Baltimore Ravens at Dallas Cowboys. Lean: Baltimore Ravens -3.5. Lean: Under 53.5.

Spread.The strongest explanation for Baltimore -3.5 is its rushing matchup against Dallas' struggling, depleted defense, with no Dallas home-field advantage in Rio. The captured move from -3 to -3.5 at several books is understandable. Nevertheless, I lean Dallas +3.5 at BetMGM's -112. Baltimore's protection problems predate Stanley's absence: its rebuilt line ranks 30th in pass-block win rate, and Carson Vinson now faces his first start. Flowers' questionable status adds another offensive limitation. Prescott gives Dallas a credible route to staying within a field goal without requiring its defense to become good. Our calibrated Baltimore -7.5 looks excessive relative to the four-point rating difference and does not incorporate these personnel constraints. This remains a LOW lean because Dallas' defensive absences and Baltimore's returning defenders can outweigh that protection advantage. I would accept the captured hook, but the conflicting matchup advantages do not justify a normal unit. Confidence is ordinal, 2/5, not a probability.

Total.The strongest case for 53-plus is efficient quarterback play against vulnerable defenses, particularly Dallas' depleted secondary. New coordinator Declan Doyle's aggressive approach also challenges a slow-game assumption. The under case is more specific: Baltimore's protection losses can interrupt drives, Flowers remains uncertain, and returning Baltimore defenders could improve resistance against Prescott. Applying the dossier's stated 1.2-point bias correction moves 52.02 to 50.82, but that is context from an inferior model, not a reliable fair total. I lean under 53.5 at DraftKings -112 after the captured rise from 51.5. Light forecast wind supplies no meaningful under adjustment, and the field passed inspection. Dallas' offensive efficiency and defensive injuries leave too much shootout risk for a normal-unit bet. Confidence is ordinal, 2/5.

Where our number looks wrong.Baltimore -7.5 overstates my assessment of the separation. Calibration expands a four-point rating difference while ignoring Stanley's absence, Flowers' hamstring and a rebuilt offensive line. The model's 47.6% development ATS performance provides no reason to trust that expansion. Dallas' defensive losses push the other way, but do not independently validate laying more than a touchdown.

• Dallas leads points per drive but ranks 30th in rushing yards allowed under new defensive coordinator Christian Parker.
• Indexed DraftKings snapshot shows Baltimore -3.5 (-105), Dallas +3.5 (-115), and Dallas receiving 59% of spread handle; snapshot timing is unspecified.
• Baltimore's touchdown production fell to three scores on 16 drives after Flowers' opening-game exit.
• Doyle described an aggressive third-and-short passing decision and emphasized establishing the run.
• Stanley is out, Flowers questionable, and Madubuike and Buchanan have no game designation.
• Friday inspection involving NFL, NFLPA and independent representation passed mandatory field compliance standards.
• Stadium-specific forecast lists cloudy conditions, 72°F, southwest wind 6 mph and 12% precipitation probability at kickoff.

Spread.Market-first: at a neutral site Baltimore -3.5 says the Ravens are about a field goal and a half better, which fits Lamar Jackson and a top run game against a Dallas team that beat Washington 37-20 but lost Week 1. The line opened 2.5 to 3 and moved to 3.5, so money has come in on Baltimore. Our number is Ravens by 7.5 with no home field, a 4.1-point gap. What the news adds is decisive for me: Dallas rules out cornerback Cobie Durant, both safeties Malik Hooker and P.J. Locke, and linebacker DeMarvion Overshown, on a defense already 29th in run success rate that has allowed 5.0 yards a carry and only two sacks. Baltimore ranks third in EPA per carry and first in run success, and gets Nnamdi Madubuike and Teddye Buchanan back on defense. The Ravens' own loss is left tackle Ronnie Stanley, with a rookie making his first start, but a Dallas front with two sacks in two games is the softest place to hide him.

Total.The total opened 51.5 to 52.5 and has risen to 52.5 to 53.5, the market pricing a depleted Dallas secondary and two good quarterbacks. Our totals model is 52.0, about 50.8 after its known high bias, so it sits a little under the market. The under case beyond the model: Baltimore's edge is a run-first game plan that shortens the game, the humid low-80s conditions wear on both sides, and a rain-softened grass surface can slow cuts and the passing game. The over case: Dallas' secondary is missing three starters, Dallas scored 37 last week, and neither defense is getting to the quarterback. I lean under 53 because the game script I expect is Baltimore running, but the Dallas secondary absences keep it a lean.

Where our number looks wrong.Our 7.5 is probably too high: the rating still carries Baltimore's prior while the Ravens have a new head coach and lost at home to New Orleans, and it does not see Stanley out. Fair is closer to 5 or 5.5 after the Dallas secondary losses, which still clears 3.5.

• Maracana field passed its mandatory tests Friday but remains under scrutiny; rain not expected during the game, possible afterward
• Total 52.5; under pick; cloudy low 80s, 75% humidity, outdoor grass potentially imperfect
• Humid start, rain possible late

Spread.Dallas is the nominal home team only because the NFL needed a home designation for its first Rio de Janeiro game - this is a true neutral site with zero home-field value baked in, per the dossier's own venue note, at Maracana. That reframes the number as a straight team-quality line, and the market has moved accordingly: the Ravens opened around -2.5/-3 and have firmed to -3.5 with sharp money reported at 71% of dollars on Baltimore against only 47% of tickets, a classic sharp-side signal. My own read agrees with that money: Dallas's defense was porous in 2025 (+0.166 EPA/play allowed, its worst unit) against a Lamar Jackson offense that finished positive on both sides of the ball, and Dallas carries its own attrition (Overshown, Durant, Hooker, Locke all out on defense) into a game away from its practice environment, on an unfamiliar field, in warm and humid conditions.

Total.The total has climbed all week - from an open near 51.5-52 to 53-53.5 now - with reports of 69-77% of bets and money on the over, and that move already reflects the same shootout logic I would otherwise cite: two offenses that graded positively in EPA/play in 2025 (Baltimore +0.048, Dallas +0.094) paired with defenses that both graded well below average, at a hot, humid site (RealFeel near 93F by day, cooling to the 70s by the 4:25pm ET kickoff) with a modest chance of rain arriving late. There is no fact here the market has not already priced - the movement itself is the public information, and I am not comfortable buying an already-inflated number on the same crowded side.

• Total moved from an open near 51.5-52 to 53-53.5 with heavy over-side money
• Maracana turf still being trimmed/prepped days before kickoff