2395 Preseason Power Rankings — projections, offense/defense grades, and the Central's third straight year as the meat grinder
By "USAFL Analytics Desk — preseason, 2393"

Off-Season 2395 (Washington, DC) - # THE MODEL SPEAKS: 2395 — The Ravens Bet the Draft, and the Math Says They Won It

*The annual preseason projection, from the same model that called last season to within 0.98 wins per team — its best year yet, and the third straight under 1.8. All twelve projections are logged and will be scored in public when the season ends, as always. The model has no friends and no enemies; it has a regression.*

## Projected Standings

**NORTH** — Pittsburgh 10-4, Buffalo 8-6, Green Bay 6-8, Minnesota 3-11
**CENTRAL** — Baltimore 12-2, Cincinnati 11-3, Las Vegas 6-8, Washington 3-11
**SOUTH** — Los Angeles 11-3, New Orleans 7-7, Atlanta 6-8, Houston 1-13

Projected playoff field: **Baltimore, Pittsburgh, Los Angeles, and Cincinnati** on the wildcard.

## The headline: Baltimore's all-in

Mark spent draft night like a man buying every lottery ticket in the gas station — five linemen, including both tackles the scouting world argued loudest about, then came back in waivers for more. Here's the thing the model wants on the record: **Baltimore projects first at 12-2 almost entirely on the veteran core that went 11-3 last year.** The rookie trench haul is variance stacked on top of a floor, which is exactly when variance is free. If even two of the five hit, this team is built for a three-year window. If none do, they're still the best team in the league today. That's not gambling; that's gambling with house money.

## The champs reload quietly

Cincinnati returns everything that mattered from the title run and drafted DT Austin McKnight 12th, who the model graded as a top-7 talent. An 11-3 projection for a defending champion is not disrespect — it's a schedule that runs through Baltimore twice. The Central is suddenly the deepest division in football, and somebody good is staying home in December.

## Los Angeles, again

Every year the model projects LAC two games worse than they finish, and every year Addam collects the difference like rent. The regression says 11-3; history says add a game and apologize later. The model is aware of its LAC problem. The model is working on it. The model projects them to the title game anyway.

## The middle class

Pittsburgh's 10-4 rides the league's most stable roster and the North's gentlest path. Buffalo retooled on the fly — shipping out veterans, grabbing G Chris Sherman at 10, a draft the model quietly loved. New Orleans drafted three defensive backs in the first two rounds, which the model notes was the first DB taken top-5 in 45 recorded years; at 7-7, the Saints are one secondary breakout from relevance. Atlanta, Vegas, and Green Bay project within a game of each other, though Green Bay enters the year having passed twice on draft night — hard to add talent with picks you don't use.

## The rebuilds

**Minnesota** made the boldest pick of the draft — QB Allen Smith, first overall, the highest-graded quarterback prospect in league history, drafted onto a roster with an eleven-year veteran at the position. At 3-11 projected, the future is the point, and the future has an IN of 83 and nowhere to go but up.

**Houston** projects 1-13 and owns the story anyway: HB Martin Shields, the first perfect prospect grade ever recorded, went second overall to a team that will hand him the ball immediately. Floor of the league, ceiling of the sport.

**Washington** projects 3-11, and the model would like to show its work. The Commanders turned over fourteen roster spots in one offseason: six draft picks — five of them linemen and linebackers — eight waiver rookies, and a depth chart where the average age of the two-deep dropped by years. The draw is merciless (Cincinnati twice, Baltimore twice, zero projected wins against either), and the model does not pay teams for potential in October. But it sees what's under the hood: the rookie trench class posted the largest single-camp endurance gains in the league, and the roster's two-year projection curve is the steepest in the Central. Three wins now. Ask again in two seasons.

## On the record

Last year's projections finished within one win on nine of twelve teams. The worst miss was Atlanta at +2.5. Every number above is logged, timestamped, and will be scored against the final standings — including a certain 3-11 that the author's own front office would love to see embarrassed.

*— The Model, as told to the Washington analytics department*