Okay, let’s talk about this soccer match ratings thing I’ve been fiddling with.
Getting Started – The Why
So, I was watching a game the other week, right? And afterwards, I checked out the player ratings online, like you do. And honestly, some of them just felt… off. Like, did we watch the same match? This one guy gets a 7, but he was invisible! Another gets a 5 but scored the winner? Made no sense to me. I figured, maybe I could track this stuff myself, see if I could come up with something that felt a bit more real, you know? Just for my own satisfaction, mostly.
First Steps and Finding Stuff
First thing I did was just grab a notepad during the next match. Seemed simple enough. I tried jotting down stuff – passes completed, tackles, shots. But man, trying to watch the game properly and count every little thing? It’s basically impossible unless you’ve got like, ten pairs of eyes or you’re just pausing and rewinding constantly, which kills the fun.
So, I thought, okay, maybe I can find some data online after the game? There are tons of stats sites. But here’s the kicker: getting raw, detailed data isn’t always straightforward. Some sites want you to pay, others have it all summarized in ways I didn’t want. It reminded me of this old project I worked on, totally unrelated, where we needed specific user behaviour data. We had tons of general info, but the exact stuff we needed was buried under layers of aggregation or just wasn’t tracked. Same feeling here. Lots of noise, hard to find the signal.
Trying to Make Sense of It
Eventually, I found a couple of places that offered decent post-match stats, like detailed passing numbers, defensive actions, that sort of thing. Okay, progress. Now, how do you turn that into a single rating, like a number out of 10?
This turned out to be the real headache. What’s more important?
- A defender making 10 clearances?
- A midfielder completing 95% of their passes?
- A striker having lots of shots, even if none went in?
- What about things stats don’t easily capture, like positioning or creating space?
I started messing around with weighting different stats. Maybe tackles are worth X points, key passes Y points, goals Z points. I built a little spreadsheet to try and calculate scores automatically based on the stats I could pull.
But here’s the thing: it started feeling really mechanical. A player could tick all the statistical boxes but still have had a stinker of a game in terms of actual impact. Or vice versa! It’s like trying to rate a meal just by listing the ingredients – you miss the actual taste, the skill of the chef.
Where I’m At Now
So, after spending a fair bit of time on this, downloading stats, tweaking formulas in my spreadsheet, I’ve got… well, I’ve got a system that spits out numbers. Are they better than the ones I complained about online? Sometimes, yeah, they feel a bit closer to what I saw. Other times, they’re just as weird, maybe even weirder because they’re based purely on my weird logic.
Honestly, it’s made me appreciate how tough it is to do properly. There’s so much subjectivity in football. My system doesn’t really account for playing under pressure, or making a crucial interception that doesn’t show up as a ‘tackle’ or ‘clearance’, or just generally having that ‘presence’ on the pitch.
It was an interesting exercise, definitely. Got me thinking more about what actually makes a player ‘good’ in a match. But am I going to keep refining this forever? Probably not. It takes a lot of effort after each game to pull the data and run the numbers. And at the end of the day, it’s still just one guy’s opinion, crunched through a spreadsheet. Maybe the professional rating systems, flawed as they sometimes seem, have teams of people and better data for a reason. Or maybe they’re just guessing too. Who knows?
