Skip to main content
Educational analyticsNot betting advice · research use only.
GameTime Picks
Today · Oct 10 · 19 eventsGames under waySettled· Oct 7

← NFL hub

Regular season · week 5 · Sunday, Oct 11, 8:20 PM ET

Baltimore Ravens at Atlanta Falcons

Baltimore RavensAtlanta FalconsSaved on this device
Research matchup

67°F · 10 mph wind from the W · 1% chance of rain · partly cloudy — the schedule carries no roof for this stadium, so this is the open-air forecast — not used by the model; no weather term enters the numbers below.

Regular season · week 5 · Sunday, Oct 11, 8:20 PM ET · Mercedes-Benz Stadium

Baltimore Ravens at Atlanta Falcons

Baltimore Ravens51.9% to win

Projected score

24–23

the middle of the outcomes, not a call on the exact final

Atlanta Falcons44.6% to win

Baltimore Ravens are the projected winner, 51.9% to win.

Expected margin

ATL -1

8 in 10 outcomes between -17 and +16

Expected total

47

8 in 10 outcomes between 30 and 64

Likeliest exact final

BAL 17–20 ATL

0.8% of outcomes — exact scores are rare

Model status: Under forward evaluation

These forecasts come from models that passed development tests on past seasons and are now being scored on games played after each forecast was frozen. Until enough of those games are graded, every number here is an estimate, not a validated pick, and none of it is a guarantee. Results are published as games are graded. Player results published so far grade the player-board model's own pre-kickoff ranges, each under the model and version that made it. From Week 5 the game pages and Weekly leaders show World Model V2's passing, rushing and receiving yards and receptions instead; those projections are frozen before every kickoff but are not graded in these results, so none of these results describe them.

Not in this number: This forecast does not account for Lamar Jackson (BAL, Out). Our team model rates teams from past results and does not adjust for who plays quarterback, so it is not compared with the sportsbooks for this game — their price already reflects the absence.

Live game state

Live beta

The left side is what the live source reports right now. The right side is the GameTime forecast made before kickoff — it is frozen and does not change while the game is played.

Live now

Checking the live feed…

Pregame GameTime · frozen

BAL
24 (52% pregame win chance)
ATL
23 (45% pregame win chance)

Forecast frozen at Oct 10, 2:48 PM ET. It does not change during the game.

Simulation story

5 chapters · read from the published artifact
Skip to the report ↓
› Loading the matchup model› Reading the outcome distribution from 10,000 simulated games› Preparing the score scenariosPress Play to step through what the model expects, chapter by chapter, or read the report below.

BAL @ ATL

Baltimore Ravens at Atlanta Falcons at Mercedes-Benz Stadium.

model nfl-regular-season-public-v110,000 simulated gamesproduced Oct 10, 2:48 PM ET

World Model V2 · 10,000 simulated games

Simulated game outcomes

Each simulated game is complete and consistent: the final score, the scoring plays, both offenses and every player line belong to the same game.

Who won the simulated games

BAL 51.7% · ATL 48.1% · went to overtime 3.4% · still tied after overtime 0.2%

This share is counted from the simulated games, which draw the score from the forecast's margin and total; the headline win chance comes from a separate team rating, so the two differ by 3.4 points for ATL.

ATL margin

median -1 · 8 in 10 between -18 and +16

2
≤-28
1
2
-24
3
4
-18
5
6
-12
7
9
-6
10
6
0
10
8
+6
7
5
+12
4
3
+18
2
1
+24
2
≥+27

Share of simulated games by ATL margin, in 3-point bands; a minus means BAL won.

Total points

median 47 · 8 in 10 between 30 and 65

4
≤24
5
7
30
11
14
40
14
14
50
12
8
60
6
3
70
2
≥75

Share of simulated games by combined points, in 5-point bands.

How each team scored, on average

Per gameBALATL
Points (median, 80% range)24 (12–36)23 (11–35)
Touchdowns2.72.6
Field goals1.81.9
Passing yards249230
Rushing yards120139

Player projections · expected statistical summaries · not one simulated game

What each player is projected to do

Passing, rushing and receiving from the simulated games; anytime touchdown from the touchdown model. Players listed as Questionable or Doubtful are marked; players ruled out are not projected.

  1. Tyler Huntley
    Tyler HuntleyQB · BAL vs ATL
    243yds80%: 143–361
  2. Michael Penix Jr.
    Michael Penix Jr.QB · ATL vs BAL
    225yds80%: 128–339
    Sportsbook · DraftKings 218.5 · o -113 / u -111

Projected = median passing yards across the simulated games; 80% of them fall in the range shown · World Model V2 · as of 2026-10-09 18:36 UTC

Model detail — who this forecast leaves out (6)
  • Tyler Huntley BAL — Not this team's passer: backup quarterback — the passing projection belongs to the depth chart's starter.
  • Lamar Jackson BAL — Not playing: Listed Out.
  • Ja'Kobi Lane BAL — Not playing: Listed Injured Reserve.
  • Jack Strand ATL — Not this team's passer: backup quarterback — the passing projection belongs to the depth chart's starter.
  • Tua Tagovailoa ATL — Role not yet observed: no game for ATL yet this season — role at ATL not yet observed.Historical prior, not used in these numbers: 16 games for MIA, 2.7 rush yds · 171 pass yds per game.
  • Cooper Rush ATL — Not this team's passer: backup quarterback — the passing projection belongs to the depth chart's starter.

Simulated game explorer

Five of the simulated games, as they were generated

Game #4,376 of 10,000 · Middle of the pack

BAL 30–29 ATL BAL win

BAL 30: 4 TD, 1 FG, 3 XP · 23/32 passing for 267 yds · 24 carries for 98 yds

PlayerPassingRushingReceivingTD
Mark AndrewsMark Andrews TE5-511 rec
Justice HillJustice Hill RB1-11-29
Rashod BatemanRashod Bateman WR2-192 rec
Zay FlowersZay Flowers WR1-09-741 rec
Derrick HenryDerrick Henry RB15-47
Rasheen AliRasheen Ali RB2-23
LaJohntay WesterLaJohntay Wester WR0-0
Tyler HuntleyTyler Huntley QB23/32, 2677-504 pass
Chris MooreChris Moore WR1-31
Matthew HibnerMatthew Hibner TE1-7
Elijah SarrattElijah Sarratt WR2-33

ATL 29: 3 TD, 2 FG, 1 XP, 1 two-point, 1 safety · 14/26 passing for 143 yds · 26 carries for 173 yds

PlayerPassingRushingReceivingTD
Kyle Pitts Sr.Kyle Pitts Sr. TE0-0
Drake LondonDrake London WR4-521 rec
Bijan RobinsonBijan Robinson RB9-66
Charlie WoernerCharlie Woerner TE1-11
Michael Penix Jr.Michael Penix Jr. QB14/26, 1433 pass
Chris BlairChris Blair WR0-0
Austin HooperAustin Hooper TE1-31 rec
Olamide ZaccheausOlamide Zaccheaus WR2-13
Jahan DotsonJahan Dotson WR2-221 rec
Brian Robinson Jr.Brian Robinson Jr. RB13-81
Zachariah BranchZachariah Branch WR4-41

A real game from this run, shown as generated. Plays by players outside the named group count in the team line only. Who scored in one game is not a touchdown probability.

Exact score

The likeliest final scores

from an event-based simulation of touchdowns and field goals, solved onto the median margin and total above

Score (BAL – ATL)Chance
17–200.80%
23–200.78%
27–240.74%
20–170.68%
24–270.67%
20–230.63%

No single scoreline is likely — these are the most common of hundreds. The projected score above is derived from the medians and answers a different question: the middle of the distribution, not its most common point.

Key numbers

How often it lands on 3, 7, 10 or 14

football's margins pile up on those four — this is the one thing a normal-draw model cannot show you

MarginThis gameLast 3 regular seasons
Decided by exactly 310.6%14.2%
Decided by exactly 76.9%8.3%
Decided by exactly 105.4%4.2%
Decided by exactly 144.7%4.4%
Any of the four27.6%31.1%

The right-hand column is what actually happened across 816 regular-season games (2023, 2024, 2025) — the same finals this engine was fitted to. It gets the shape right, with 3 the most common margin by a distance, and it is not calibrated to the number: it puts less weight on 3 and more on 10 than those seasons did.

BAL touchdowns
2.6
ATL touchdowns
2.5
BAL field goals
1.8
ATL field goals
1.8
Tied after regulation
3.9%
Comparison

Us versus the sportsbooks

two independent reads, shown side by side

This forecast does not account for Lamar Jackson (BAL, Out). Our team model rates teams from past results and does not adjust for who plays quarterback, so it is not compared with the sportsbooks for this game — their price already reflects the absence.

Methodology

How these numbers are made

Sources, versions, timestamps and evaluation status
NumberModelVersion / runProduced
Win chance, projected score, margin, totalGame Time Forecastnfl-regular-season-public-v12026-10-10 18:48 UTC
Simulated outcomes, sampled games, passing / rushing / receiving yards, receptionsWorld Model V22.2.0 · run 8bfd9b6a4ce5b82e · 10,000 games2026-10-09 18:36 UTC
Anytime touchdownTouchdown model (player board)opportunity model2026-10-10 18:48 UTC

The Game Time Forecast rates both teams from past results and play efficiency and produces the win chance and the score centre. World Model V2 simulates 10,000 complete games: each one draws a final score around that centre, splits it into touchdowns, kicks and safeties, gives each offense its passes and carries, and hands those to the players expected to be active — so in every simulated game the passing yards equal the receivers' yards and every touchdown has a ball carrier or receiver. Player projections are counted from those games. From the margin-of-victory Elo win head, published exactly as evaluated on held-out 2006–2021 — no shrink toward 50% is applied, and no claim to out-predict the sportsbook market is made. Until kickoff a later run may update these numbers when an input changes (each version is kept); at kickoff the last version is frozen.

Inputs as of: forecast 2026-10-09 18:36 UTC · injuries 2026-10-09 18:36 UTC · rosters 2026-10-09 18:36 UTC · player history through the last completed week (2026-week05-20261009T0100Z.json).

Quarterbacks: BAL Tyler Huntley · ATL Michael Penix Jr. (depth chart; the first quarterback not ruled out throws).

Tested on past seasonsYesplayer receptions, receiving / rushing / passing yards from game worlds: development look PROCEED_TO_FORWARD_SHADOW (2022–2025)
Frozen before kickoff and recordedYesWeek 5 2026 candidates frozen before kickoff (private captures 2026-10-08T18:21Z, T18:27Z)
Graded on games played sinceNot yetno graded forward sample yet
Eligible for picks productsNot yetneeds forward evaluation against the preregistered bars
Promoted as the official forecastNot yetthe published game forecast remains the forecast of record

Results: Player results published so far grade the player-board model's own pre-kickoff ranges, each under the model and version that made it. From Week 5 the game pages and Weekly leaders show World Model V2's passing, rushing and receiving yards and receptions instead; those projections are frozen before every kickoff but are not graded in these results, so none of these results describe them.

Not published:

  • anytime td: Touchdown scorer probabilities from these worlds failed their evaluation (log loss 0.5138 vs 0.5109 for the red-zone model; calibration error 0.034). Diagnosis: a scorer must touch the ball in the same world, which double-counts opportunity and moves touchdowns from part-time to every-down players. Not published as probabilities; anytime-touchdown chances shown on the site come from the separate touchdown model.
  • passing td: Passing touchdowns are credited to the passer in every world, but no evaluation of passing-touchdown distributions exists. Not published as probabilities.
  • interceptions, sacks, kicking and defensive player stats: Not simulated.
  • First touchdown scorer: needs the order of scoring plays, which the simulated games do not record.

Known limitations:

  • Scores come from the published margin and total heads; they do not see injuries or who plays quarterback.
  • Simulated final scores do not cluster on 3- and 7-point margins the way real NFL scores do, so exact-score frequencies from these games are not published.
  • The win chance here is counted from these simulated games and can differ from the published win chance, which comes from a separate rating.
  • Overtime outcomes are drawn from the 2017–2025 record, not simulated play by play.
  • Player volume does not yet adjust to an opposing defense beyond the team-level volume fit.
  • Red-zone history ends with 2025; 2026 games add no red-zone evidence.
  • Plays by players outside the named active set are counted in team totals only.

Availability used: Lamar Jackson (BAL, Out, not simulated); Zay Flowers (BAL, Questionable, simulated as playing).

20,000 simulated team-games checked for consistency, 0 inconsistent.

Simulation V2 · experimental · separate from this forecast

See 10,000 coherent simulated games of this matchup

Reading key

What these numbers mean

Projected score
The middle outcome across every simulated game — not a prediction of the exact final.
Win chance
The model's chance for each side to win, from its team-strength rating — not a count of simulated wins. The simulated games show their own win share, and the difference, in Simulated game outcomes.
80% range
Eight in ten simulated games landed inside this band. Real games land outside it too.
pp (percentage points)
The plain difference between two percentages. A gap is a difference, not an advantage.
Under forward evaluation
Published while its record on games played after each forecast was frozen is still accumulating. Every forecast is frozen before kickoff and graded against the official result.
Provenance

Where this came from

model
nfl-regular-season-public-v1 v2
simulations
10,000
input hash
f6a5f2c3e33ce4e4
generated
2026-10-10T18:48:02Z
kickoff
2026-10-12T00:20Z
state
UPCOMING
Player-family model provenance (3 published)
  • Receiving yards: props-v1 evaluation: n=3575, all promotion bars pass
  • Receptions: Share-level model: cleared every bar when re-tested on 2014–2021 (27,134 player-games) — a second look at seasons examined once before, not a blind test. Each week's forecast is frozen before the first kickoff and graded; its blind 2026 record is holding every bar (586 graded player-games).
  • Anytime touchdown: Opportunity touchdown model: cleared every bar on a blind test of 2014–2021 (35,128 player-games never used to build it). Probabilities condition on playing — a player who does not play settles void. Each week's forecast is frozen before the first kickoff and graded; its blind 2026 record is still accumulating (782 graded player-games so far).

Our win chances come from a team rating that weights how decisively teams win. Tested on 4,281 past games it had never seen (2006–2021), it scored 0.629 on log loss (lower is better) — better than our previous rating's 0.642 and a coin flip's 0.693, but short of the sportsbooks' own odds at 0.610. It has NOT been shown to beat the sportsbook market. Where our win and margin ratings disagree on the favourite, the previous rating is used for that game. The 2026 season evaluation still runs under bars frozen in August, so every forecast stays experimental. These ratings are built from results, not from who is playing: they do not know that a starting quarterback has been ruled out. Our player projections do remove players who cannot play; the team win chance beside them does not. We tested whether adding that would help, using perfect hindsight about whether each team's main quarterback actually played — on 855 games from 2023 to 2025 it improved the win rating on average but was WORSE in one of the three seasons, so it did not clear the bar we hold these ratings to and we left the rating alone.

Educational and paper-only — not betting advice.

More

Other games with a simulation