that moment.
    Then:
  2. T-24hOne day out, 47 lines beatable

    Their fair side went on to win 58.5% of the time. Every other line: 50.7%. The warning was there a day early.
    Then:
  3. Kickoff13 of 47 fixed, 70 beatable

    Only 13 of the 47 moved toward the books. The count still grew to 70, because new gaps opened in the last 24 hours. Those 70 won on their fair side 53.1% of the time.

Lines that moved between one day out and kickoff

Beatable one day outAll other lines

About 52% of yardage lines moved at all in the last day, but the moves weren't aimed at the beatable ones. Receptions is the clearest case: 4 of 26 beatable lines moved, against almost none of the rest. Lines do move; the moves just aren't aimed at the lines that leak money.

Is the fair chance right?

The best test is goblin lines (easier lines that pay less) and demon lines (harder lines that pay more), because their fair chances range from 18% to 71%.

Line typeStatLinesWonFair chance
demonRec Yds1,40427.4%26.6%
demonRush Yds67526.5%26.5%
demonReceptions64217.3%18.2%
demonPass Yds24440.6%39.8%
goblinRec Yds98468.3%67.4%
goblinRush Yds46267.3%67.5%
goblinReceptions31868.2%71.2%
goblinPass Yds24670.7%71.4%

Across 4,975 goblin and demon lines, results landed within 0.8 points of the fair chance on average. The one gap is Receptions goblins (68.2% won vs 71.2% fair), where whole-number lines and ties interact.

Standard lines: a weaker test
Actual More win ratePerfect match
40%50%60%35%45%55%65%Fair chance of More at kickoff

885 of 924 standard lines sit in the two middle groups, so there's little range to test. Those landed at 47% (fair 51%) and 48% (fair 48%), in line with More's overall 48%. The outer groups hold only 22 and 17 lines, too few to read either way.

Did lines lean More or Less?

StatLinesMiddle lineActual minus lineAverage missMore wonFair More
Rec Yds35627+4.220.148.6%50.0%
Receptions2193.5+0.11.648.0%50.1%
Rush Yds17329.5+1.717.345.7%49.9%
Rush+Rec Yds9165.5+1.424.446.2%50.2%
Pass Yds62221−1.161.948.4%49.6%
Pass TDs261.5+0.11.045.5%49.4%

No stat leans. More won 48.0% of 974 graded lines against a fair 50%, a gap of 1.2 standard errors (small enough to be chance). The week-to-week swing (51.5% in Week 1, 44.8% in Week 2) looks bigger than it is: lines in the same game ride one game script, so a week is a smaller sample than its line count suggests.

Watch item: low Rec Yds lines
More win rate by line level within each stat (lines in parentheses)
StatLow lineMiddle lineHigh line
Pass TDs29% (9)56% (9)50% (8)
Pass Yds62% (21)38% (21)45% (20)
Rec Yds55% (134)49% (133)47% (133)
Receptions46% (75)43% (75)51% (74)
Rush Yds46% (61)41% (61)48% (61)
Rush+Rec Yds47% (36)43% (35)54% (35)

Each stat's lines split into thirds by size (a stand-in for player role). Low Rec Yds lines (role players) went More 55% on 134 lines, the largest lean. That's only about 1.2 standard errors, so it's one to track for 4+ weeks, not act on yet. Blue cells went More more than half the time; orange cells went Less.

Also checked: ties and Underdog lines

About 1 in 5 whole-number lines tied.

About half of Receptions and Pass TD lines were whole numbers. A tie voids the pick and drops the entry to a smaller payout, so a 6-pick becomes a 5-pick. Worth sizing, and worth testing .5 lines on low counts.

Ties by stat
StatWhole-number linesOf those, tied
Pass TDs54%29%
Receptions49%21%

Yardage lines matched Underdog only 41%60% of the time.

Every mismatch at kickoff lets a player with both apps take the better number. That's the line-shopper group on Player Groups.

Underdog match by stat
StatLinesSame lineAverage gap to books
Rec Yds35160%0.43
Rush Yds16555%0.50
Rush+Rec Yds8641%0.91
Pass Yds6244%1.52
Receptions45100%0.00
Pass TDs7100%0.00

Most beatable lines at kickoff

WeekGamePlayerStatPrizePicksBooksFair side · chanceActualResult
2NYG @ LATutu AtwellRec Yds3.56.5More 70.1%19Fair side won
1CLE @ JAXTravis HunterRec Yds7.511.5More 64.6%8Fair side won
1DAL @ NYGKaVontae TurpinRush+Rec Yds15.513.5Less 63.4%8Fair side won
1DEN @ KCAdam TrautmanRec Yds3.54.5More 59.8%10Fair side won
1TB @ CINCade OttonReceptions33.5More 59.5%3Tie (void)
1TB @ CINTed Hurst IIIRec Yds17.522.5More 59.5%36Fair side won
1DAL @ NYGTyrone Tracy Jr.Rush Yds19.516.5Less 59.3%14Fair side won
2NYG @ LABlake CorumRec Yds3.52.5Less 58.8%13Fair side lost
1DEN @ KCJ.K. DobbinsRec Yds3.54.5More 58.4%0Fair side lost
1BUF @ HOUKayshon BoutteReceptions21.5Less 58.4%1Fair side won
1CLE @ JAXBhayshul TutenReceptions21.5Less 58.4%1Fair side won
2CAR @ ATLJonathon BrooksRush+Rec Yds35.532.5Less 58.1%11Fair side won