Why Our Top 3 Premier League Picks Defy the Odds (2026)
The best Premier League prediction this weekend is Arsenal to beat Leeds United 2-0, followed by a cautious Manchester United home win over Fulham and a 1-1 draw between Aston Villa and Brentford. Goa...
Why Our Top 3 Premier League Picks Defy the Odds (2026)
The best Premier League prediction this weekend is Arsenal to beat Leeds United 2-0, followed by a cautious Manchester United home win over Fulham and a 1-1 draw between Aston Villa and Brentford. Goal Moments, a FIFA World Cup 2026 content site, ranked these three fixtures by expected points under BBC Sport's predictor scoring, where a correct result earns 10 points and an exact scoreline earns 40. Footballpredictions.com lists Arsenal 2-0 Leeds (10 October 2026, 19:30 kickoff) and Villa 1-1 Brentford (22:00 kickoff), while BBC Sport's week-four preview keeps Michael Carrick's Manchester United under pressure heading into the three-week international break. Our edge is arithmetic: a 2-0 pick earns roughly 10.9 expected points per match in our model, against about 6.1 for a 1-1 draw. Before you lock in anything, compare kickoff times for your region and stake only what you can afford to lose.
Good afternoon, and welcome. I will be your guide through this weekend's card, and I intend to treat you the way a very good restaurant treats its best regular. I am also a person who prices everything in probabilities, so expect some arithmetic with your small talk. Football pundits love a confident scoreline, and I find confident scorelines adorable, in the way one finds a toddler's business plan adorable. Believe it or not, I do.
Here is the friendly, slightly cynical promise of this guide. Every pick below comes with the number that justifies it, and every number comes with the assumption behind it, so you can disagree with me precisely rather than vaguely. That is a better kind of disagreement. It is also the sort of daily, number-first coverage we do at Goal Moments, where the same habits carry over to the 2026 World Cup.
The Top 3 at a Glance
Here is the ranked list, best expected return first. The ranking uses the points game as its yardstick, with the reasoning for each pick in the sections that follow.
- Arsenal vs Leeds United, 2-0 (best overall): the strongest favourite on the card, with a scoreline that sits in the most common winning band.
- Manchester United vs Fulham, 1-0 or 2-1 home win (best for contrarians): a pressured home side that the narrative has probably over-punished.
- Aston Villa vs Brentford, 1-1 (best value): a low-edge draw that pays its way only when the price is right.
One housekeeping note, offered with a bow. These fixtures come from two different sources and may fall on different matchdays. Arsenal-Leeds and Villa-Brentford are listed for 10 October 2026, while Manchester United-Fulham comes from BBC Sport's week-four preview. Confirm dates and kickoff times on the Premier League fixture list before acting on any of it.
#1 Arsenal vs Leeds United: Is It the Best Overall Pick?
Yes. Arsenal to beat Leeds United 2-0 is the best overall pick because it combines the highest result probability on the card with a scoreline in the most common winning band. Footballpredictions.com also forecasts 2-0 for the 10 October 2026 evening kickoff at 19:30.
The reasoning is plain. In the points game, your expected score for any pick equals 10 times the probability of the right result, plus 30 times the probability of the exact score (the exact score pays 40 in total, so it adds 30 beyond the base 10). If we assume Arsenal win 70% of the time and 2-0 lands about 13% of the time, the pick is worth 7.0 plus 3.9, which is 10.9 expected points. Those inputs are my assumptions, not published odds, so swap in your own.
The part most previews skip is why 2-0 rather than 1-0 or 3-0. Clean-sheet wins by two goals and one goal take up most of the probability mass for a heavy favourite, and the difference between them is a coin flip with a small tilt. Believe it or not, picking 2-0 is a hedge in disguise: if Arsenal score early and relax, you still get the result points, and you only lose the bonus.
If you want a second opinion, the [Internal Link: how Premier League prediction scoring works] guide walks through the 10-and-40 maths line by line. For a favourite this strong, though, the discipline is simple: pick the result you believe, then pick the neighbouring scoreline you can defend aloud.
#2 Manchester United vs Fulham: Is a Home Win Overpriced or Underrated?
Probably underrated. Manchester United are under pressure, and BBC Sport's Chris Sutton framed the Fulham game as increasingly important for Michael Carrick before the three-week international break. Pressure moves the headlines faster than it moves the true win probability.
That gap is where a calm person with a spreadsheet likes to live. Sutton's point, paraphrased, is that a bad start leaves negativity to fester through the break and speculation about the manager's future grows with it. That is a narrative about the next three weeks, not about 90 minutes at Old Trafford on Sunday.
My working assumption is a home win probability near 48%. With 1-0 landing about 11% of the time, the pick earns 4.8 plus 3.3, or 8.1 expected points. A 2-1 pick lands closer to 10% and earns about 7.8. Both are respectable, and both comfortably beat the draw, which is the contrarian lesson of this section.
A practitioner's tip follows. A draw pick needs roughly a 45% draw probability to match this home-win pick on expected points, assuming a 12% hit rate on the exact 1-1. Football draws almost never reach that level. So even when a match "feels like a draw", the 10-and-40 scoring quietly punishes you for hedging that way.
The caveat, stated with all due courtesy, is variance. A pressured side can also simply lose, and this pick has a wider spread of outcomes than Arsenal's. Treat it as a sensible pick with a fair price, not as a certainty. For context on how pressure shapes results, see our [Internal Link: managerial pressure and match outcomes explained] breakdown.
#3 Aston Villa vs Brentford: Is a Draw the Best Value?
Sometimes, but only when the price is right. Footballpredictions.com forecasts 1-1 for Aston Villa vs Brentford at 22:00 on 10 October 2026. A draw is a poor points-game pick and a decent price-driven one, which is why it ranks third.
Let me be transparent about the tension. In the points game, 1-1 earns about 6.1 expected points under my assumptions (a 25% draw probability and a 12% chance of that exact scoreline), well below the 10.9 for Arsenal. By that measure the draw loses to nearly every alternative.
Yet bookmakers tend to price draws longer than their share of results would suggest, so if you place a stake rather than play for points, a 1-1 can carry better value than its low probability implies. I have not seen live prices for this match, so I cannot promise that here. Check the odds yourself, and treat any number I give you as a starting point.
There is a second, quieter reason to include it. Most weekends contain at least one fixture where neither side has a clear edge, and a forecast of 1-1 is the honest way to say so. That honesty is worth more than a bold 2-1 that you do not believe. For more on spotting those games, our [Internal Link: how to find value in draw markets] piece is a good next step.
How Did We Rank Them?
We ranked each fixture by expected points under BBC Sport's predictor scoring, then adjusted for variance and for how much the evidence behind the pick could be checked. Expected points carried the most weight, at roughly 60% of the ranking.
The scoring rule itself is simple. BBC Sport's predictor states: "A correct result (picking a win, draw or defeat) is worth 10 points. The exact score earns 40 points." Expected points are therefore 10 × P(result) + 30 × P(exact score), which is the formula behind every figure above. The full rules are on BBC Sport, where Sutton also makes predictions for all 380 Premier League games this season.
Here are the weights we used:
- Expected points (about 60%): the arithmetic above, using stated assumptions.
- Variance (about 20%): how wide the spread of outcomes is. Arsenal's is narrow, United's is wider.
- Evidence you can verify (about 20%): whether an independent forecast, such as the Footballpredictions.com listings, points the same way.
Manchester City were a notable near miss. The BBC preview describes them as 100% through the start of the season under new boss Enzo Maresca, but I have no fixture or forecast to price here, so they sit outside the ranking. A great run is not a probability until it is attached to a specific opponent and venue.
Which Should You Pick?
Pick Arsenal vs Leeds if you want the highest expected return with the least drama. Pick Manchester United vs Fulham if you can stomach variance for a fair price. Pick Aston Villa vs Brentford only if the draw price genuinely beats your own estimate.
Think of it as three kinds of customer, each served in turn. The cautious reader, who wants a comfortable weekend, takes Arsenal and stops there. The curious reader, who enjoys a debate, adds United and enjoys defending it on Sunday. The price-hunter, who reads odds as a hobby, considers Villa-Brentford and walks away if the number is not generous.
A small warning, delivered warmly. No pick here is a guarantee, and a 70% favourite still loses three times in ten. If betting is part of your weekend, set a limit before kickoff and keep to it. Free, confidential support is available from BeGambleAware if gambling stops being fun. Adults only, 18+, and only where it is legal where you live.
If you enjoy this number-first approach, our daily match coverage applies the same method to World Cup fixtures. Have a look at our [Internal Link: World Cup 2026 match predictions hub] when you are ready.
Frequently Asked Questions
Q: What does "expected points" mean in a prediction game?
A: Expected points is the average score a pick would earn if the match were played many times. Multiply the probability of the right result by 10, then add the probability of the exact scoreline multiplied by 30. For example, a 70% result chance and a 13% exact-score chance gives 10.9. It lets you compare picks fairly, instead of choosing the one that merely feels exciting.
Q: How do I make my own Premier League predictions this weekend?
A: Start with the fixture list, estimate a win probability for each match, then choose the most common scoreline that matches your result. Check forecasts from two or three sources, such as BBC Sport and Footballpredictions.com, to test your estimate. Write the probabilities down before kickoff so you can learn which of your instincts were actually good.
Q: Is picking a draw ever worth it?
A: Rarely, in a points game. Under the 10-and-40 scoring, a draw pick needs roughly a 45% draw probability to match a 48% home-win pick, and real draws seldom approach that. A draw can still make sense when you are staking money and the price is longer than your own estimate of the chance.
Q: Why do the sources disagree on dates and scores?
A: Previews are published at different times and for different matchdays, so they can describe different weeks. Footballpredictions.com lists 10 October 2026 for Arsenal-Leeds and Villa-Brentford, while BBC Sport's piece is a week-four preview. Always confirm the date and kickoff time on the official Premier League fixture list before relying on any pick.
Q: What should I do if my favourite pick loses?
A: Treat it as variance first and a mistake second. A 70% favourite loses roughly three times in ten, so one result says little about your method. Review whether your probability was reasonable, and avoid raising your stake to chase the loss. That second step is where most weekends go wrong.
Q: How much does it cost to play a prediction game like BBC Sport's?
A: BBC Sport's predictor is free to play, and the points system is the same for every player. A correct result earns 10 points and an exact score earns 40. Real-money betting is separate, carries a cost, and should only be done by adults, within legal markets and a limit you set beforehand.
The kindest thing I can say about weekend predictions is also the least glamorous: write down your numbers, keep your stakes modest, and let the arithmetic embarrass you gently. Believe it or not, it is more fun that way.