Unders Do the Work While the DFS Chaos Was Something Else

Fifteen games on a Sunday. That's a lot of baseball, and a lot of chances for the model to look either very smart or very humbled. Sunday split the difference in the most honest way possible: we went 3-1 on totals, 0-0 on the moneyline side, and the DFS slate produced some of the wildest scorelines of the summer. The run environments were all over the place — Fenway played tight, Comerica played tight, Citi Field went over on a Dodgers-Mets game we faded, and Milwaukee looked like a Little League scoreboard. On balance, a good day. Not a perfect one. That's the job.

We published four totals with material edge on Sunday. Here's how they graded out.

• TOR @ BOS — Under 8: Final was TOR 1, BOS 6 — seven total runs. Quiet pitching duel in Boston, the offense never got going on either side. WIN.
• KC @ DET — Under 9.25: Final was KC 5, DET 4 — nine total runs. This one barely breathed. Slid under the number by a quarter run, which is the kind of finish that makes you exhale slowly. WIN.
• LAD @ NYM — Under 9.5: Final was LAD 3, NYM 8 — eleven total runs. The New York Mets had other ideas. When a team drops eight runs in a game we're fading, there's not much to say except the model missed the environment. Own it. LOSS.
• COL @ MIL — Over 7.5: Final was COL 2, MIL 11 — thirteen total runs. The Milwaukee Brewers did the heavy lifting almost entirely on their own. The Colorado Rockies managed two, but thirteen runs clears 7.5 with room to spare. WIN.

Totals record for Sunday: 3-1. No moneyline plays reached the threshold, so the card was totals-only — that's the right call when the edge isn't there on the sides.

470 players scored across the Sunday slate. We matched projections to 222 of them. Some of those matchups were very good. A few were genuinely rough. Let's get into it.

Top Scorers

• Munetaka Murakami (Chicago White Sox) — 41 DraftKings points. Line: 3-for-5, 2 HR, 6 RBI. We did not have a projection on Murakami for Sunday. He was the best hitter on the slate and we're simply tipping our cap to a monster performance.
• Jacob Misiorowski (Milwaukee Brewers) — 36 DK points. Line: 5.0 IP, 12 strikeouts, 1 ER. Our projection: 26.62 points at a $11,300 salary. He was the COL/MIL Over engine and then some — double-digit strikeouts for a pitcher that expensive is the kind of night that justifies the price.
• Ryan Kreidler (Minnesota Twins) — 36 DK points. Line: 3-for-4, 1 RBI, 2 SB. Our projection: 11.94 points at a $3,000 salary. More on this in a moment.
• Drew Rasmussen (Tampa Bay Rays) — 35.9 DK points. Line: 7.0 IP, 9 strikeouts, 0 ER. Our projection: 18.28 points. Also more on this in a moment.
• CJ Abrams (Washington Nationals) — 33 DK points. Line: 3-for-5, 2 HR, 3 RBI. Our projection: 9.44 points at a $6,200 salary. Another one below.

Where We Hit

• Ryan Kreidler, Twins — projected 11.94, scored 36. At $3,000, that's a 24-point outperformance from the minimum salary tier. Two stolen bases and three hits from a guy most lineups weren't paying up for. That's what the model is supposed to find, and on Sunday it found it.
• Drew Rasmussen, Rays — projected 18.28, scored 35.9. Seven innings, nine strikeouts, a shutout. Nearly doubled our projection. When a pitcher goes that deep and that clean, the point total stacks fast — he outperformed by almost 18 points.
• CJ Abrams, Nationals — projected 9.44, scored 33. Two home runs will do that. We had him as a lean-value play and he turned in one of the better offensive performances on the slate. Good call, better result.

Where We Missed

• Ronel Blanco (Houston Astros) — projected 15.56, scored 5.8. Line: 5.1 IP, 4 K, 5 ER. Five earned runs will crater a pitching score in a hurry, and they did. This one wasn't close.
• Will Warren (New York Yankees) — projected 15.2, scored 0. Line: 3.0 IP, 5 K, 6 ER. Three innings, six earned runs — he was gone before the game had a chance to stabilize. When a starter gets knocked out that early, the DK score hits zero in a way that's genuinely painful to look at.
• Jameson Taillon (Chicago Cubs) — projected 12.08, scored -3.8. Line: 4.0 IP, 3 K, 7 ER. Negative points. Seven earned runs. There's no spin on that. We were wrong, the blowup was real, and it cost lineups that rostered him.

The pitching misses on Sunday were concentrated and they were bad. Blanco, Warren, and Taillon combined to give up 18 earned runs. That's not a model problem in isolation — pitchers blow up — but when three starters we liked all go sideways on the same day, it's worth noting.

The honest takeaway from Sunday is that the totals model is working and the pitching projections need to hold up on days when we're stacking them. Three of four totals cashing is a good day by any measure. The COL/MIL over was the clearest call and it wasn't close. The KC/DET under barely cleared but cleared.

The DFS lesson is the one that stings a little and encourages at the same time. Kreidler at $3,000 going for 36 points is exactly what low-salary targeting is supposed to produce. The model liked him relative to price, and he delivered four times what we projected. That's leverage. Meanwhile, three pitchers we projected in the 12-to-16 point range combined for about two DK points total. Pitching volatility is real and it is ruthless, and on high-variance Sundays with fifteen games, it will find you eventually.

Going forward: lean harder into salary relief valuation when the model finds a gap like Kreidler's. And when multiple starters in a similar projection band are on the same slate, spreading the exposure — rather than concentrating it — is the smarter construction. Sunday was a 3-1 day on totals and a humbling one for the pitching builds. Both things are true.