Columbus Crew @ CF Montréal
CCAwayColumbus Crew38%
@CMHomeCF Montréal36%
Model PickColumbus Crew 38%Market edge — disclosed after closing odds are collectedEvidence 12 structured · 1 qualitative
AI Summary grounded in real data · zero hallucination
We project a win for Columbus Crew, holding a slight edge in our model probabilities. While 'Home Advantage' favors Montreal, the 'Season Form (record)' and 'Defense (goals against/game)' metrics support the visitors. As our model does not consistently beat the market long-term, please approach this pick with caution.
Logistic-regression statistical model — predicts Columbus Crew vs CF Montréal by accumulating the log-odds contributions of 12 proven structured metrics. Only structured metrics enter the probability; qualitative and folklore factors get a label only. Predictions are sealed with SHA-256 before the game.
Model factor contribution How the 36% home win was derived
Home Advantage+3.7
Season Form (record)-2.7
Attack (goals/game)-1
Defense (goals against/game)-2.2
Shots on Target %+0.7
Baseline 50% → final home win48.4%
※ This is the head-to-head home-win probability (home vs away). After carving out the 26% draw share, the 3-way home probability shown above is 36%.
Each factor's log-odds contribution accumulated · only structured metrics reflected (qualitative/folklore excluded) · each %p converts log-odds to probability, so it's non-linear — a simple sum may not exactly equal (final − baseline)
Structured metric analysis 12 metrics · fully disclosed
| Metric | Analysis (home vs away) | Source |
|---|---|---|
| Home Advantage | CF Montréal at home — MLS home-field edge applied | ESPN MLS |
| Season Form (record) | Home CF Montréal 5-6-14 vs Away Columbus Crew 6-5-14 | ESPN MLS |
| Home/Away Split | Home team's home record ? · away team's away record ? | ESPN MLS |
| Recent Form | Home LLLLD vs Away WLLWL (recent results) | ESPN MLS |
| Attack (goals/game) | Home 1.24 vs Away 1.4 — goals per game | ESPN MLS |
| Defense (goals against/game) | Home 2.04 vs Away 1.68 — goals conceded per game (lower is better) | ESPN MLS |
| Shots on Target % | Home 37.27% vs Away 32.88% — share of total shots on target (shot quality) | ESPN MLS |
| Finishing (conversion rate) | Home 10.3% vs Away 13.4% — shot-to-goal conversion (finishing efficiency) | ESPN MLS |
| Possession % | Home 48.47% vs Away 57.35% — ball control (game-control style) | ESPN MLS |
| Pass Accuracy % | Home 82.58% vs Away 88.11% — build-up stability | ESPN MLS |
| Tackle Success % | Home 67.91% vs Away 66.91% — pressing-defense efficiency | ESPN MLS |
| xG (Expected Goals) | Not available on ESPN's free tier — will be added when a paid xG source is connected. Currently proxied by shots on target and conversion rate | ESPN MLS |
Qualitative & folklore factors 1 factors · fully disclosed
Qualitative · Mutual Slump Struggle
Both sides are currently struggling in league play and are desperate for a spark. They enter this match looking to build momentum following recent cup victories.
Basis: LLM · news · struggling MLS sides seek spark after cup wins
12 source headlines (auto-collected · basis disclosed)
- Columbus Crew visit CF Montreal as struggling MLS sides seek spark after cup wins
- Columbus Crew visit CF Montreal as struggling MLS sides seek spark after cup wins
- Columbus Crew visit CF Montreal as struggling MLS sides seek spark after cup wins
- CF Montréal hosts Columbus Crew on Saturday at Stade Saputo
- MLS success hard to come by as Crew visit CF Montreal
- CF Montreal vs Columbus Crew: Where to watch, live stream, TV channel and kick-off time | Goal.com US
- MLS success hard to come by as Crew visit CF Montreal
- CF Montreal v Columbus Odds
Frequently asked questions This game's prediction summary
Columbus Crew vs CF Montréal — who's more likely to win?
SealedSlate's logistic-regression statistical model gives Columbus Crew a 38% chance to win (Home 36% · Draw 26% · Away 38%). MLS · 2026-09-20 08:30 KST.
How was the 48.4% home-win probability calculated?
Starting from a 50% baseline, we additively accumulate the log-odds contributions of 12 proven structured statistical metrics to reach the final 48.4%. Each factor's %p contribution is fully disclosed on the page; qualitative and folklore factors are not reflected in the probability.
Can this prediction be altered later?
No. Every prediction is sealed with a SHA-256 hash before the game starts and permanently recorded in the track record, making after-the-fact changes impossible. SealedSlate runs a transparent ledger that never inflates the hit rate and leaves missed predictions exactly as they were.
We don't compete on hit rate — every factor is fully disclosed and labeled. Only 📊 structured stats feed the probability (each factor's contribution shown in %p); ⚔️🔮 are context to weigh. Predictions are locked and sealed before the game — verify the hits directly on the track record.