SSILVERMINECollege basketball stats
LIVE BOARD
2026–27 SEASON
2026–27 / Basketball scouting brief

C Michigan
vs Buffalo

The projected game, the historical pressure points and the personnel questions to take into preparation. Generated from published statistics and reviewed announcements.

Feb 16, 2027 · start time unconfirmedBuffalo designated home · Broadview Arena

C Michigan73.2Projected away score
Preseason baselineBuffalo by 2.9Projected winning margin · 68.8 possessions / 40 min

Buffalo win estimate 61.8%

Buffalo76.1Projected home score

Projected total 149.4. Nominal 80% range for home scoring margin: -13.0 to +18.8. Either team winning falls within that range.

The model uses historical opponent-adjusted efficiency and tempo. It has no current roster or injury inputs. Schedule details are partial and may change. Model edition: Sep 12, 2026.

Checking the latest registered D1 forecast for this game…

Pre-tip readiness / Four checks

Know what is settled before you scout.

Updated with this brief edition
Forecast signal
Model baseline

Home win estimate 61.8% · 31.8-point nominal range.

Model notebook
Schedule status
Start time unconfirmed

The source schedule has not confirmed a tip time; recheck before distributing a plan.

Source schedule
Roster evidence
Both teams observed

C Michigan 100% prior minutes represented · Buffalo 100% prior minutes represented

Roster evidence
Market evidence
No licensed quote

No qualifying bookmaker observation is published; an edge cannot be reported.

Record a manual check

These checks describe source coverage and model state. They do not convert a roster listing into eligibility, a schedule into availability, or a market quote into a recommendation.

Reader tool / Manual quote check

Compare a line you observed.

Browser only · never published · checking model edition

Enter a home spread, game total or both sides of a moneyline from a source you are reviewing. Results use the current forecast when available and stay in this browser; they are not a verified market observation, recommendation or ledger record.

Up to 20 notes · never uploaded
Spread edge+2.9 ptsModel is more favorable to the home side.Normal approximation: 59.3% chance the home margin clears this line.
Total edge+149.4 ptsModel total is above the observed number.
Moneyline probability edgeEnter both moneylines to calculate a no-vig market probability.
Moneyline expected returnEnter both valid prices to compare home and away returns.

Convention: spread edge = model home margin + home spread; total edge = model total − observed total. This arithmetic does not account for vig, pushes, limits, timing or player availability. The spread-cover probability is a normal approximation from the published nominal 80% margin interval; it is not a calibrated betting probability. Moneyline edge uses both sides to remove two-way vig. Expected return is a hypothetical arithmetic output from the model probability and quoted price; it is not a recommendation or a realized result.

01 / Prior-season evidence → Film questions

Where the styles meet.

2025–26 observations

These are historical questions, not forecast adjustments. Each direction highlights up to two Four Factors with the largest difference in favorable percentile between one offense and the opposing defense. Rates came against different schedules and personnel; their difference is not an expected matchup outcome.

01 / A contrast to investigate

Shot-making

C Michigan offense Buffalo defense

C Michigan · offense52.3%31 games · rank 155/366
Buffalo · opponent rate54.8%32 games · rank 338/366

Effective FG%. Favorable percentile: 57.8 offense / 7.7 defense. Higher favorable percentile is better, including for statistics where a lower raw rate is better.

On film: Which actions create efficient shots, and which coverages take those looks away? Separate rim attempts, catch-and-shoot threes and contested pull-ups on film.

02 / A contrast to investigate

Protecting the possession

C Michigan offense Buffalo defense

C Michigan · offense18.3%31 games · rank 288/366
Buffalo · turnovers forced17.8%32 games · rank 121/366

Turnovers / possessions. Favorable percentile: 21.4 offense / 67.1 defense. Higher favorable percentile is better, including for statistics where a lower raw rate is better.

On film: Do the turnovers come from ball pressure, passing-lane help or unforced mistakes? Identify the ball handlers and distinguish live-ball losses from dead-ball turnovers.

03 / Offensive strength / defensive concern

Shot-making

Buffalo offense C Michigan defense

Buffalo · offense56.0%32 games · rank 27/366
C Michigan · opponent rate53.0%31 games · rank 292/366

Effective FG%. Favorable percentile: 92.9 offense / 20.3 defense. Higher favorable percentile is better, including for statistics where a lower raw rate is better.

On film: Which actions create efficient shots, and which coverages take those looks away? Separate rim attempts, catch-and-shoot threes and contested pull-ups on film.

04 / Offensive strength / defensive concern

Keeping the game off the line

Buffalo offense C Michigan defense

Buffalo · offense39.4%32 games · rank 66/366
C Michigan · opponent rate38.2%31 games · rank 282/366

Free-throw attempts / FGA. Favorable percentile: 82.2 offense / 23.0 defense. Higher favorable percentile is better, including for statistics where a lower raw rate is better.

On film: Which actions draw shooting fouls, and which defenders can contain them without fouling? Check the personnel and foul context before assigning a coverage.

02 / Both directions, all the evidence

The possession matchup.

C Michigan with the ball.

Historical rateC MichiganOffenseBuffaloDefense
Effective FG%52.3%31 games · rank 155/36654.8%32 games · rank 338/366
Turnovers / possessions18.3%31 games · rank 288/36617.8%32 games · rank 121/366
Offensive rebound rate29.8%31 games · rank 226/36630.5%32 games · rank 210/366
Free-throw attempts / FGA31.9%31 games · rank 259/36631.3%32 games · rank 97/366
Three-point attempt share33.9%31 games41.7%32 games
Two-point FG%53.7%31 games · rank 125/36656.5%32 games · rank 343/366
Three-point FG%33.1%31 games · rank 234/36635.0%32 games · rank 267/366

Defense reports opponent rates, except turnovers forced. Each statistic pools its own valid numerators and denominators; game counts can differ. Three-point attempt share describes style and has no favorable rank.

Open C Michigan’s full dossier →

Buffalo with the ball.

Historical rateBuffaloOffenseC MichiganDefense
Effective FG%56.0%32 games · rank 27/36653.0%31 games · rank 292/366
Turnovers / possessions16.1%32 games · rank 147/36617.7%31 games · rank 127/366
Offensive rebound rate27.8%32 games · rank 292/36633.9%31 games · rank 330/366
Free-throw attempts / FGA39.4%32 games · rank 66/36638.2%31 games · rank 282/366
Three-point attempt share45.4%32 games39.2%31 games
Two-point FG%54.8%32 games · rank 79/36654.1%31 games · rank 303/366
Three-point FG%38.3%32 games · rank 14/36634.2%31 games · rank 228/366

Defense reports opponent rates, except turnovers forced. Each statistic pools its own valid numerators and denominators; game counts can differ. Three-point attempt share describes style and has no favorable rank.

Open Buffalo’s full dossier →
Stat definitions and denominators

Effective FG%: (FGM + 0.5 × 3PM) / FGA.

Turnovers / possessions: Turnovers / estimated possessions.

Offensive rebound rate: ORB / (ORB + opponent DRB).

Free-throw attempts / FGA: FTA / FGA; displayed per 100 field-goal attempts.

Rates are pooled from the source records, not averages of game percentages. Rankings compare the 366 programs in the model field. Some source games include opponents outside Division I.

The pace question.

C Michigan recorded 69.3 possessions per 40 minutes across 31 paired games; Buffalo recorded 67.4 across 32. The adjusted forecast for this matchup is 68.8. Identify which team’s personnel can dictate the first pass, early offense and transition balance; the historical tempo gap alone does not identify a tactical winner.

Five-v-five archive / source-native handoff

See which lineups actually shared the floor.

The matchup-stint archive keeps each publisher-native five-player group beside the opposing five, with possessions, repeat games and scoring margin. Search these teams to turn the factor question into a film assignment; the source IDs remain separate from the forecast and recruiting identities.

03 / Prior meetings / exact source IDs

What happened when these programs met?

2025–26 source schedule

This is a descriptive look at completed meetings where the home and away source team IDs matched exactly in the scouting archive. It is context for preparation, not a head-to-head forecast; a missing row means this source edition did not retain a qualifying meeting.

1Recorded meetings
01Buffalo record
-5.0Average Buffalo margin
Feb 28, 2026Latest shown
DateLocationScoreResultMarginEvidence
Feb 28, 2026Game 401814588Buffalo home7075L-5.0ESPN source ↗
04 / Schedule-adjusted lens

What survives a harder schedule?

These four-factor role estimates adjust for opponent and season recency using the same team-and-venue design as the efficiency model. A positive gap means the offense’s favorable rate is stronger than the opposing defense’s corresponding rate. This is a descriptive lens for preparation, not a new forecast or a claim about the next rotation.

C Michigan offense vs Buffalo defense.

Adjusted rateC MichiganOffenseBuffaloDefenseFavorable gap
Effective FG%49.9%54.1%-4.2 ppfavors this defense
Free-throw attempt rate33.9%32.4%+1.6 ppfavors this offense
Offensive rebound rate29.7%31.2%-1.5 ppfavors this defense
Turnover rate18.0%17.2%-0.8 ppfavors this defense

For turnover rate, lower offensive turnover rate and higher forced-turnover rate are favorable. The other rows favor more efficient offense and lower opponent rates.

Buffalo offense vs C Michigan defense.

Adjusted rateBuffaloOffenseC MichiganDefenseFavorable gap
Offensive rebound rate27.2%32.2%-5.0 ppfavors this defense
Effective FG%51.1%52.4%-1.2 ppfavors this defense
Free-throw attempt rate35.5%35.9%-0.5 ppfavors this defense
Turnover rate17.9%17.5%-0.4 ppfavors this defense

For turnover rate, lower offensive turnover rate and higher forced-turnover rate are favorable. The other rows favor more efficient offense and lower opponent rates.

05 / Historical workload, then roster verification

Who carried the old possessions?

The three largest recorded minute totals on each 2025–26 team, among players with at least 200 minutes. These are historical contributors, not a projected 2026–27 rotation. Follow the player history and verify availability before assigning matchups.

C Michigan / 2025–26

Nathan ClaerbautC · 31 games · 903 total minutes
29.1 MPG12.3 PPG20.5% est. usage59.3% TS

Usage sample: 31 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Tamario AdleyG · 31 games · 900 total minutes
29.0 MPG12.8 PPG23.9% est. usage56.0% TS

Usage sample: 31 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Phat Phat BrooksG · 31 games · 870 total minutes
28.1 MPG11.0 PPG23.9% est. usage50.1% TS

Usage sample: 31 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Inspect the historical shot map →

Buffalo / 2025–26

Ryan SabolG · 32 games · 1,159 total minutes
36.2 MPG18.8 PPG25.7% est. usage59.0% TS

Usage sample: 32 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Noah BatchelorG · 32 games · 985 total minutes
30.8 MPG7.3 PPG11.1% est. usage69.0% TS

Usage sample: 32 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Angelo BrizziG · 32 games · 954 total minutes
29.8 MPG14.5 PPG24.2% est. usage62.1% TS

Usage sample: 32 games. An opportunity estimate from matched minutes and boxes, not measured on-court possessions.

Inspect the historical shot map →
06 / Source roster observation

How much old workload is represented?

2026–27 listing · unconfirmed

This is an exact-ID comparison between the published 2026–27 roster listing and recorded 2025–26 minutes. It describes what the source currently represents; it does not establish eligibility, availability, a returning decision or a projected rotation, and it does not change the forecast.

C Michigan

Players listed15
Roster shape · G / F / C8 / 5 / 2
Listed as same program12
New to source dataset3
Returning IDs with prior minutes12
Prior minutes represented6,200 · 100.0%
Role workloadPrior minReturningIncomingUnrepresented
Guard8 listed3,707100%
Forward5 listed1,590100%
Center2 listed903100%

Roster shape uses only the source-reported position labels on this listing; it is descriptive and does not identify a viable lineup or positional matchup.

Role rows use prior production attached to the currently listed source records. They are a narrower sample than the team denominator above; a missing or unclassified listing remains unknown.

The percentage uses 6,200 recorded 2025–26 minutes for this program as its denominator. A missing listing is not evidence of a departure; a new entry is not proof of a freshman or transfer.

Review roster observations and announcements →Build a role shortlist for C Michigan

Buffalo

Players listed14
Roster shape · G / F / C9 / 4 / 1
Listed as same program13
New to source dataset1
Returning IDs with prior minutes13
Prior minutes represented6,450 · 100.0%
Role workloadPrior minReturningIncomingUnrepresented
Guard9 listed4,894100%
Forward4 listed883100%
Center1 listed673100%

Roster shape uses only the source-reported position labels on this listing; it is descriptive and does not identify a viable lineup or positional matchup.

Role rows use prior production attached to the currently listed source records. They are a narrower sample than the team denominator above; a missing or unclassified listing remains unknown.

The percentage uses 6,450 recorded 2025–26 minutes for this program as its denominator. A missing listing is not evidence of a departure; a new entry is not proof of a freshman or transfer.

Review roster observations and announcements →Build a role shortlist for Buffalo
07 / Dated school evidence

What changed after those box scores?

Announcement board reviewed Sep 8, 2026

This board covers a limited set of schools. An announced addition does not by itself prove game-day availability. Later statements remain in the player’s timeline; a missing entry is not evidence that a player left or that a roster stayed unchanged.

C Michigan

This program is outside the reviewed school-announcement coverage in this edition. Check its official roster and availability reports.

Open the recruiting evidence board →

Buffalo

This program is outside the reviewed school-announcement coverage in this edition. Check its official roster and availability reports.

Open the recruiting evidence board →
09 / Compare only matching records

The forecast and market trail.

Published ledger edition Sep 15, 2026. Only a record matching this game’s model, participants, start and prediction can appear here. Quotes are last qualifying observations, not verified closing lines or a live price feed.

Forecast registered . Status: Excluded · Start time unconfirmed.

No qualifying timestamped market comparison is available for this forecast in this ledger edition. A model-versus-market edge cannot be reported.

For spreads, a positive difference is model home margin plus the observed home spread; for totals it is model total minus the line; for moneylines it is model home probability minus the two-way normalized market probability. These are model disagreements, not placed bets or profit estimates.

Inspect this game’s complete forecast and source history →

Retained reading snapshots of this brief →

Your preparation / Private browser notes

Take it into the film room.

Loading browser notes…

How to read this brief.

The preseason model’s independent 2025–26 evaluation covered 5,734 games, with 10.26-point margin MAE and 79.1% empirical coverage for nominal 80% ranges. The weekly model experiment is separate from these forecasts.

Historical scouting and player statistics: Sep 12, 2026 source edition. Model: basketball-efficiency-v2-b7df2377c580. Training cutoff: 2026-09-12T12:53:26.061472Z. Model notebook and source receipts →

Bulk observations: SportsDataverse, whose publisher labels its datasets CC BY 4.0. Independent calculations and text templates: Silvermine. School announcements are linked individually. The film questions are hypotheses to verify, not measured descriptions of a current lineup.