BBMI MLB Power Rankings, Week of July 15
All 30 teams ranked by schedule-adjusted BBMI Score at the All-Star break, with two weeks of movement since the July 1 board and the changes worth explaining.
BBMI Research Desk
·
July 15, 2026
Break edition
Sixty-five years ago this week, baseball played an All-Star Game at Candlestick Park and a 165-pound reliever named Stu Miller became a trivia answer. A gust off the bay rocked him mid-delivery, the umpires called a balk, and by the time the newspapers were finished the story was that Miller had been blown clean off the mound. He hadn’t — he swayed, the runners moved up, and the National League won anyway — but “blown off the mound” is the version the sport decided to keep, because it was better. That’s what the All-Star break is for: the games stop, the retelling starts, and a half-season of results hardens into stories that are a little tidier than what actually happened.
The board’s job is to resist the retelling. The BBMI Score takes a team’s banked wins, projects the rest of the schedule off the current roster, and corrects for the strength of the opponents already faced — a credit for a tough slate, a debit for a soft one. BBMI is the ranking; Talent is the lens beside it, what a healthy roster is worth record-neutral. Housekeeping: the desk spent last week on the awards races, so the Chg column covers two weeks of baseball, diffed against the July 1 board. The shape of the two weeks: the top two both won more than they lost and both shed score, a team that hasn’t lost since the last board jumped seven spots, and the biggest score drop in baseball moved its owner all of two ranks. The break board is where the record and the score argue in public.
The full board
| # | Chg | Team | BBMI | 7/1 | Δ | SOS | Talent | Record | L15 |
|---|
Top of the board
The Dodgers (0.860) and Brewers (0.814) held first and second, and here is the oddity of the break board: both shed exactly −0.029 of score while going a combined 12–12. Nothing broke; they just stopped outrunning their own pace. BBMI is expected wins measured against the 116-win ceiling, and a .500-ish two weeks at the top bleeds projection even while the win column grows. The Dodgers also now carry the deepest schedule debit on the board (−2.3): the model has clawed back more than two full wins for the softness of the slate they banked their record against, and they are still first by .046. That’s what a No. 1 looks like. But the cushion is thinning — the gap from first to third was .100 on July 1 and is .063 at the break. The top is compressing.
The Rays (▲2, into 3rd) are the biggest gainers in the top six at +0.023, a 10–5 stretch while their schedule debit nearly vanished (−1.1 to −0.1): better baseball against firmer competition, priced accordingly. They stepped over the Braves, whose 6–9 slide is the regression our Running Hot, Running Cold piece has been collecting on since June — though Atlanta’s score barely moved (+0.001) because the talent lens rose while the record fell. And an honest accounting on the Yankees: two weeks ago we wrote that their rank would catch up to a −0.053 collapse. It caught one spot. The score stabilized instead (+0.008 on a 6–9 stretch, mostly roster inputs firming), and they sit 5th with the second-best talent in baseball (.584). The collapse got priced, and then it stopped. We’ll take being half right about the direction and wrong about the destination.
One more promise kept: on July 1 we said the Cubs’ score would come to their record if the run was real. It has started — +0.019, up to 7th, on a 10–5 run the model no longer discounts.
+.041
The movers
Red Sox ▲7 (20th to 13th). The jump of the break, and the cleanest kind: Boston has not lost a game since the July 1 board. Nine straight wins, a 13–2 last-fifteen that is the hottest stretch in baseball, and +0.041 of score, the biggest gain on the board. The contrast with the Cubs streak we discounted two weeks ago is the point: this run carries a +0.5 tough-slate credit, so there is no schedule asterisk to hang on it. The caution is the same as ever — a .529 Talent (10th) says this is a good roster, not a great one, and at 46–48 the climb is all projection, no banked position. But you don’t get to 9–0 against real competition by accident, and the model isn’t pretending otherwise.
Guardians ▲5 (15th to 10th). The quiet reversal. Cleveland fell three spots on the July 1 board and took all of them back with interest: +0.022 on a 9–6 stretch with a +0.6 credit for the slate it came against. Back in the top ten with the board’s usual caveat — a .472 Talent says the roster is renting the neighborhood, not buying.
Reds ▲3 and Angels ▲2 — the risers who lost. Cincinnati went 5–10 down the stretch and gained +0.031. Los Angeles went 4–11 and gained +0.036. Neither climb was earned on the field: both teams’ Talent inputs jumped hard (.389 to .459; .335 to .436), which means the model now grades the rosters themselves meaningfully better than it did two weeks ago, and the rest-of-season projection rose with them. We flag these because they are the opposite of the Red Sox move — input-driven, not results-driven — and input-driven climbs are the ones that evaporate if the inputs were wrong.
Mariners ▼5 (9th to 14th) and Athletics ▼5 (21st to 26th). The falls of the break, for opposite reasons. Seattle’s is the gentle kind: −0.019 on a 7–8 stretch, in a middle of the board so tight that two hundredths of score is five ranks. The roster case for them is below. Oakland’s is not gentle: 2–13, the coldest baseball anyone played into the break, −0.038 of score, and a slide most of the way to the floor. On July 1 the correction collected from the Orioles; this time it collected from the A’s.
Royals ▼2 (27th to 29th) — the fall the rank is hiding. Kansas City lost more score than any team in baseball, −0.048, and fell two spots. This is the Yankees pattern from the July 1 board relocated to the basement: the rank only moves if someone is there to pass you, and at the bottom the traffic is thin — the two teams that did pass them were the input-driven risers above. The score says the Royals were repriced as hard as anyone on the board; 3–8 since the last board, a 4–11 last-fifteen, and a Talent input that slid from .482 to .419 all point the same direction. The two-spot fall flatters them.
Biggest movers — rank change since 7/1
Spots gained or lost since July 1 · two weeks of baseball, one board
Up in green, down in red. The Red Sox’ seven-spot jump is the loudest move on the board — and the rarest kind, earned 9–0 at full schedule price.
The watch list (talent vs. record)
Mariners (14th, .554 Talent, 48–49). The new headliner of roster-over-record. The Talent column grades Seattle 7th in baseball; the board has them 14th; the record is a game under .500. Something has to give, and the model’s money is on the roster: the rotation inputs (a 3.83 staff FIP only the Phillies grade ahead of) never stopped grading out, the slide was two hundredths of score in a congested middle, and nothing in the underlying quality moved. This is the profile the watch list exists for — the disagreement most likely to resolve upward in the second half.
Tigers (17th, .528 Talent, 44–52). The July 1 thesis is playing out on schedule: we said the record would climb toward the talent, and Detroit went 10–5 into the break and picked up two more spots. The gap has narrowed from both ends — the record improving, and the Talent input easing from .562 to .528 — so this is now a good roster catching up to itself rather than a great one being ignored. The variable is unchanged: the trade deadline. What this front office does in the next three weeks decides whether the climb continues or the thesis gets liquidated mid-proof.
Mets (27th, .521 Talent, 40–57). Still the widest talent-versus-rank gap on the board, and still closing in the wrong direction. Another 6–9 stretch, another −0.007 of score, and a roster the model grades 13th in baseball sitting 27th on the board. On July 1 we wrote that the new-manager bounce would have to show up as wins before the score believed in it. Two weeks later the score is still waiting, and at 17 games under .500 the second half stops being a runway and starts being an epitaph. The model prices rosters, not funerals, so the gap stays on the watch list — but it is getting hard to call this one “undersold” with a straight face.
Talent vs. ranking — the roster the record won’t admit to
Where each team’s healthy-roster talent ranks vs. where BBMI ranks it · of 30
The line is the gap. Each roster grades out higher than the ranking — talent the record hasn’t paid off yet.
What the board is telling you
The break freezes the argument mid-sentence. The top is compressing — first-to-third went from .100 to .063 in two weeks where the top two won more than they lost — and the middle is as violent as ever, with sixteen teams packed inside .075 of score from 8th to 23rd. The second half opens with three live questions: whether a 9–0 Red Sox run survives contact with a .529 roster, whether the Royals’ rank finishes the fall their score already took, and whether the two input-driven risers earn prices the model advanced them on credit. The standings will spend the break being retold into stories. The board will be here repricing them.
The bottom line
Compressed top, frozen middle, and two teams to watch out of the break: the Red Sox, whose 9–0 run was banked at full schedule price and now has to be sustained by a merely good roster, and the Royals, holding a rank their score has already abandoned. The second half settles both.
The Chg column diffs against the prior published board — July 1 this time, two weeks back, since the desk spent last week on the awards races. Full sortable table lives on the MLB Rankings page; prior weeks are archived below.