TACTICAL
The Sabres own the underlying play
The Sabres own the underlying play — 48.3% 5v5 xG vs 33.9%.
Preview · CHI at BUF
The Sabres are the model's lean over the Blackhawks — with PDO regression in play
The model's leanSat, Oct 3
Written from the numbers
The read
Edited for clarity
◆ By the ModelOct 3, 9:02 AMEdited for clarityDrafted from NHL play-by-play by Backcheck's model, worded for clarity by an AI editor that may not change a number. Nobody at Backcheck was at the rink. How this works
Buffalo is the model's clear favorite — a 2.9 to 3.5 projection with the Sabres at 61% win probability. The case is workload, not luck.
The Sabres own the underlying play at 5v5, carrying 48.3% expected goals to Chicago's 33.9%. Territory compounds the gap: Buffalo drives 56.5% of shot attempts while the Blackhawks manage 42.2%, pressing Chicago into a territorial disadvantage the hosts cannot overcome.
Both teams are running cold on conversion. Chicago's PDO of 0.937 sits in regression territory. Buffalo is colder at 0.891 — a clear sign that positive regression waits if the Sabres keep generating looks. This is the edge: Buffalo outworks Chicago, and the score eventually reflects the workload.
The goaltending matchup presents no separation. A. Soderblom arrives at −2.1 expected goals against per game in 1 GP on 81.2% save percentage. U. Luukkonen is marginally worse: −2.2 expected goals against in 1 GP, 77.3% save percentage. This is not where the game is decided.
Discipline at the rims determines the margin. Chicago's sharpest scoring zone is the slot, where they have 4 goals on 29 shots (14%); Buffalo must deny that space. Buffalo's edge is the left circle — 4 goals on 21 shots (19%) — where they own both volume and finish. How Chicago defends the left circle determines whether Buffalo's expected-goals advantage becomes a lopsided scoreline.
The model favors Buffalo because the fundamentals align: territory, volume, and expected production all point the same direction. It is not declaring the Sabres elite — only that they are better than Chicago tonight, and the numbers are coherent enough to trust.
Live from the league
The game, as it happens
Puck drop 7:00 PM ET
Once the puck drops, every goal and every shot lands here as it happens — the scorer, the clock, and the shot map with the replay — straight from the league's feed. The desk writes the recap after the final horn.
Win probability
The model's lean
BUF 61%
61%
BUF
Model favorite
39%
61%
CHI50BUF
BUF by 22 points of win probability — a clear lean. Projected score CHI 2.9–3.5 BUF; 16% chance it's tied after 60.
Source · Backcheck Poisson projection · season scoring rates
New to these numbers?
This is a preview, so every number on it is about form and rates — what each club has been doing, not what it will do tonight.
The strip above is the model's lean, built from each side's chance creation and prevention, goaltending, rest and home ice. It is not a betting line and nobody here is telling you to wager on it. Full glossary →
- Win prob
- How often this matchup ends in a win for that club, over many simulated tries. It is a lean, not a prediction — 60% teams lose plenty.
- xGF/60
- Expected goals a club creates per sixty minutes of play — its scoring-chance rate, independent of how many games it has played.
- xGA/60
- The same rate for chances it gives up. Lower is better.
- PDO
- Shooting % plus save % — a luck gauge that drifts back toward 1.000. Well above it flags a club that has been running hot.
Season rates · 5v5
Tale of the tape
NHL stats API · our xG
CHI← edge
edge →BUF
33.9%48.3%
42.2%56.5%
2.272.77
4.442.96
1.003.00
5.506.00
0.9370.891
BUF holds 5 of 6 process categories — and the luck row (0.937 vs 0.891) says CHI has been getting the bounces, which tends not to last.
Source · NHL stats API + our xG · 5v5
Form · Last 1–2
CHI
0-2-0
-9 GD · 1d rest
BUF
0-1-0
-3 GD · 1d rest
0.937
CHI PDO
Luck gauge · drifts to 1.000
0.891
BUF PDO
Luck gauge · drifts to 1.000
What decides it
Keys to the game
TACTICAL
The Sabres drive Corsi at 56.5%
The Sabres drive Corsi at 56.5% — expect territorial tilt.
TACTICALBUF
The model's clear lean
The Sabres project at 61% win probability.
WARNINGBUF
Regression watch: Sabres
The Sabres' PDO is 0.891 — running cold, due for positive regression.
Weight the underlying numbers over the recent results.
Where the edges live
- The Sabres own the underlying play — 48.3% 5v5 xG vs 33.9%.
- The Sabres drive Corsi at 56.5% — expect territorial tilt.
Regression watch
- The Sabres' PDO is 0.891 — running cold, due for positive regression.
- The Blackhawks' PDO is 0.937 — running cold, due for positive regression.
Things to watch
- The Sabres own the underlying play — 48.3% 5v5 xG vs 33.9%.
- The Sabres drive Corsi at 56.5% — expect territorial tilt.
- Goalie matchup: A. Soderblom (-2.1 GSAx in 1 GP, 81.2% SV) vs U. Luukkonen (-2.2 GSAx in 1 GP, 77.3% SV).
- The Sabres' PDO is 0.891 — running cold, due for positive regression.
Last game's lineup · confirmed at warm-up
Lineups
NHL box score
CHI
vs UTA · Thu, Oct 1
Expected in goal
A. Soderblom
1 day rest.813 last start60:00 played
Backup Spencer Knight · +0.5 GSAx this season
Forwards · 12
- R. Donato11:54
- O. Moore11:25
- J. Greenway13:52
- A. Frondell18:10
- R. Greene17:04
- C. Smith14:13
- T. Bertuzzi18:15
- N. Lardis10:18
- R. Kantserov16:36
- T. Teravainen13:00
- P. Kane19:19
- F. Nazar17:23
Defence · 6
- S. Rinzel17:30
- B. Byram24:49
- I. Cole14:52
- W. Kaiser19:49
- A. Levshunov19:51
- A. Vlasic18:18
BUF
vs CBJ · Thu, Oct 1
Expected in goal
U. Luukkonen
1 day rest.773 last start59:08 played
Forwards · 12
- Z. Benson18:34
- J. Norris16:19
- S. Carrick07:43
- P. Krebs14:58
- J.Kulich08:45
- J. Quinn20:46
- B. Malenstyn08:11
- R. McLeod16:41
- T. Thompson19:46
- N. Ostlund16:42
- J. Doan19:06
- K. Helenius15:09
Defence · 6
- M. Samuelsson18:21
- M. Grzelcyk14:27
- O. Power25:25
- R. Dahlin25:58
- L. Crevier16:14
- Z. Metsa12:54
Lineups are the players who dressed in each club's last game, with their minutes; clubs confirm tonight's at warm-up. Every player this preview names was in that lineup.
Keep going