By Mark Wallis, Co-Host of Weekday Wagers and CEO of JustTheStatsSports.com
There’s a difference between taking an underdog because the number looks big and taking one because the number may actually be wrong.
That’s the idea behind Walking The Dogs. Rather than blindly grabbing points, the goal is to find underdogs where the Just The Stats Sports model sees a tighter game than the sportsbook does.
The updated 2026 results have made one group particularly interesting. JTSS Cover Dogs — teams carrying roughly a 30% to 42% model win probability while receiving at least 1.5 more points than the model’s projected losing margin — have covered at approximately a 70% rate this season.
The goal isn’t necessarily to find teams that win outright. It’s to find teams that get more points than the model thinks they need.
Week 9 gives us three.
Temple +6.5 at East Carolina
Temple isn’t the kind of underdog that jumps off the board, which is exactly why the number is interesting. JTSS gives the Owls a 34.7% chance to win outright and projects East Carolina to win 27.4-23.2. That’s a projected margin of 4.2 points. The market is giving Temple 6.5. That creates more than two points of additional protection beyond the JTSS projection.
Temple’s defense has taken some hits recently, allowing 72 combined points over its last two games, but this wager doesn’t require the Owls to suddenly become dominant defensively. It simply asks whether East Carolina should be laying almost a touchdown in a game the model projects closer to four points. The ATS splits add another layer. Temple has covered 100% of its road games, while East Carolina has covered just 33.3% of its home games.
East Carolina can win and Temple can still cash.
That’s exactly what we’re looking for.
The play: Temple +6.5.
Colorado State +6.5 at Utah State
This may be the cleanest model-versus-market discrepancy of the three. JTSS gives Colorado State a 39.6% win probability and projects Utah State to win 31.9-28.7. That’s only a 3.2-point game. We’re getting 6.5. That’s more than three full points of additional cushion, and the expected game environment makes those points even more attractive.
With the total sitting around 63.5, this has the makings of a game where both teams get opportunities to score. In that kind of matchup, nearly a touchdown with the underdog carries real value. Colorado State doesn’t have to control the game. It just has to keep answering. Utah State can win by three, four, five or six and the ticket still cashes.
When the model sees barely more than a field goal separating the teams and the market is offering 6.5, the difference becomes difficult to ignore.
The play: Colorado State +6.5.
South Carolina +5.5 at Oklahoma
This one comes with an important qualifier:
Shop the number.
South Carolina was available at +5.5, while much of the market had already moved to +4.5. That point matters. JTSS gives South Carolina a 37% chance to win outright and projects Oklahoma to win 23.9-20.5. That’s a model margin of only 3.4 points. At +5.5, South Carolina gets 2.1 points of additional cushion beyond the projection and fits the Cover Dog profile. At +4.5, that cushion obviously shrinks.
The matchup also gives South Carolina a path to staying close. Quarterback LaNorris Sellers can create outside the structure of the offense when pressure arrives, while the Gamecocks’ pass rush gets an Oklahoma offensive line that has been dealing with injuries. Then there’s the total.
At roughly 44.5, every point becomes more valuable. Scoring 5.5 points in a game expected to stay in the low 20s is very different from doing the same in a shootout. South Carolina also enters with confidence after beating Tennessee. The model still makes Oklahoma the winner.
It just doesn’t see the Sooners winning by much.
The play: South Carolina +5.5.
The Week 9 Card
The Week 9 Walking The Dogs card:
Temple +6.5
Colorado State +6.5
South Carolina +5.5
All three fit the updated JTSS Cover Dog profile, and all three are receiving more points from the sportsbook than the model projects they should need. That’s the entire concept. We’re not asking these teams to win the damn game. We’re asking them to stay inside a number that may be too large. The sportsbook sets the spread.
The model tells us whether it makes sense.
And when the gap gets big enough, sometimes it’s time to grab the leash and walk the dog.
Model projections, historical model data and 2026 Cover Dog results: JustTheStatsSports.com. Odds and market numbers referenced were current at the time of publication and are subject to change.
Leave a comment