Every NFL game this week with its spread, moneyline, game total and, most usefully for lineup building, the implied team total behind each side. The board is built from the consensus schedule feed and refreshed weekly, so it does not go stale between seasons.
Why implied team totals matter more than the spread. A game total tells you how many points the market expects on the field. The implied team total splits that between the two offenses, and that split is what actually drives fantasy scoring. A 27-point implied total on a road underdog is a better stacking spot than a 44-point game total on its own would suggest.
Log in daily to earn Bacon Bits toward free tool passes. Log in or join free
How Our Projections Are Actually Doing
The board above is what we expect. Below is what happened. Every completed game carries its final score beside the projection, the actual DraftKings and FanDuel points every projected player scored, and a grade for how well the model ranked them.
Grades use the correlation between projected and actual points rather than raw error, because ranking players correctly is what matters for a lineup, and a game where everyone outscored expectations should not be marked down for it. A single game is a small sample, so the season figure is the number worth reading.
How the projections have actually done. Across 64 completed games and 1386 player scores this season, mean absolute error is 4.31 DraftKings points with a bias of -0.39, correlation between projected and actual is +0.703, and that is +36.2 percent better than simply projecting every player at the slate average. Season grade B. A positive bias means the projections run high. A single game is far too small a sample to judge, which is why the season figure leads.
Projected 20.7 to 18.1 for a total of 38.7, against a market total of 38.5. Actual total 51.
3.56Mean abs error
-2.15Bias
+0.815Correlation
+45.8 percentvs baseline
23Players graded
Player
Proj DK
Actual DK
Diff
Actual FD
Own
Aaron RodgersPIT QB
14.5
24.0
+9.5
24.0
0.4%
Quinshon JudkinsCLE RB
12.4
21.6
+9.2
18.6
1.6%
DK MetcalfPIT WR
11.3
19.5
+8.2
14.0
1.9%
Roman WilsonPIT WR
5.8
16.4
+10.7
14.9
0.3%
Jaylen WarrenPIT RB
18.9
15.6
-3.3
14.1
25.9%
Deshaun WatsonCLE QB
15.5
14.9
-0.6
13.9
1.4%
Denzel BostonCLE WR
8.8
12.9
+4.2
10.9
0.1%
Darnell WashingtonPIT TE
4.8
11.7
+6.9
10.2
0.1%
Harold Fannin Jr.CLE TE
10.7
11.7
+1.0
10.2
3.0%
KC Concepcion Jr.CLE WR
10.2
11.4
+1.2
8.9
13.2%
How to read the board
Games are grouped by day in kickoff order. The favorite on each card is picked out in gold, and the chip in the top right tags the game shape: a grind at 42 points or under, a shootout at 49 or over, average in between. Those are the same bands the FTA showdown research uses, so the language carries across the site.
Win probability is de-vigged from the two moneylines, so the pair adds to 100 rather than the 104 or so the raw prices imply. The second tab ranks all 32 implied team totals on the slate, which is the fastest way to find the offences the market likes before you start building.
What this pairs with
Once you have the spots, the NFL Strength of Schedule tool grades the matchups, the NFL Bacon Station carries defense-allowed and red zone usage, and the BaconLab optimizer builds the lineups. FTA+ members can export the whole board to CSV and drop it straight into their own model.