I Tested 4 Scoreline Picks: 2-0 Beats 1-1 on Value
I Tested 4 Scoreline Picks: 2-0 Beats 1-1 on Value
The best Premier League prediction this weekend is the one with the highest expected value, not the boldest scoreline. Football Compass compared four common picks (2-0, 1-0, 2-1 and 1-1) under BBC Sport's prediction scoring, where a correct result earns 10 points and an exact score earns 40, and found that a home-win scoreline beats a draw pick by roughly two to one. For the 10 October 2026 fixtures, footballpredictions.com lists Arsenal 2-0 Leeds United, kicking off at 19:30, and Aston Villa 1-1 Brentford, kicking off at 22:00. Using a simple Poisson model with illustrative expected goals of 2.0 versus 0.8, a 2-0 or 1-0 pick returns about 10.2 points per game, while 1-1 returns about 5.0. Draws are the trap: they feel safe yet pay least. Before locking in your picks, check the probability of the result first, then choose the scoreline that sits inside it.
Have you ever wondered why a scoreline you were "sure about" felt so right and still lost you points? I did, around 1 a.m. last Thursday, with a spreadsheet open and a cold cup of tea (I know, I know, the maths nerd's idea of a wild night). So I stopped trusting my gut and tested four scorelines against the same scoring rule, one at a time, with the same assumed match strength. What follows is the honest result, including the bits that surprised me. Think of it as a late-night chat between friends who both want to stop guessing. By the end you will know which scoreline earns the most over a season, why draws are quietly expensive, and how to read the weekend's published picks as probabilities instead of vibes. Football Compass built this walkthrough for fans who follow the 2026 World Cup afterglow into the club season and want a sharper routine for weekend predictions.
Ready to see how the numbers play out across a whole weekend? Here is the next step.
Before 2025: How Did Premier League Predictions Work?
Before 2025, weekend Premier League predictions were mostly a single pundit's scoreline list with a short rationale and no shared scoring system. Readers could agree or argue, but nobody could measure who was actually better over a season, so confidence was cheap and accuracy was rarely audited.
In that older format, a column might tell you Manchester United would edge Fulham 2-1, give two lines of reasoning, and move on. Next week, nobody remembered whether the call landed. (You know how it is: a bold prediction is memorable for ten minutes and forgotten by Monday.) The missing ingredient was a scoreboard. Without points, there was no way to compare a pundit against a coin flip, against a bookmaker's implied probability, or against the readers themselves. There was also no shared language for the thing I care about most: the difference between being right and being right for good reasons. A pick that wins at 8% probability and a pick that wins at 65% look identical in a results table, yet they say very different things about the thinking behind them.
That is why the old format produced so much noise. Scorelines were treated as opinions rather than probability statements, and the payoff for exactness was never spelled out. Once you see a prediction as a bet on a distribution of outcomes, the whole genre looks different, and the newer formats start to make a lot more sense.
The 2026 Shift: What Actually Changed?
The 2026 shift is accountability. BBC Sport now has Chris Sutton predicting all 380 Premier League games against AI, readers and guests, with points awarded per fixture, while sites like footballpredictions.com publish a scoreline first and add analysis nearer kickoff. Predictions have become scoreboards you can check.
The BBC Sport weekly prediction column is the clearest example. In week four, Sutton faced guitarist and Manchester City fan Johnny Marr, whose new album The Age Of Everything is out on 2 October. The same piece noted pressure on Manchester United manager Michael Carrick ahead of the Fulham game on Sunday, which is a reminder that predictions now sit inside a running narrative rather than a one-off list. The most popular scoreline picked by readers for each game feeds the scoreboards and tables at the bottom of the page, so the crowd itself becomes a competitor. Meanwhile, the footballpredictions.com Premier League hub lists its picks early and states plainly that "the full analysis will be added closer to kickoff." Both approaches share one idea: commit to a number publicly, then get scored.
- Public scoring: every pick earns or loses points, so track records exist.
- Competitors beyond humans: AI, readers and guests all sit on one leaderboard.
- Staged publishing: the scoreline arrives first, the reasoning arrives later, so you can judge the call before the narrative colours it.
If you like seeing how a pick holds up before kickoff, take a look at the full toolkit.
What Changed for Players?
Players now face a scoring rule that rewards precision: under BBC Sport's system a correct result earns 10 points and an exact score earns 40. That four-to-one payoff means the best pick is not the likeliest result alone, but the scoreline that maximises probability-weighted points.
The BBC states it directly: "A correct result (picking a win, draw or defeat) is worth 10 points. The exact score earns 40 points." I read the 40 as the total for an exact score, not an add-on to the 10. If it stacks to 50, the ranking below stays the same and the gap widens a little, so the conclusion survives either reading. Here is where the maths gets fun (yes, I said fun). The standard way to model goal counts is the Poisson distribution, and I fed it illustrative expected goals of 2.0 for the home side and 0.8 for the away side. Those numbers are my own assumption, not a forecast for any real match. They give a home win of about 65%, a draw of about 20% and an away win of about 14%. The expected points per game, using 10 for the result and 40 for the exact score, come out like this:
- 2-0 home win: exact-score chance about 12.2%, expected return about 10.2 points.
- 1-0 home win: exact-score chance about 12.2%, expected return about 10.2 points.
- 2-1 home win: exact-score chance about 9.7%, expected return about 9.5 points.
- 1-1 draw: exact-score chance about 9.7%, expected return about 5.0 points.
Notice something a typical preview would never tell you: 2-0 and 1-0 are exactly tied here, because a Poisson distribution with a whole-number mean of 2.0 makes one goal and two goals equally likely. Nudge the favourite to 2.5 expected goals and 2-0 pulls ahead. So the real question is not "how many goals," it is "how strong is the favourite." (Tiny detail, big difference, and that is precisely the kind of thing I cannot leave alone.)
What Does This Mean for This Weekend's Picks?
It means treating each published scoreline as a probability statement. Arsenal 2-0 Leeds United sits inside a likely home win, so it scores well on expected value. Aston Villa 1-1 Brentford is a defensible draw call but carries a lower expected return per game under 10/40 scoring.
Per footballpredictions.com, both fixtures are listed for 10 October 2026, with Arsenal against Leeds United at 19:30 and Aston Villa against Brentford at 22:00 (check the kickoff time in your own time zone, since listings vary). Here is how I would read them. The Arsenal pick is a favourite scoreline, so the question is only whether you prefer 2-0 or 1-0, and that depends on how strongly you rate Arsenal's attack against Leeds' defence. The Villa pick is more interesting. A 1-1 call needs the draw to be unusually likely before it beats a home-win scoreline, and in my test the draw returned roughly half the points of the favourite picks. That does not make it wrong (football refuses to follow models), but it does mean you should hold a draw only when you have a specific reason, such as two sides that both defend deep and rarely open up.
There is also a contrarian point worth keeping in mind. If your league scores each pick in absolute points, as the BBC does, copying the crowd's most popular scoreline costs you nothing. If your pool ranks you only against friends, a slightly less popular but still sensible scoreline can separate you from the pack. Know which game you are playing before you choose.
Want a steadier routine for every matchweek, not just this one? Here is where to go next.
Three Predictions for Next Quarter
These are my own forecasts for the final quarter of 2026, built on the reasoning above rather than on any published data. Treat them as hypotheses you can check, because the entire point of this article is that predictions should be checkable.
- Draw picks will stay rare. Roughly a quarter of top-flight matches end level, yet published scoreline lists lean heavily towards home wins. Expect that gap to persist, which means a well-argued draw call will remain a genuine edge in relative leaderboards and a poor bet in absolute-points ones.
- Human-versus-AI scoreboards will spread. BBC Sport's format of one pundit against AI, readers and guests is easy to copy. I expect more outlets to publish season-long tallies, and the interesting stories will be the weeks when the crowd's most popular scoreline beats both.
- World Cup habits will carry into the club season. The FIFA World Cup 2026 put eight matches and 39 days of fan celebrations into Los Angeles alone, according to the Los Angeles host committee. Fans who learned to pick scorelines in a knockout setting will bring that habit back to weekly league football, and tools like Football Compass will be where many of them compare notes.
So, what is the one thing to remember? Choose the result first, then the scoreline inside it, and let the payoff rule (10 for the result, 40 for the exact score) decide how bold you can afford to be. If you do only that, your season will look better than it did last year (trust me on this one, I have the spreadsheet to prove it).
For more on the numbers behind these picks, read our [Internal Link: beginner's guide to expected goals], our [Internal Link: how Poisson scoreline models work], our [Internal Link: weekly World Cup and league prediction archive] and our [Internal Link: matchday betting bankroll basics].
Frequently Asked Questions
Q: What is expected value in football predictions?
A: Expected value is the average points a pick earns if you could repeat the same match many times. Multiply each possible payoff by its probability and add them up. Under BBC Sport scoring, a pick with a 65% result chance and a 12% exact-score chance returns about 10.2 points. It lets you compare a safe-looking scoreline against a bold one on equal terms, which is why I use it for every weekend slate.
Q: How do I choose a scoreline for a Premier League game?
A: Pick the most likely result first, then choose the scoreline inside it that matches the favourite's strength. Estimate expected goals for both sides, note that a favourite near 2.0 makes 1-0 and 2-0 equally likely, and move to 2-0 or 2-1 if you rate the attack higher. A simple spreadsheet with a Poisson formula handles this in about ten minutes per matchweek.
Q: Is a 1-1 draw a bad prediction?
A: A 1-1 draw is not bad, but it usually earns less than a favourite's scoreline. In my test it returned about 5.0 points per game against about 10.2 for 2-0. Choose it only when you have a concrete reason to think the draw is unusually likely, such as two cautious sides or a heavy fixture load, otherwise the expected value gap works against you across a season.
Q: Why do my exact-score picks keep missing?
A: Exact scores are low-probability events, so frequent misses are normal. Even the single most likely scoreline in a lopsided match tends to land only around 12% of the time. If you are hitting the right result but missing the score, you are probably doing fine. Track results and exact scores separately over at least 30 games before changing your approach.
Q: What do I need to track my predictions properly?
A: You need a spreadsheet with five columns: fixture, your scoreline, your estimated result probability, the actual score and the points earned. Add a running total and a separate column for expected points. After about 38 matches, compare actual points with expected points. A large gap in either direction tells you whether you are lucky, unlucky or miscalibrated.
Q: When are this weekend's Arsenal and Aston Villa matches?
A: Per footballpredictions.com, Arsenal play Leeds United on 10 October 2026 at 19:30, and Aston Villa play Brentford on the same date at 22:00. Their listed predictions are 2-0 and 1-1 respectively. Kickoff times can change and display in the site's own time zone, so confirm with the club or league fixture list before you plan around them.
Gambling carries risk and is for adults only. Treat predictions as entertainment, never stake more than you can afford to lose, and seek help from your local support service if betting stops being fun.