BetRivers 200 Preview: Dover Xfinity Picks, Odds & Power Rankings

NASCAR's Xfinity Series rolls into Dover Motor Speedway for the BetRivers 200 on Saturday, May 16 at 4:00 PM ET, 200 laps around the only concrete short oval on the schedule. Practice runs Saturday at 9:30 AM ET with qualifying at 10:35 AM ET.

There's a twist this week: NASCAR is rolling out a new right-side tire compound for the Xfinity field at Dover. That changes how we read our model's tire-rank signal. We'll get into that below. Justin Allgaier sits as the heavy market favorite, but Brent Crews carries the cleanest edge on the board.

The track: Dover Xfinity, with a new compound

The Monster Mile is the schedule's only 1.0-mile concrete oval banked at 24°, passing is at a premium, tire fall-off rules the late runs of each stage, and pit-road timing is amplified by the short window between cycles. Stage lengths are 45 / 90 / 200.

Xfinity will run the D-6106 left-side compound and a new D-6148 right-side compound this weekend. The right-side compound is the same one the Truck Series runs at Dover the day before, which gives us at least a cross-series read on how it behaves at this exact banking and surface. The catch: our model's Xfinity-specific right-tire rank has limited historical depth on this compound, so we're leaning more on track-type rank and recent form than on the tire-combo signal alone.

See the full Dover Xfinity historical track stats: https://fastlap.io/oreilly/track-stats

Top contenders by the book

Justin Allgaier, +325 (book), 20.5 percent Win (model)

The heaviest book favorite of the weekend at +325 (23.5 percent implied). Allgaier has 3 Dover starts and 51 short-track total in our data, the kind of veteran sample that anchors a confident projection. Our sim has him at 20.5 percent to win, a touch under the book's implied 23.5 percent, with a 65.3 percent Top-5 and an 86.5 percent Top-10. Mean DK score of 59.8. The model has him as the most likely outright winner on the board. The slight gap to the implied is just calibration, not a knock on the play.

Run your own BetRivers 200 simulation: https://fastlap.io/oreilly/simulator

Brent Crews, +550 (book), 18.0 percent Win (model)

This is the best price on the board. Crews comes in at +550 (15.4 percent implied), and our model gives him an 18.0 percent chance to win, a +2.6 point edge over the market, the largest positive gap of any priced driver. Top-5 sits at 61.6 percent and Top-10 at 85.3 percent. Mean DK score 59.7, essentially tied with Allgaier.

Crews has 0 starts at Dover and only 4 short-track starts in O’Reilly, so the projection leans heavily on his caliber and recent form rather than track-specific history. If you're hunting for one favorite to overweight against the field this week, this is the line.

Sheldon Creed, +800 (book), 11.1 percent Win (model)

The market and the model are in lockstep here. The book's +800 line implies 11.1 percent, and our sim returns exactly 11.1 percent. Creed sits at 3 Dover starts and 51 short-track total with a 50.3 percent Top-5 and an 80.2 percent Top-10. No edge to chase, but no premium to pay either. Mean DK score of 53.3 with a 34.6 percent Optimal share makes him a tournament anchor at his price.

Get the full Xfinity Series power rankings: https://fastlap.io/oreilly/preview

Three sleepers our model likes

Drivers at longer prices where the long-format finish probabilities and DFS Optimal share justify exposure even with thin Win percent edges.

Sam Mayer, +1800 (book), 5.9 percent Win (model)

The Optimal percent leader of the entire field, 35.6 percent Optimal at a $8,300 salary, the single best DFS price on the board. Model gives him a 5.9 percent chance to win versus the book's 5.3 percent implied, a small but real +0.6 point edge. His 72.9 percent Top-10 at $8,300 is the kind of value the DFS optimizer hugs.

Mayer has 3 Dover starts and 50 short-track total, a solid sample for an under-the-radar play.

Austin Hill, +2000 (book), 3.4 percent Win (model)

Hill brings the most Dover experience of any priced sleeper, 4 starts at the track and 55 short-track total, the deepest historical sample in this price tier. Model has him at 3.4 percent to win with a 51.7 percent Top-10 lock at a $7,700 salary. The Top-10 probability at this price is the play; the outright is gravy.

Sammy Smith, +3000 (book), 2.9 percent Win (model)

The deepest track-experience play of the week, 5 Dover starts and 84 short-track total, the most of anyone in the field. At +3000 (3.2 percent implied), our model gives him 2.9 percent to win, essentially market-priced, with a 48.8 percent Top-10. A longshot with real track equity at an $8,500 salary.

Compare any two of these drivers head-to-head: https://fastlap.io/oreilly/head-to-head

Where our model sees the biggest edges

This is an efficient market. The book has most of the field priced close to where our sim sees them. Two drivers carry meaningful positive Win percent edges this week:

Brent Crews, model 18.0 percent vs. market 15.4 percent implied (+2.6 point edge). The single biggest positive gap on the board.

Sam Mayer, model 5.9 percent vs. market 5.3 percent implied (+0.6 point edge). Modest on the win line, but the No. 1 Optimal percent on the board makes it a strong DFS-correlated play.

A third name worth flagging: Sheldon Creed sits with zero edge against the book's number, but the model and the market agree exactly on him at 11.1 percent. That level of agreement on a competitive market is itself a high-confidence signal.

See the full odds + edge board across all three series: https://fastlap.io/edges

How to use FASTLAP this race week

Run your own 10k sim at https://fastlap.io/oreilly/simulator with custom boosts, equipment handicaps, and dominator picks. Toggle drivers in and out of the field, override starts.

Browse the Xfinity Series power rankings at https://fastlap.io/oreilly/preview, full short-track, tire-combo, and pit-crew sub-ranks for every driver in the field.

Check the live Top Edges board at https://fastlap.io/edges, every priced driver across Cup, Xfinity, and Trucks ranked by where our model agrees and disagrees with the book.

Sign in free during beta at https://fastlap.io/pricing, every tool is unlocked for signed-in users while we're tuning the model.