The NASCAR Cup Series has finally found a format to keep Kyle Larson from stinking up the show. They're going to use it this week for the NASCAR Cup Series All-Star Race.

If he wins a segment, stick him in the back.

Sunday's All-Star Race is broken into six different "rounds." After each of the first three rounds, at least some portion of the field will be inverted. Your finishing position at the end of each round helps determine where you start the fifth round, so there's motivation to push hard; but whoever wins each of those first few rounds is going to get to deal with traffic for the next.

The rules behind this format are, quite honestly, mind-numbingly complicated. But you don't need a TI-89 to figure out implications for DFS. Those are actually pretty simple. Let's run through them now so we know how to fill out lineups for Sunday's unique event.

Logistics

Before we actually outline roster construction, it's important to note the pool of drivers we'll get to choose from. Not everybody who ultimately runs the All-Star Race will be available for us on FanDuel.

Before the All-Star Race is the All-Star Open. Here, all cars not already locked into the All-Star Race will compete to advance to the big show. Three drivers will race their way in, and one more will join them via a fan vote.

Those four drivers will be added to the field, but they won't be added to the player pool. This is pertinent for our upcoming discussion because A) they will occupy the final four spots in the starting order, and B) it means the eventual winner may not be in the player pool. That happened in 2019 when Larson had to race his way in and then wound up winning the race.

So, in other words, you should account for the eventual presence of those drivers. However, you don't need to sweat scrambling to fill out lineups after the Open on Sunday. You won't get to use those drivers, but nobody else will, either.

The Format's Implications

Outside of the weird player pool machinations and the inverts, the other way in which this event is unique is that it's a small field and a short race.

There are just 100 laps in total for Sunday night. That leaves 10.0 FanDuel points for laps led, which is basically a number we get exclusively on road courses among points-paying races.

That's one factor pushing us toward prioritizing place-differential. The other factor is the aforementioned inversions. We're not going to see one driver dominate the entire race; if they win an early segment, they're getting pushed to the back.

As a result, the biggest upside will come via drivers who can scoop place-differential points. However, with just 17 drivers available to us, that's easier said than done.

These two elements combined means that a disproportionate number of the total points scored on Sunday will come via finishing position. Upside is capped for both lap-leaders and place-differential cars, but the finishing points remain the same. It means that although we should emphasize finding place-differential, our overall goal will be to identify the fastest drivers.

The ideal situation is when those two overlap. Denny Hamlin and Ryan Blaney drew the 16th and 17th starting spots, respectively, but rank inside the top six in my model. That's going to make them high-quality DFS plays. Alex Bowman (starting 15th), Martin Truex Jr. (10th), and William Byron (8th) are also starting deeper than where my model ranks them.

As a result, those drivers are going to grade out well. If you can find drivers who have both finishing upside and some place-differential juice, they should be your core. But you'll still want to identify the eventual winner.

The 43 points for a win will likely guarantee the winner a spot in the tournament-winning lineups Sunday night. That's true even if that driver is Larson, who starts on the pole. This means it's impossible for him to get positive place-differential points; he'll either get zero or negative. But given his speed and race-winning upside (he has 22.3% win odds in my simulations; nobody else is in the double digits), he's still a quality DFS play.

Basically, the way to view starting positions is this: bump up drivers starting deeper than their true strength. Even the slightest boost via place-differential is worthwhile in a race that will likely be so slim on points. You just shouldn't sacrifice those ever-valuable finishing points to get there.

The other key thing to note for this week is that you'll have to work harder to differentiate your lineups from others. With just 17 drivers in the player pool, there are fewer total combinations of lineups we can make. If we have a lineup in a contest that's duplicated by an opponent, the expected value of that lineup automatically goes down. So, whether that means leaving salary on the table or consciously using potentially sub-optimal plays, you need to actively try harder to be different than you would for a full-field race.

Being different while also being considerate of finishing position and place differential is a tough line to walk. This might be the most difficult slate of the entire year. But those are complications our opponents will deal with, too, and not all of them will take the proper time to consider the unique format. Even if it makes our head hurt, we've still got an opportunity to profit on Sunday, and considering all three of those things simultaneously will be necessary if you want to do so.