Pick Sheet — 2026 Week 19

Enter the confidence number next to each game. Highest number = most confident. Strategy: Always-favourite (baseline).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

No games found for 2026 Week 19.

How often the favourite loses outright

Every 1999–2026 regular-season and playoff game with a closing spread: favourites lost 33.4% of 7,261.

Fav byGamesLostLoss % 95% rangeSince 2015
1 546 266 48.7% 45–53% 51.9% (206)
1.5 207 98 47.3% 41–54% 45.7% (105)
2 169 71 42.0% 35–50% 43.8% (80)
2.5 532 247 46.4% 42–51% 45.8% (312)
3 1,158 484 41.8% 39–45% 39.7% (438)
3.5 684 248 36.3% 33–40% 36.5% (310)
4 342 116 33.9% 29–39% 44.4% (133)
4.5 269 91 33.8% 28–40% 34.0% (106)
5 165 49 29.7% 23–37% 28.6% (63)
5.5 288 99 34.4% 29–40% 35.2% (142)
6 355 112 31.5% 27–37% 30.1% (143)
6.5 341 96 28.2% 24–33% 21.2% (137)
7 492 117 23.8% 20–28% 23.7% (173)
7.5 291 69 23.7% 19–29% 24.1% (166)
8 100 29 29.0% 21–39% 26.8% (41)
8.5 104 26 25.0% 18–34% 27.5% (51)
9 153 37 24.2% 18–32% 31.2% (16)
9.5 162 33 20.4% 15–27% 14.6% (48)
10 213 40 18.8% 14–25% 18.8% (80)
10.5 136 22 16.2% 11–23% 14.1% (64)
11 80 14 17.5% 11–27% 30.8% (26)
11.5 42 10 23.8% 13–39% 20.0% (15)
12 33 8 24.2% 13–41% 14.3% (7)
12.5 47 5 10.6% 5–23% 4.5% (22)
13 73 8 11.0% 6–20% 7.1% (28)
13.5 64 10 15.6% 9–26% 11.1% (27)
14+ 215 19 8.8% 6–13% 9.9% (101)

Ties count as not lost; pick'em lines are left out. Rows past 8 points hold a few dozen to two hundred games, so read neighbouring rows together rather than any one. The rate at a given spread has not changed since 2015: adjusted for spread, the difference is −0.05 points (95% range −1.8 to +1.8).

Tiebreakers

Highest scoring team
—
guess 44 points
right 13% of weeks (random 3%)
Lowest scoring team
—
guess 4 points
right 18% of weeks (random 3%)
GamePosted totalGuess this

Why the guess is under the line. As a forecast the posted total is already the best number available — nothing beats it, and shifting it costs accuracy. But a tiebreaker does not score accuracy, it scores being closest, and if everyone writes down the line then the line is the most contested number on the board. Against a field clustered there, guessing 7 points off turns a 1% win rate into about 30% in a twenty-person pool. It goes under because the median total lands half a point below the line even though the average sits above it, so the under side is the denser one at this distance.

Expect the team picks to miss. The highest-implied team tops the week only 13% of the time and the lowest-implied bottoms it 18%. That is four to six times better than guessing and nothing more. Adding a team's scoring volatility was tested and made both picks worse.

The deviation rests on an assumption that could not be measured here — that the rest of your pool guesses near the line. If they scatter instead, the right distance grows. The direction is the robust part: do not write down the consensus. Study at backtests/pickem_tiebreak/FINDINGS.md.

Why this sheet is boring on purpose. Across 2,119 games (2018–2025) the always-favourite baseline scored 72.39% of maximum points and went 66.4% straight up. Five alternatives were tested against it and none won: a devigged-moneyline ranking ties it at +0.00pp despite disagreeing about the order in 138 of 141 weeks, any re-weighting of spread magnitude is arithmetically the same sheet, and flipping small home dogs is worse by 0.52pp with a 99% interval excluding zero.

So the edge here is not the picks — it is not deviating from them. In a season-long pool the ordinary way to lose is a handful of gut overrides on close games, which is exactly where the market is hardest to beat and where the old “hybrid” strategy on this page used to override it.

The one exception is the flagged game above, and it is an exception about standings, not about football: winning a pool is not the same as maximising points, so a single deviation at the bottom of the sheet buys insurance against tying a field that all picked the same chalk. One game. Never more.