Ravens vs Cowboys: Rio Neutral-Site Showdown
Ravens vs. Cowboys: Baltimore's Ground Game Is the Edge in Rio
The NFL's international slate lands at MaracanΓ£ Stadium this week with a game that looks like a mismatch on paper and a toss-up on the scoreboard. Baltimore (1-1) is the designated road team, Dallas (1-1) the nominal home side, but neither team gets a real home-field edge in Rio. That matters, because this line is built almost entirely on the teams themselves β and the Ravens are the better team.
Baltimore's Offense Is Built for This Matchup
The Ravens are averaging 29.0 points and 434.5 yards per game, and the shape of that production is what should worry Dallas. Baltimore runs for 155.0 yards per game at a healthy clip and has already scored 2.5 rushing touchdowns per game β five in two outings. Derrick Henry is chewing up 106.0 yards per game at 5.3 yards per carry, and that's the single best matchup on the board.
Dallas is allowing 24.0 points per game and just 1.0 sack per game. The Cowboys lost cornerback Cobie Durant, which thins out a secondary that was already the soft spot of this defense. When you can't generate pressure and you're down a cover man, the answer is to lean on the run β and Baltimore has the personnel to do exactly that.
Lamar Jackson's raw numbers are quieter than usual: 559 yards, 2 touchdowns, 1 interception, 67.9 percent, 104.7 passer rating. The yardage is fine at 279.5 per game; the scoring through the air is not. Only one passing touchdown per game is why Baltimore sits at 1-1 despite out-gaining both opponents. The good news is he hasn't needed to force it β he's added 74 rushing yards and a score on the ground, and Henry is carrying the red-zone load.
Dallas Can Move It β Until the Run Game Kills a Drive
Dak Prescott has been genuinely excellent. 454 yards, 6 touchdowns, 1 interception, 73.8 percent completion rate, 117.1 rating β and Week 2 against Washington was close to flawless: 279 yards, four touchdowns, no picks, 83.9 percent completion. That's the version of Prescott that makes this spread uncomfortable.
CeeDee Lamb is doing Lamb things: 6.5 catches for 98.5 yards per game at 15.2 yards per reception. George Pickens (4.5 rec, 34.0 yds) and Jake Ferguson (3.0 rec, 24.5 yds) give Prescott enough chain-movers underneath.
The problem is everything after that. Dallas is averaging 294.5 total yards per game β that's 140.0 fewer than Baltimore's offense, and the gap is entirely on the ground. The Cowboys rush for just 67.5 yards per game. Javonte Williams is at 35.5 yards per game on 3.0 yards per carry. A 3.0 YPC back as your lead runner means second-and-long, which means Prescott throwing into obvious passing downs, which means Baltimore's 2.5 sacks per game pass rush gets to pin its ears back.
Dallas is running 52.3 percent pass on 56.6 plays per game. Baltimore runs 61.8 plays per game at a nearly identical pass rate. The Ravens play faster and have a functional run game. That's the whole story of the pace and possession battle.
The Defensive Side of the Equation
Baltimore's defense is allowing 23.5 points per game with 2.5 sacks and 0.5 interceptions per contest. Not elite, but functional and disruptive. The concern is real, though: Baltimore has now blown a fourth-quarter lead, and the revamped coaching staff made some questionable in-game decisions in the Week 2 loss to New Orleans. That's a legitimate fade angle β this team has a habit of letting opponents hang around.
Dallas has been better defensively than the raw scoring suggests, but it's still leaky, and losing Durant takes away the one thing a thin secondary can't afford to lose: bodies. The Cowboys are averaging just 1.0 takeaway per game, and their plus-0.5 turnover differential is doing more work than the actual defense. Baltimore's differential sits at 0.0, but with Lamar's 1 interception in 56 attempts, the Ravens aren't handing the ball away.
The Market and the Model
The number has moved. Baltimore opened at -3.0 and is now -3.5, and the total has climbed from 51.5 to 52.5. Both moves point the same direction: money likes Baltimore and expects points.
The model isn't buying the total. Its projected score is Ravens 29, Cowboys 21 β an eight-point margin, well clear of the 3.5, but a combined 50 points, which lands under the 52.5. That's the interesting tension here. The market is pricing a shootout; the model sees a Baltimore win with a slightly slower clock than the public expects, driven by a Ravens run game that chews possessions.
ATS Pick: Baltimore -3.5. Moneyline: Baltimore (-178). Confidence: 0.68.
The moneyline price is fair. -178 implies about 64 percent, and the model's confidence sits at 0.68 β a small but genuine edge. Dallas at +150 is the value side if you believe Prescott stays hot, but you're betting on a team that can't run the ball and just lost a starting corner.
Situational Notes
Rest is a wash β both teams are on normal weeks. Both are 1-1, both are 1-1 in conference, and neither team has a division game on the line here. The field conditions at MaracanΓ£ drew some noise earlier in the week, but the NFL's field director has said there are "no concerns right now," and this is a grass surface with no weather data of note to factor.
One thing worth flagging: Dallas is technically the home team, but this is a neutral-site international game. The Cowboys' 1-0 home record is meaningless. For Baltimore, the 0-1 "home" record is equally meaningless β they're 1-0 on the road.
Worth a small nod to special teams: Tyler Loop has been a bright spot for Baltimore, but the rest of the unit has been shaky. In a game where the spread is 3.5, a special-teams lapse is exactly the kind of thing that burns a favorite.
The Verdict
Baltimore runs for 155 yards per game against a Dallas defense that can't pressure the quarterback and just lost a corner. Dallas runs for 67.5 yards per game at 3.0 yards a carry, which puts Prescott in obvious passing situations against a Ravens front averaging 2.5 sacks. The Cowboys will score β Prescott is playing too well not to β but the path to a Dallas cover requires him to be nearly perfect for four quarters with no run support.
Take Baltimore -3.5, with the model projecting an eight-point win. If you're playing the total, the model's 50-point projection against a 52.5 line points under, even though the market has been creeping up all week. Confidence: 0.68 on the Ravens.
