Boston needs more offense without spending its pitching cushion

The decision

Boston allowed 65 fewer runs in 2026 and won two fewer games. Its offense lost 97 runs. The next move should address that imbalance while protecting the pitching that kept an 87-win club in the postseason. Our preferred plan buys an established second-base bat and brings Sonny Gray back on a new agreement. It also places a firm limit on how much recovery work the rotation can assume.

Offer Brandon Lowe three years and $66-75 million for a defined second-base job. Offer Gray one year and $20-24 million for 28 starts and 160 innings. Include a separate $6-8 million depth reserve. Annual additions total $48-57 million. With existing commitments and the roster pool, the resulting player-allocation envelope is about $269.11-282.11 million. These are S&S negotiating ceilings, not reported offers or an approved ownership budget.

That cost is the strongest argument against the recommendation. Boston already carries substantial commitments to players with limited recent availability, and the reconstructed 2027 base is higher than the headline payroll estimate. The preferred plan is defensible only if ownership accepts that allocation without requiring an unsupported salary dump or an undisclosed medical bet. The cheaper alternatives are complete rosters with identifiable compromises.

Plan B replaces Gray with Nick Martinez at two years and $26-30 million. Annual additions fall to $41-48 million, but the club buys a second guaranteed year and a different performance profile. Plan C signs neither outside player and retains only the $6-8 million reserve. Campbell, Durbin, Sogard and Mead absorb Lowe's proposed work; Bello, Bennett and the depth starters absorb Gray's. Neither branch receives imaginary production from an unfilled allowance.

A good season with a shrinking margin

The Red Sox moved from 81-81 in 2024 to 89-73 in 2025, then finished 87-75 in 2026. Runs allowed declined in each season, from 747 to 676 to 611. Runs scored moved in the opposite direction in the most recent year, falling from 786 to 689. The result was a plus-78 run differential, down from plus-110. That is a strong starting position, but a thinner one.

Boston went 21-21 in one-run games and 7-4 in extra innings, counted from 162 unique official regular-season finals. There is little basis for making bad close-game fortune the central explanation. The simple squared-run formula puts 689 runs scored and 611 allowed at roughly 91 wins. That describes the season's run balance; it does not establish a four-win entitlement for the next club.

The postseason ended with two clear defeats: Yankees 9, Red Sox 0, followed by Yankees 9, Red Sox 2. Those games matter as the actual end of the season, but two games cannot diagnose every winter need. The full-season offensive decline supplies the more durable case for improvement. Gray's departure and the rotation's medical uncertainty supply the reason not to spend every available dollar on hitters.

The roster changed while the season was being evaluated

The November 3 official major-roster snapshot contains 40 names. Marcelo Mayer and Connelly Early are no longer Boston development solutions. Mayer went to San Francisco in the August 3 transaction involving Erik Miller; Early went to Washington in the July 26 Curtis Mead trade. Carlos Narvaez departed in the August 3 Baltimore deal that brought Adley Rutschman. Historical Boston statistics remain in the evidence tables, with no retained workload assigned to those departures.

Gray elected free agency after his option was declined. Patrick Sandoval and Isiah Kiner-Falefa are also free agents. Zack Kelly was designated for assignment on November 2, with his outcome still requiring reconciliation. Eduardo Rivera was designated and then traded to St. Louis on November 3. Nate Eaton's Athletics trade and Tsung-Che Cheng's Detroit trade further change the depth list. A familiar name on an old depth chart does not fill a current job.

There is no presumed empty 40-man place for Lowe, Gray, Franklin Arias or Blake Wehunt. Their proposed additions and selections require a legal roster plan at the time of execution. Kelly's uncertain cost remains a financial contingency without a baseball role. The analysis does not solve roster pressure by silently waiving a useful player or assuming a non-tender has already occurred.

The offense has a small group of dependable anchors

Wilyer Abreu supplied 681 plate appearances, 24 home runs and a .773 OPS. His three-year OPS sequence of .781, .786 and .773 shows why he should retain a substantial role. Willson Contreras was the most productive established bat in the 2026 lineup, with 27 home runs and a .902 OPS in 535 plate appearances. The plan assigns him 550, while preserving his first-base job and some designated-hitter work.

Ceddanne Rafaela's .765 OPS was another real gain. Jarren Duran's .627 was a major loss of production compared with .834 in 2024 and .774 in 2025. The proposed 450 plate appearances for Duran preserve his speed, range and rebound opportunity while recognizing that his recent offensive decline cannot simply be erased. His role spans all three outfield positions, giving the club a way to rest or replace another outfielder without creating an entirely new defensive arrangement.

Trevor Story's 266-plate-appearance season and .622 OPS do not support planning around an uninterrupted full season at shortstop. Roman Anthony's .735 OPS in 242 plate appearances was below his 2025 debut performance. His wrist and hand history makes both workload and recovery relevant. The plan preserves a meaningful place for each, then names the players who absorb their work if that place cannot be filled.

Expected statistics sharpen the offensive argument

Abreu's .335 wOBA and .346 expected wOBA support continued confidence. Contreras's .391 and .376 remain strong even after acknowledging that his actual outcome exceeded the expected measure. Anthony's .326 and .341 suggest a more encouraging underlying sample than the raw decline alone. These measures describe contact and plate appearances through a particular model; they are not guaranteed future slash lines.

Rafaela's .332 wOBA against .285 expected wOBA is a reason to temper offensive extrapolation, even while preserving his exceptional range. Durbin's .317 against .291 creates a similar caution. Duran's .274 against .301 leaves room for a better outcome without supplying evidence for a full return to his 2024 peak. Story's .272 and .248 make availability only one part of his question.

Mead presents a particularly important split between sample and affiliation. His full MLB season covered 329 plate appearances with a .851 OPS, .370 wOBA and .358 expected wOBA. Only two of those plate appearances came for Boston. The acquired player brings encouraging evidence, but Boston has almost no direct major-league sample of him after the trade. His wrist recovery and defensive fit remain necessary parts of the recommendation.

Handedness should guide specific lineup choices

Lowe's 2026 split is large: a .648 OPS against left-handed pitching in 211 plate appearances and .910 against right-handers in 445. Buying him does not mean ignoring that distinction. Campbell, Mead and Durbin provide ways to reduce his most difficult matchups while preserving 550 total plate appearances. The proposed position schedule includes only 120 starts at second base, plus ten at designated hitter.

Anthony also had a meaningful split, .588 against left-handers and .817 against right-handers, with much smaller samples. Rutschman produced .649 and .792. Abreu, however, hit left-handers well in this season, with an .854 OPS in 217 plate appearances. A rote platoon assumption based on reputation would discard useful current evidence. The manager should combine these samples with longer history, health and actual player development.

Yoshida's 30 plate appearances against left-handers are too few to establish a broad conclusion. Mead's results against both sides were encouraging, but his total role must account for defensive cost and recovery. This is a flexible lineup with specific matchup choices, not a promise that every right-handed hitter will solve every left-handed pitcher.

Protect the center of the field

Rafaela's Outs Above Average moved from minus-2 in 2024 to plus-21 in 2025 and plus-16 in 2026. His 2024 work included different positions, so the sequence is not a pure three-year center-field rating. The recent positive range is nevertheless a strong reason to preserve 135 center-field starts. Duran's plus-5 in 2026 and Abreu's plus-2 help support the outfield arrangement.

The infield is less settled. Durbin's plus-6 is useful. Mead's minus-9 and Monasterio's minus-15 limit the appeal of simply filling offensive vacancies with more of their defensive innings. Sogard's plus-3 supports a utility role, subject to position and sample. Story's minus-2 followed minus-9 in 2025, which argues for evaluating his movement rather than assuming the old defensive reputation has returned.

Lowe's minus-14 in 2025 followed by plus-15 in 2026 illustrates how uncertain a one-season range estimate can be. The purchase is not credited with an automatic 15-run defensive dividend. He must hold the position well enough that the offensive upgrade survives its full cost. The comparison with Torres later in the report explicitly includes range rather than comparing only their OPS figures.

Rutschman changes the catching calculation

Rutschman's 126 Boston plate appearances produced a .688 OPS. His full MLB season was 410 plate appearances at .741. The distinction matters: evaluating only the Boston sample would overlook most of the season being purchased through his retained arbitration rights. His full-season framing value was plus-6.42 runs and total catching value plus-12.24. The broader catching measure includes components beyond framing and should not be added to it.

Wong caught 570 innings for Boston, with minus-7.72 framing and minus-9.11 total catching value across his MLB sample. That does not remove his practical value as the established backup, but it makes a Rutschman-led workload sensible. The plan assigns 105 catching starts to Rutschman, 45 to Wong and 12 to Gasper, with separate first-base and DH work where specified.

Gasper's biceps surgery carries a reported four-to-six-month recovery window from the September procedure. His 100 proposed plate appearances and 12 catching starts are conditional. If he is unavailable, Wong receives 50 additional plate appearances and the 12 catching starts; Romy Gonzalez receives the other 50 plate appearances and ten first-base starts. No new signing is required for that individual branch.

A strong rotation result contains several different stories

Boston starters worked 837.2 innings with a 3.92 ERA, 813 strikeouts and 263 walks. Gray's 169 innings at 2.72 ERA were a major contribution. Payton Tolle supplied 148.2 innings at 3.03, and Ranger Suarez delivered 148.1 at 3.22. Jake Bennett added 113.1 innings at 3.65. The club has credible starters to build around, but it cannot retain Gray's work merely by retaining his name in a historical table.

The expected figures distinguish those seasons. Tolle's 2.70 xERA and Bennett's 3.42 support the quality of their results. Suarez's 3.63 and Gray's 3.87 are less dominant than their actual ERAs but still useful starting points. Bello's 4.33 ERA and 4.89 xERA support a more cautious 20-start, 110-inning assignment. More work for Bello is an explicit consequence of the cheaper plan, not a free improvement.

Crochet made six starts and worked 30 innings with a 6.30 ERA. The plan assigns him only 20 starts and 100 innings, conditional on shoulder recovery and a genuine starting progression. That is already a substantial increase. It cannot be described as the conservative certainty that makes a second outside starter unnecessary. Houck, Crawford and Oviedo receive no principal innings because their recovery evidence does not establish reliable 2027 workloads.

The bullpen earned credit, but cannot absorb every uncertainty

The bullpen worked 602.1 innings at a 3.00 ERA. Chapman and Whitlock were central to that success: Chapman struck out 71 in 51.2 innings with a 1.92 ERA; Whitlock struck out 59 and walked nine in 52.1 innings with a 1.72 ERA. Their expected ERAs of 2.64 and 2.62 still support important roles while suggesting that repeating the exact run prevention should not be assumed.

The next tier matters because the plan reduces the relief burden to 545 innings without pretending the starters will become unusually durable. Weissert receives 65, Moran 75 and Miller 60. Moran's 33 walks in 65.2 Boston innings are a clear command risk. Miller's full-season evidence is separate from his 18.2 Boston innings. Olds's low ERA came in only 15.2 innings, with a 3.47 xERA; the proposed 45 innings is an opportunity to test, not an established performance floor.

Guerrero's 50 innings depend on a shoulder review. If he cannot supply them, Miller receives 15 more, Weissert ten, Olds 15 and Samaniego ten. Slaten receives no principal innings while his flexor condition remains unresolved. Kelly's designation removes him from the workload model. The club should preserve bullpen flexibility rather than spending its reserve on a named closer role already supported by multiple retained pitchers.

Administrative activation is not a recovery forecast

Anthony's hand and wrist discomfort, Story's sports-hernia history, Mead's wrist injury and Gasper's surgery all affect the position-player schedule. The latest injury update expects Anthony to recover for 2027, but an expectation remains different from demonstrated readiness. The plan gives him 450 plate appearances and an explicit replacement branch. Story receives 350 and 80 shortstop starts, leaving Arias and Sogard real work rather than merely listing them as insurance.

Crawford's current flexor reporting differs from older wrist-related transaction language. Oviedo's flexor reporting and elbow-list wording also need careful treatment. Casas is the most obvious unresolved source conflict: the team page's heading, body text and API injury description refer to different body areas and recovery histories. This report does not choose one unsupported medical narrative. Casas keeps his estimated cost but receives no principal plate appearances until the record and readiness are reconciled.

Crochet's September activation followed a long shoulder absence. Houck's rehabilitation and Slaten's flexor treatment do not justify filling the remaining pitching budget with optimistic returns. Whitlock's knee and elbow interruptions also warrant current review despite his excellent final statistics. These are baseball-planning limitations, not diagnoses or claims about private medical information.

The farm can fill defined jobs

Arias's Double-A line was .994 OPS in 332 plate appearances; his Triple-A line was .894 in 188. The plan gives him 250 major-league plate appearances and 60 shortstop starts after a roster selection and readiness decision. It does not translate those minor-league OPS figures directly into an MLB projection. Campbell's 462 Triple-A plate appearances at .763 OPS support continued development, but his 126 strikeouts and prior defensive difficulty require a real evaluation before expanding his role.

Wehunt worked 100.1 innings across Double-A and Triple-A. His 20 walks in 46.2 Triple-A innings make command an important promotion condition. The preferred plan asks only for two starts and ten innings; the internal plan expands that to eight starts and 45. Gamboa and Burgos have actual Triple-A starting experience, allowing their swing roles to be tested against evidence rather than invented after the budget is written.

Ziehl's combined 122 innings make him a possible conditional depth source after selection. Mullins's 132 strikeouts and 49 walks in 92.2 Double-A innings present both promise and risk. Romero, Castro, Brannon and Johanfran Garcia remain development reserves without automatic major-league quotas. Protecting those players from a premature assignment is compatible with giving Arias and Campbell substantial opportunities.

Why Lowe, and why not another familiar name

Lowe supplies the most direct offensive fit among the candidates tested here: 35 home runs, an .824 OPS and a legitimate second-base role in 656 plate appearances. His .355 wOBA and .331 expected wOBA call for a restrained projection, while 189 strikeouts and the handedness split define the risks. Three years and $66-75 million is a ceiling for a meaningful everyday job, not a price for a bench complement.

Torres is a real lower-cost alternative at two years and $24-28 million, with the same proposed 550 plate appearances and 120 second-base starts. His .727 OPS, 57 walks and eight home runs in 420 plate appearances offer a different offensive shape. His minus-5 OAA and lower power make the compromise visible. Replacing Lowe with Torres reduces annual additions by $10-11 million at the paired endpoints, but the club should not claim equivalent run production or range.

Suarez's 26 home runs came with 169 strikeouts and negative range. Bohm's 15 home runs and minus-11 OAA do not make him the preferred solution to this particular roster. Arraez adds contact but creates a less useful first-base and DH overlap with Contreras, Yoshida and other retained players. Trevor Rogers remains a separate pitching inquiry; Wacha is a contracted Royal whose availability is not assumed. Giolito's small sample and elbow history exclude him from the principal purchase.

Repeated targets must have repeated scrutiny

The private Pittsburgh analysis also proposes three years and $66-75 million for Lowe. In Pittsburgh that is a retention decision for an established role; in Boston it is an external purchase that reduces second-base opportunities for Campbell and Durbin and reshapes Mead's work. The same negotiating band does not mean both clubs can sign him. Boston must accept the failed-negotiation branch if Pittsburgh, another club or the player chooses a different outcome.

Martinez has appeared in the private Washington and Colorado preparations. Washington asks for 30 starts and 165 innings as a primary purchase; Boston's alternative asks for 28 and 160. Colorado uses him as a more expensive replacement for Fedde. The common two-year, $26-30 million ceiling reflects the same underlying candidate evidence, not three independently obtainable pitchers. His 3.10 ERA alongside a 3.98 xERA and 109 strikeouts in 180 innings argues for workload value without selling an ace profile.

Gray's proposed return is a new free-agent agreement after the option decision. The plan does not credit him with both retained salary and new salary, or treat the historical Cardinals credit as a repeatable source of money. These are free-agent proposals with no outgoing talent. Any later trade inquiry must name the outgoing players, control, replacement workload and resulting payroll before it can replace one of these branches. No player is added merely to make Boston's candidate list look different.

The contract bridge changes the apparent room

The retained contract allocations total $146.408 million across 11 players, including Chapman's $13 million and Whitlock's $8.25 million. Whitlock's option exercise is confirmed by the November 1 team announcement. Chapman's option is marked vested in the current payroll record after the reported innings trigger; potential award escalators still require a final check. The allocations include prorated bonuses in several cases and are not identical to either base salary cash or competitive-balance-tax AAV.

The 16 current-roster arbitration estimates total $47.9 million. No non-tender saving is assumed. Kelly's separate $0.8 million contingency remains until his designation resolves. Houck, Crawford, Oviedo and Casas retain estimated costs despite receiving no principal workload. A player can occupy a payroll obligation without being a dependable part of the baseball plan.

The bridge also carries a $10 million Gray buyout reserve, $4 million Hicks obligation, $2 million Devers bonus allocation and a $4 million Cardinals credit for Contreras. Gray's current settlement and timing need reconciliation before public release. The earlier reported contractual provision and current payroll table support preserving the $10 million reserve; a conflicting secondary snippet is not a basis for quietly cutting it in half.

Three budgets, with cash and tax kept separate

The RosterResource 2027 headline of $186.358 million is not the complete operating base used here. It excludes the Chapman and Whitlock salaries while carrying their potential $0.3 million and $1 million buyouts. Adding $13 million and $8.25 million and removing those two buyouts produces a $19.95 million correction. Add Kelly's contingency and a $14-18 million pre-arbitration, minor-league and callup pool to reach $221.108-225.108 million.

Plan A adds Lowe at $22-25 million annually, Gray at $20-24 million and the $6-8 million reserve. New multiyear commitment, including the first-year reserve, is $92-107 million. Plan B's lower annual Martinez allocation produces $98-113 million of new commitment because it buys a second pitching year. A lower first-year total therefore does not mean a smaller total obligation.

The preferred allocation is substantially above the historical 2026 estimate of $203.398 million. The 2026 tax estimate was a different figure, $271.752 million, under a different calculation. Benefits, actual payment timing, future labor terms, tax rates and roster adjustments prevent a responsible exact 2027 tax bill here. The recommendation requires an explicit budget decision; it does not assume revenue, ownership approval or a new labor outcome. The reserve earns no plate appearances or innings until an actual player is identified.

Every proposed plate appearance has a job

The principal plan allocates 6,200 plate appearances across a season. That is a planning convention, not a prediction of an exact team total or a simultaneous active roster. Abreu receives 600; Contreras, Lowe and Rafaela 550 each; Rutschman and Durbin 500 each. Anthony and Duran receive 450 each. Story, Mead, Yoshida, Arias and Campbell have smaller but meaningful roles.

Casas receives zero principal plate appearances. Jones remains retained at an estimated arbitration cost without a guaranteed quota in this plan; his short Boston offensive surge does not compel a role ahead of the fuller-season evidence. Utility players move through active-roster combinations over time. The club must manage actual options, selections and roster limits rather than assume all 20 planned hitters are active together.

The positional schedule makes the crowded areas visible. Contreras has 125 first-base starts and 15 at DH. Rutschman has 105 catching starts and 25 at DH. Campbell has 30 second-base and 30 DH starts. Anthony has 85 in left and 17 at DH. Those assignments prevent the report from treating designated hitter as an unlimited outlet for every player who needs rest or lacks a defensive position.

A rotation budget with a recovery limit

The preferred staff allocates 162 starts and 895 starter innings. Tolle receives 30 starts and 170 innings; Suarez and Gray 28 and 160 each; Bennett 24 and 135; Bello 20 and 110; Crochet 20 and 100. Gamboa, Burgos and Wehunt cover the remaining 12 starts. Tolle's increase over 2026 and Crochet's recovery workload require monitoring rather than a claim of automatic durability.

If Crochet cannot provide any of his 20 starts and 100 innings, the branch assigns Bello four starts and 20 innings, Bennett two and ten, Tolle two and ten, Suarez two and ten, Wehunt six and 30, and Ziehl four and 20 after selection. That branch also works arithmetically after the no-Gray plan. It still raises the developmental and workload risks, particularly for the depth pitchers.

If Wehunt is unavailable in the preferred plan, Gamboa takes his two starts and ten innings. Gamboa's relief assignment falls by ten and Olds's rises by ten, keeping the total workload intact. Houck, Crawford and Oviedo can improve the choices if their recovery becomes demonstrable. They are not necessary anonymous names used to make the current arithmetic balance.

Relief innings must survive the starter alternatives

The 545-inning relief plan leaves Chapman and Whitlock at 50 each, while spreading the rest across nine pitchers. Gamboa's 30 starting and 45 relief innings total 75. Burgos's 20 and 45 total 65. Counting those combined assignments prevents the common mistake of using a swing pitcher twice at full capacity.

If Gray is not signed, Gamboa adds 30 starter innings and Burgos adds 15. Their relief work falls by the same amounts. Weissert adds five relief innings, Moran 20, Miller ten and Olds ten. The no-Gray branch therefore preserves 1,440 planned staff innings without quietly enlarging Gamboa's or Burgos's total workload. It relies more heavily on less established pitchers, which is the actual price of saving the salary.

The 1,440-inning staff total is a planning convention, not the innings in 162 nine-inning games. Actual schedules include road losses, extra innings and other variation. If several pitchers fail together, the individual replacement branches require a fresh combined workload and roster review. The report does not assume that a reserve of dollars can pitch an inning.

The internal plan has a complete cost

Without Lowe, Campbell adds 200 plate appearances, Durbin 150, Sogard 100 and Mead 100. Campbell takes 50 additional second-base starts and ten at DH; Durbin takes 40 at second and gives up 40 at third; Sogard adds 30 at second; Mead takes those 40 third-base starts. The arithmetic preserves 550 plate appearances, 120 second-base starts and ten DH starts. The baseball tradeoff is a greater dependence on Campbell's development and Mead's defense.

Without Gray, Bello adds eight starts and 45 innings, Bennett six and 35, Gamboa six and 30, Wehunt six and 35, and Burgos two and 15. The corresponding relief shifts appear above. This is a real lower-cost path, not evidence that Gray has no marginal value. It turns several uncertain depth cases into more important parts of an 87-win team's attempt to improve.

The individual hitting failures are also explicit. Anthony's 450 plate appearances go to Duran, White and Yoshida. Arias's 250 go to Sogard and Monasterio. Gasper's 100 go to Wong and Gonzalez. Mead's 350 go to Durbin, Monasterio, Gonzalez and Yoshida. These branches are not additive without recalculation. Losing several of those players together could exceed the retained roster's practical capacity even when each isolated branch is numerically complete.

What would make the recommendation wrong

Lowe could decline, cost more than the ceiling, attach draft compensation that materially changes the value, or lose enough defensive range that his bat no longer justifies the full package. Gray could prefer another club or require a second guaranteed year Boston should not grant at this price. Crochet's shoulder and the other recovery cases could leave a larger pitching gap than the principal plan anticipates. Those are concrete reasons to retain the alternatives rather than sell the preferred roster as inevitable.

The run examples isolate the scale of the problem. Forty more runs scored with the same prevention improves the descriptive run-ratio result, while allowing 60 more with unchanged offense gives away much of Boston's current advantage. These are not projections assigned to Lowe or Gray. Adding their past production to the team total would ignore displaced players, playing time, park, age, opponents and uncertainty.

The decision should be revisited if ownership's actual budget cannot support the reconstructed base plus the additions, if medical evidence changes the workload assumptions, or if a genuinely available trade produces a better fully costed outcome. An internal season can still be defensible if Campbell, Arias and the young starters advance. It should be chosen for that developmental and financial case, with the offensive risk acknowledged.

The retained roster and the recommendation

Boston has enough established pitching and defensive strength to justify an attempt to improve now. Its most useful purchase is an offensive role the current roster has not securely filled, accompanied by a defined replacement for Gray's departed innings. The preferred plan is expensive because the existing commitments are expensive. Concealing that starting point would make the recommendation easier to sell and less useful to evaluate.

Keep the strongest core roles, preserve the cost of injured players, and make the younger players earn defined opportunities. Buy Lowe only within the stated ceiling and with current compensation and health review. Re-sign Gray only through a new agreement that fits the one-year limit. If either negotiation fails, use the named alternative and recalculate the combined roster. The goal is a better offense that leaves Boston's hard-won pitching advantage intact.

Definitions and sources

The report separates Boston-only 2026 contributions from full MLB seasons, current roster from departed players, and major-league from minor-league levels. Innings notation counts outs: 46.2 means 46 innings and two outs. Expected statistics and OAA are descriptive measures with sampling and model limitations. Catching run value overlaps framing and must not be summed with it. Handedness samples are dated evidence, not fixed future platoon identities.

Contract allocations, cash payments and tax AAV are distinct. Arbitration is estimated until agreed. Proposed prices, plate appearances, starts, innings, reserves and run sensitivities are S&S scenarios. No reported offer, confidential medical finding, trade availability, tax outcome or ownership budget is inferred. Research and roster checks are dated November 3, 2026; public release requires a fresh current-source review.