I Didn’t Get 27% Worse at Padel – The Rating System Just Let Me Down
“I recently had a short run of games booked through Playtomic. Nothing unusual. Same standard of opposition, different partners, normal club padel. However, over the course of four matches, my rating dropped from 4.42 to 4.12.
On the surface, that doesn’t sound dramatic. A few tenths either way feels fairly normal, and most players would barely think twice about it. But when I actually sat down and worked through what that change meant in percentage terms, using our mapping to the PadelLevels scale, it revealed something far more significant. That movement equated to roughly a 27 percent drop in playing level. That’s the bit that should make you stop and think.
A swing of that size isn’t form. It isn’t confidence. It isn’t a slightly off day. It’s the difference between players who shouldn’t realistically be matched together. Different rally tolerance, different shot quality, different outcomes.
I didn’t suddenly become 27 percent worse as a player because I had four games with different partners.
This isn’t about calling out any one platform or saying anyone is doing things “wrong”. It’s about understanding what different rating systems optimise for, and where the trade-offs sit.
Why Context Matters More Than Just Results
This is one of the core issues with simple ELO-style rating systems, particularly in doubles. They react quickly to results, but they struggle to separate individual performance from context. Partner variance, opponent mix and short-term swings all get baked into the number.
The rating moves, but the signal gets diluted.
At PadelLevels, we’ve taken a deliberately different approach. Rather than focusing purely on wins and losses, we measure performance relative to expected outcome. If you win when you’re expected to win, with a result that you’re expected to have, your level barely moves. If you lose narrowly against stronger opposition, that’s informative. If you consistently outperform expectations, your level rises in a controlled, meaningful way.
Consistency is the real signal. Not just in padel, but in anything where progress actually matters. Good systems reward what you do over time, not what happens on a single day.
PadelLevels is built to recognise sustained performance, while also giving players the confidence to compete without fear. A bad day shouldn’t define you, and a good system shouldn’t punish you for having one.
Ratings also shouldn’t be static. If players improve, their level should move, and it should move promptly. We’re very much in favour of dynamism. The challenge is finding the balance between responsiveness and accuracy. Move too slowly and ratings lag reality. Move too fast and they stop meaning anything.
We believe PadelLevels has hit that sweet spot, with ratings updating regularly but in sensible, proportionate steps that reflect true change rather than short-term noise. The aim isn’t to slow ratings down for the sake of it. It’s to make sure they move for the right reasons.
We didn’t arrive at this approach by accident.
Better Ratings. Better Matches. Better Experience.
Before padel, we spent years building and validating the same underlying system in squash with SquashLevels. That work now spans millions of recorded matches across club, junior, national and professional levels. In junior squash in particular, we apply additional calibration and progression boosts, because development isn’t linear and young players improve in bursts rather than steady increments.
First, much higher predictive accuracy. Ratings become a reliable reflection of true playing level, not short-term noise.
Second, and just as importantly, far better match quality. When players are rated accurately, the majority of matches are genuinely competitive. Not perfect coin flips, but close enough that both players feel tested and engaged.
That second point matters more than most people realise.
You see the consequences when rating systems don’t get this right. Self-rating is a good example. I recently played a match that finished 6–0, 6–1 because a player had self-rated into a bracket that didn’t reflect their true level. Nobody benefited from that game. It wasn’t competitive, and it wasn’t enjoyable. That’s not a player problem. It’s a system problem.
To be clear, simple 1–7 scales exist for a reason. They’re familiar, accessible and give players an easy reference point. We support that as a front-facing layer because it helps players orient themselves quickly. But simplicity shouldn’t come at the expense of accuracy.
Underneath that 1–7 reference, PadelLevels runs a far more granular level scale. One that allows us to show real progression, percentage improvement and consistency across matches, clubs and regions. Think of 1–7 as a signpost, and PadelLevels as the map.
The goal of a rating system isn’t to produce a number that looks tidy. It’s to create better matches. When ratings swing wildly over a handful of games, the result isn’t insight, it’s mistrust. And mistrust leads to mismatches, frustration and players quietly disengaging.
Good ratings don’t just describe ability. They actively improve the playing experience.“
That’s why we built PadelLevels the way we did.
✍️ Jethro Binns, Co-Founder, PadelLevels.
Please drop us an email if you’d like to bring PadelLevels to your club, we can walk you through the benefits and even get you set up with free box leagues ⬇️