This site contains betting-related content and is intended for users aged 18 and over. You must be 18 or older to gamble.
Jay Thompson · 30 July 2026
Last updated 31 July 2026. Seven days out. Predictions data is in. Let's get into it.
Stoke City vs Oldham Athletic. Carabao Cup. Saturday 8 August, 2pm kick-off. On paper this looks like a comfortable home win and you move on with your life. Stoke are Championship level, Oldham are down in the National League or thereabouts, and that should be that. Job done. Goodnight.
Except... the data is doing something funny. And I can't ignore it. So bear with me.
Look, Stoke's Carabao Cup record coming in is fine. One game, one win, one goal scored, nothing conceded. Lovely. But look at the fixtures in their actual league campaign and it falls apart completely. In their last five Championship games, Stoke have won zero, drawn one, and lost four. They've shipped eleven goals and only scored three. Eleven goals in five games. That is not a defence, that is a suggestion of a defence.
The shots numbers don't help either. Sixteen shots per game but only two on target per game on average. So they're having the ball, they're getting into positions, and then absolutely nothing is happening. The conversion is non-existent. That 54 percent possession figure with a 0 percent clean sheet rate in that run tells you everything. Lots of the ball, not a lot of end product, and wide open at the back.
Oldham are hardly setting the world alight themselves. Away from home in their own league, they've lost five from five in the last five. Five losses. Zero wins. Zero draws. They've conceded eleven away from home and scored six. Not great reading. At all.
But here's the thing. At home in their own league? Completely different animal. Last ten home games, Oldham have gone W-L-D-D-W-W-W-W-W-W. Seven wins in ten at home, 16 goals scored, only four conceded, and a 70 percent clean sheet rate at home. Now obviously this isn't a home game for them. But it does tell me this squad has some quality in it somewhere. They're not just a team turning up to make the numbers.
Right. So the model we use here at SportSignals has given Oldham Athletic a 40.1 percent chance of winning this game. Forty percent. That is not a small number for a team coming from a lower division away from home in a cup tie.
Now the bookies have them at 5.5 with bet365. That price implies roughly an 18 percent chance of Oldham winning. The model says 40 percent. That gap, 25.9 percentage points, is the kind of edge that makes you sit up straight.
I'm not saying back it with your mortgage. The confidence rating on the Oldham win signal is only 40 percent, which is the model basically saying "I think this but I'm not sure." Fine. Honest. I respect that more than false certainty.
But seven to one? At 7.0 with Boylesports? On a team the model gives a genuine four in ten shot? That is interesting, mate. You heard it here first.
This is where it gets a bit contradictory and I love it. The model likes BTTS Yes at 57 percent probability. The market has it at 1.8, implying 55.6 percent. Tiny edge. Not a screaming value play but the model agrees with the direction.
Look at Oldham's overall numbers and BTTS has landed in 70 percent of their last ten games and 80 percent of their last five overall. Stoke are conceding goals for fun right now. Eleven against in five league games. The conditions for BTTS feel right even if the price is a bit skinny.
The Under 2.5 signal is also flagged at 2.4 odds, model giving it 43 percent. Honestly I'm less convinced by that one. Oldham's last five overall games have been over 2.5 one hundred percent of the time. One hundred. Every single one. That doesn't scream Under 2.5 to me. The model sees some value there but my gut says goals.
Now I did actually look at the numbers for once, as Marcus would say, and the xG stuff made me laugh. The model has Oldham's xG for at 2 and xG against at 4. The xG, for those of you who don't know, stands for... actually I have no idea what it fully stands for but someone in a podcast once tried to explain it to me for forty five minutes and I still don't fully get it. What I do know is that those numbers suggest Oldham are creating less than they're conceding, which, yeah, checks out.
Right so here's where I have to be straight with you. The injury data on this one is completely clean. Nothing flagged. No reported absences from either camp at this stage. That either means everyone is fit and raring to go, or the information just hasn't come through yet seven days out. Keep an eye on both clubs' socials through the week because Carabao Cup team news tends to drop late and rotation is always a factor in these early rounds. Stoke will almost certainly rest players. That could work in Oldham's favour more than people think.
Look, I'll be honest. Stoke should win this and probably will. The home advantage, the class gap on paper, the fact that Oldham away from home have been dreadful... all points to a Stoke win. The 1.4 on Stoke is too short for me to touch. No way.
But there is genuine chaos energy around this fixture. Stoke are leaking goals. Oldham score goals even when they lose. And the model thinks an upset is far more likely than the market does.
I'm going big on this... BTTS Yes at 1.8. It's not the flashiest pick but both these teams are giving goals away right now and Oldham have scored in games even when getting battered away from home. Six away goals in five losses tells you they don't just roll over.
For the chaos merchants among you, don't @ me, but Oldham to win at 7.0 with Boylesports is the punt of the week if you fancy a fiver on something with proper upside. A 40 percent shot at sevens is value. Simple as. The vibes say Stoke win comfortably. The data says it might be scenes.
Back to the drawing board if it goes wrong. It usually does. But that's football, innit.
BTTS Yes, bet365, 1.8
Oldham Athletic to win (cheeky punt), Boylesports, 7.0
