Per90
Loading live football data…
Per90
Loading live football data…
DFB Pokal · Germany
lineups not announcedBest edge
—
no call published
Angles
—
none found
Referee
—
Martin Petersen · fouls v league
No value call published on this fixture
No edges published on this fixture. An edge is published only when a bookmaker's price beats the model's fair price by that market's full bar — most fixtures never produce one. The value board explains every gate a candidate has to clear.
No lineup yet. Minutes come from each player’s start rate, which is the widest of the three states — and why rows are marked provisional.
Published at the price shown, scored against the closing line.
No edges published on this fixture. An edge is published only when a bookmaker's price beats the model's fair price by that market's full bar — most fixtures never produce one. The value board explains every gate a candidate has to clear.
The feed has published no teamsheet for this fixture yet.
Lineups are captured every fifteen minutes from three hours before kick-off. An expected XI normally lands well before that; a confirmed one about an hour out. Until one does, every projection on this page takes its minutes from each player’s start rate — which is what the provisional marks in the props table mean.
Last six completed matches per side, newest first. Half-time score in brackets where the feed carries one.
Model projections. Nothing here has been compared to a bookmaker.
How this is calculated →No projections for this fixture yet. The projection job runs against fixtures inside the odds window, and only for players with at least 10 completed matches on record.
Form against the lines the books are pricing, what the opposition concede, and the referee against his league.
No angles on this fixture. An angle needs priced lines in the feed and a run long enough to be worth stating — most fixtures produce some once the books put player props up.
§5.1: the dominant covariate for fouls, and the reason two identical players price differently on different days.
0 matches on record
Fouls per game
—
Cards per game
—
Cards, home side
—
Würzburger Kickers
Cards, away side
—
1. FC Köln
Fouls — every foul committed by both teams in a match he refereed, averaged over his completed matches. The match being previewed is never in it. Under 15 matches on record, so the model itself falls back toward the league mean here.
Cards — any card, yellow or red, both teams. The same definition the card props settle against, so this figure and the Cards market count the same events. A dismissal that the feed records as a second yellow counts as two.
Both league averages are the mean across every competition Per90 covers, not this fixture's league alone — the fouls comparison has always been that, and the cards comparison is built the same way so the two percentages beside each other mean the same thing.
What the numbers on this page are, and what they are not.
Not tips and not prices. Each row is the model's expected count for one player in this match — how many fouls, tackles or shots it thinks he records — plus the full spread of outcomes around it. Nothing in the projections has been compared to a bookmaker; that comparison is what the Value tab is.
Step 1
His own record
Every completed match in the dataset, as a rate per 90 minutes. The n column is how many matches that rests on.
Step 2
Pulled toward his position
A thin record gets pulled hard toward other players in his position, in his league — a goalkeeper toward goalkeepers, not toward the outfielders around him. How hard is fitted per market, not chosen: tackles pull about four times harder than shots on target.
Step 3
This fixture's conditions
How much of the stat the opponent concedes, what the referee’s matches look like, and home or away. This is where a projection stops being a season average.
Step 4
Scaled to expected minutes
A rate becomes a count only once you say how long he plays. 60 minutes carries two-thirds the exposure of 90, and the distribution is built at that exposure.
How to read each column
Counts are modelled as negative binomial rather than Poisson, because fouls and tackles are overdispersed and a Poisson would systematically underprice the tails — which is exactly where over bets live. Every market on this page cleared its calibration gate before it was allowed to produce a projection: when the model says 60%, it lands within a point or so of 60% out of sample.
The full write-up — every gate, every fitted parameter and what the model refuses to price — is on the methodology page.