Colorado had Monday's second-highest expected offense of 20 lineups — and scored the fewest runs of all 20: none

You saw a 9-0 rout, and the box score backs the memory: Arizona beat Colorado 9-0 at Chase Field on Monday, August 10. Arizona out-hit Colorado 10 to 5, Nolan Arenado hit a first-inning grand slam, and Arizona scored 7 of its 9 runs in the first two innings — 4 in the first, 3 in the second — before a combined shutout closed it. That is one game, read one game at a time. We read the whole slate. We pulled Statcast's pitch-and-batted-ball record for all 10 games that played and re-derived each team's expected weighted on-base average (xwOBA) ^mmodeled. xwOBA is an expected-offense rate on the on-base scale: it estimates the on-base value a team's plate appearances "should" produce, built from the quality of contact (exit velocity and launch angle) and from the non-contact outcomes — walks and strikeouts — in the same plate appearances. It is not a measure of how hard a team hit the ball, and it does not see where the fielders stood or when the hits came. Read all 20 team-performances at once and Colorado is the shape no single box score holds. Colorado's team xwOBA was .387 ^mmodeled. Of the 20, that ranked second — behind only Arizona's .412 and ahead of Philadelphia (.383), Minnesota (.379) and every other lineup that took the field ^mmodeled. Seven of the eight highest expected-offense performances of the day scored at least six runs — Tampa Bay 10, Minnesota 9, Arizona 9, the Mets 8, and so on. The eighth, Colorado's, scored zero. The Rockies reached base 10 times — 5 hits, 5 walks — and stranded all of it. Two outcomes in the batted-ball log carry the same shape in miniature. Colorado's two hardest-hit balls in play were a 104.5-mph drive at a 25-degree launch angle and a 103.4-mph drive at 33 degrees — batted-ball profiles the model scores at 1.53 and 1.18 estimated wOBA per ball, near the top of the scale for a single struck ball ^mmodeled. The first was caught for a fly out; the second became a double play. Colorado's 22 balls in play averaged 86.5 mph off the bat and totaled 5 bases. Arizona's 29 balls in play averaged 89.1 mph and totaled 18. The gap between Colorado's team xwOBA (.387) and the .242 wOBA it actually produced — a shortfall of .145 on the rate — was the largest such expected-to-actual shortfall of any team on the slate ^mmodeled.

The following is analysis, not fact. A box score shows one game; we read the whole slate. Read one game at a time, this is a shutout and nothing more. Read all 20 team-performances at once, the way only a complete read allows, and Colorado is the day's outlier — the point where an expected-offense rate and an actual run total came apart hardest. Across the slate the two tracked loosely: the lineups with the highest team xwOBA mostly scored, and the lineups with the lowest mostly did not ^mmodeled. Colorado broke the pattern from the top. It posted the second-highest expected-offense rate of the 20 and scored fewer runs than any of them. The two measures are not the same axis, and the gap between them is not itself explained by these data. Team xwOBA is a rate on the on-base scale (Arizona .412, Colorado .387); the scoreboard is a count of runs (9 to 0). A .025 difference in the rate and a 9-run difference in the count are different units and do not convert into one another without a run-expectancy model, which we do not have ^mmodeled. So we can say what happened — Colorado generated the day's second-highest expected-offense rate and scored zero — but not why. Sequencing, the timing of Arizona's hits, the Diamondbacks' pitching, Colorado's stranded runners, and ordinary single-game variance are all candidates; none is established by the batted-ball record alone ^sspeculative. What the record does show is narrow and firm: a team can produce a high expected-offense rate and still be shut out, and on this slate Colorado is the clearest instance of it.

We are not claiming Colorado out-hit Arizona. Arizona's team xwOBA was higher (.412 to .387) ^mmodeled, it put twice as many balls in play at 95-plus mph (10 to 5), and its average exit velocity was 2.6 mph higher (89.1 to 86.5). The claim is narrower: Colorado posted the day's second-highest expected-offense rate and scored zero runs, the sharpest such decoupling on the slate. The strongest counter is that the rate is not a promise of runs, and one game is noise. xwOBA is a rate, not a run count; a .387 team figure describes the on-base value the plate appearances "should" carry on average, not the runs any single game must yield ^mmodeled. Over 33 plate appearances that rate is volatile, and it does not see defensive positioning, the ballpark, or a good pitch in a good spot ^mmodeled. A 9-0 shutout on 5 hits and 10 baserunners is also a story about pitching and about stranding runners: Arizona's staff walked 5, struck out 6, and allowed nothing. And a 104.5-mph ball hit at 25 degrees is a good outcome for the hitter on average, not on every swing — defenses convert some of that contact into outs by design ^mmodeled. The scoreboard, a skeptic would say, is the only number that clears. It read 9-0, and no rate overturns it.

We were never in the ballpark, and we did not watch a pitch. We read the record of every plate appearance across all 10 games after the fact, whole, as a single object — and from that vantage Colorado's day looks the way we are describing it, which no single seat in Chase Field supplied. A spectator saw a first-inning grand slam and a lopsided line, not a slate-wide ranking. That side-by-side read is the finding. The same shutout a box score files under "didn't hit" a whole-slate read files as the day's second-highest expected offense returning the fewest runs of the 20.