Our positive-edge calls: four from thirteen

Thirteen BAI8 calls on the public record carried a positive edge when we published them. Four won and nine lost. Our own chances on those calls added up to 5.8 expected wins, so we are about two wins short of what we said. One unit on each at the quoted price would be 3.68 units down. That is a bad start, it is on the record, and it is not yet evidence either way.

All thirteen

Date	Fixture	Pick	Our chance	Price quoted	Edge	Result
6 Oct	Braves v Dodgers (MLB)	Dodgers	51.6%	1.99, Matchbook	+1.3 pts	Won
7 Oct	White Sox v Guardians (MLB)	White Sox	52.4%	1.92, Betfair	+0.4 pts	Lost
8 Oct	White Sox v Guardians (MLB)	White Sox	52.3%	2.03, 1xBet	+3.1 pts	Lost
9 Oct	Málaga v Espanyol	Espanyol	38.9%	2.86, Betfair	+3.9 pts	Lost
9 Oct	Blue Jackets v Penguins (NHL)	Penguins	51.3%	2.06, Betfair	+2.7 pts	Lost
10 Oct	Gençlerbirliği v Amedspor	Amedspor	40.6%	2.67, 1xBet	+3.1 pts	Lost
10 Oct	Aston Villa v Brentford	Brentford	37.5%	2.69, 1xBet	+0.3 pts	Lost
11 Oct	Elche v Celta Vigo	Celta Vigo	36.6%	2.75, Coolbet	+0.3 pts	Lost
11 Oct	Crystal Palace v Nottingham Forest	Palace	36.9%	2.80, Betfair	+1.1 pts	Won
11 Oct	Nice v Strasbourg	Strasbourg	37.7%	3.07, 1xBet	+5.1 pts	Lost
11 Oct	Seahawks v 49ers (NFL)	49ers	52.2%	2.58, Betfair	+13.5 pts	Won
12 Oct	Brewers v Dodgers (MLB)	Brewers	51.6%	1.95, 1xBet	+0.4 pts	Won
12 Oct	Levante v Sevilla	Sevilla	37.6%	2.84, Betfair	+2.4 pts	Lost

The average edge was 2.9 points at an average price of 2.48. How each of these was graded is in how a BAI8 call gets settled on the record.

Why four from thirteen proves little

	Value
Expected wins (sum of our chances)	5.8
Actual wins	4
One standard error on the win count	1.8
Flat one-unit result at quoted prices	−3.68 units

Being 1.8 wins short with a standard error of 1.8 is ordinary. If every one of our chances were exactly right, four wins or fewer would still happen about 24% of the time. Seven of the thirteen were football picks at 36% to 41%, which lose most of the time even when the chance is right. A record this short cannot separate a real 3-point edge from none at all; the arithmetic is in how many bets before a record means anything.

Where it could genuinely be wrong

We would rather say this before the numbers force it. Three things would make these edges smaller than we stated:

• Stale quotes. Several prices were days old at kick-off. If the best price had gone by then, the edge was never really available.
• Draws. Six of the nine losses were football picks in three-way markets, where a draw also loses (Sevilla's 3–3 on Monday was one). If we underrate draws, it shows up exactly here.
• Small edges are thin. Four of the thirteen were under half a point. At that size the edge is inside quote noise.

How the chance itself is built is in how BAI8 prices a fixture.

What this does not tell you

It does not tell you the edges are real, and it does not tell you they are not. It covers thirteen calls in one week across four sports. We will publish the same table as it grows, wins and losses alike.

18+. Bet only what you can afford to lose.