Advertisement · 728 × 90

Posts by Mátrai Márk

Post image

PSG and Bayern had the toughest opponent fixture list based on the average Elo rating of the opponents in the tournament. English teams can't play against each other in the league phase, which helps them a lot. This mitigation needs to be done.
Source: Club Elo

2 weeks ago 0 0 0 0
Post image Post image Post image Post image

Champions League QF matchups. Liverpool are the best so far, they have the best numbers. Bayern should be considered the best by their numbers and taking into account their opponents.
Below 👇opponent quality, which impacts these numbers due to low sample size. @optaanalyst.optajoe.com

2 weeks ago 0 0 1 0

Anfield Index, especially the Under Pressure podcast with an award-winning journalist, sports scientist and a professor from a university.

2 weeks ago 2 0 1 0
Preview
Bevásárlólista 2.0 9x28

A tavalyi népszerű adás után idén is beszéltem a Liverpool lehetséges nyári átigazolásairól a PoolBarátok podcastben. / After last year's popular episode, I talked about Liverpool's potential summer transfer targets again for the PoolBarátok podcast. HU open.spotify.com/episode/3R0O...

3 weeks ago 1 0 0 0
Preview
How to expect more from xG Getting from shot probabilities to team performance metrics

Expected goals are everywhere now, but there's so much more we can do with them. Here's how futi turns basic xG into team-level metrics including deserved results, projected points and projected finish.

1 month ago 40 8 1 0
Preview
Állapotfelmérő 9x25

A legújabb PoolBarátokban beszéltem a szezon eddigi alakulásáról, illetve annak hátralévő részéről. Köszönöm a lehetőséget! / In the latest PoolBarátok podcast, I talked about the season so far and about the rest of the season. Thank you for having me! (HU) open.spotify.com/episode/53p8...

2 months ago 0 0 0 0
Preview
Tension Tension drives growth and it also fuels frustration. While tension often gets a bad rap, the right kind of tension is hugely beneficial for building new skills, expanding trust, and collective problem solving. If we believe that our best relationships in work and life generate friction, then it makes sense that tension isn’t just good, it’s also necessary for human development.

Tension

Tension drives growth and it also fuels frustration. While tension often gets a bad rap, the right kind of tension is hugely beneficial for building new skills, expanding trust, and collective problem solving. If we believe that our best relationships in work and life generate friction,…

2 months ago 1 1 0 0
Advertisement

If I could listen to only one song during my entire life, this would be it.

2 months ago 0 0 0 0

Thank you for @profootballfocus.bsky.social @pro-football-reference.com and @rbsdm.com for the stats! It is amazing that the NFL has free public data available for complete causals to play with and to improve themselves.

2 months ago 0 0 0 0

MLB voters are far ahead of NFL voters. Metrics such as WAR, or LEBRON by BBall-Index in the NBA, can eradicate these biases and account for more things than people can. The NFL needs to come up with their own WAR, so prestigious awards get to the rightful winners.

2 months ago 0 0 1 0

As we saw it with the Hall of Fame's shameful decision, voter education is at a low point in the NFL. The NFL has to educate voters, to not use old stats anymore (TD-INT ratio, common opponents, picking stats by a biased friend aka Dan Orlovsky). Biases should be fought consciously by voters.

2 months ago 0 0 1 0
Post image Post image Post image

Drake Maye should have been the NFL MVP in 2025 based on the advanced stats. There were some stats in Stafford's favour, but overall, Maye should've been the winner, even after accounting for schedule difficulty, which was everybody favourite stick to beat Maye with.

2 months ago 0 0 1 0
Preview
What Makes A Data Visualisation Elegant?, Nightingale On a dedicated channel, #dvs-topics-in-data-viz, in the Data Visualization Society Slack, our members discuss questions and issues pertinent to the field...

What makes a data visualization elegant? 📊

6 criteria by @visualisingdata.com

3 months ago 27 5 1 1
Maye's schedule has to play a starring role in the debate if you're arguing for Stafford. And while I lean toward Maye, there's no denying that the Patriots have played a breezy slate in 2025. As I mentioned last month, per Pro-Football-Reference.com's Simple Rating System, the Patriots were in line to play one of the 10 easiest schedules in NFL history since the merger. Obviously, that's going to help make life much easier for Maye.

The context missing from the schedule argument, though, is that overall team strength isn't the same thing as the strength of the opposing pass defense. As an example, Stafford played two games against the 49ers, who are a good team on the whole but field one of the worst pass defenses in the league. And Maye went up against the Browns; Cleveland has a horrible offense, but its Myles Garrett-led defense is excellent against the pass.

Using each team's finish in pass defense EPA per play from NFL Next Gen Stats, Maye's average opponent ranked 20th against the pass this season. Stafford's average opponent was 17th. Do the same test with QBR instead of pass defense EPA, and their schedules were virtually identical; Maye's average opponent ranked 18.6 out of 32 defenses, while Stafford's average pass defense was at 18.4. While these two quarterbacks faced massively different opposition on the whole, that doesn't hold true for the pass defenses.

Maye's schedule has to play a starring role in the debate if you're arguing for Stafford. And while I lean toward Maye, there's no denying that the Patriots have played a breezy slate in 2025. As I mentioned last month, per Pro-Football-Reference.com's Simple Rating System, the Patriots were in line to play one of the 10 easiest schedules in NFL history since the merger. Obviously, that's going to help make life much easier for Maye. The context missing from the schedule argument, though, is that overall team strength isn't the same thing as the strength of the opposing pass defense. As an example, Stafford played two games against the 49ers, who are a good team on the whole but field one of the worst pass defenses in the league. And Maye went up against the Browns; Cleveland has a horrible offense, but its Myles Garrett-led defense is excellent against the pass. Using each team's finish in pass defense EPA per play from NFL Next Gen Stats, Maye's average opponent ranked 20th against the pass this season. Stafford's average opponent was 17th. Do the same test with QBR instead of pass defense EPA, and their schedules were virtually identical; Maye's average opponent ranked 18.6 out of 32 defenses, while Stafford's average pass defense was at 18.4. While these two quarterbacks faced massively different opposition on the whole, that doesn't hold true for the pass defenses.

Last week, I talked about Maye and Stafford against common opponents, but I'm going to take that one step further. Maye outperforming Stafford against the same opposition tells part of the story, but it leaves out a piece of the equation. If everyone got to play Maye's slate this season, would other quarterbacks also be producing career seasons?

Overwhelmingly, the answer to that question is no. Maye posted a 77.0 QBR against his schedule this season. Stafford posted a 57.7 QBR in six games against common opponents, losing to the Falcons and Panthers. The rest of the league posted a 58.5 QBR against those same teams on the whole throughout the season. The only quarterback who threw more than 100 passes against those teams and posted a better QBR than Maye is Dak Prescott, who was 0.8 points higher than Maye over a much smaller sample.

If Maye was really a product of his schedule, his performance would be much more in line with what average quarterbacks did against the same opponents in 2025. Instead, not only have we found that Maye's schedule of pass opponents wasn't much easier than Stafford's slate, we've also found that Maye grossly outplayed Stafford and the rest of the NFL against the same opponents.

Last week, I talked about Maye and Stafford against common opponents, but I'm going to take that one step further. Maye outperforming Stafford against the same opposition tells part of the story, but it leaves out a piece of the equation. If everyone got to play Maye's slate this season, would other quarterbacks also be producing career seasons? Overwhelmingly, the answer to that question is no. Maye posted a 77.0 QBR against his schedule this season. Stafford posted a 57.7 QBR in six games against common opponents, losing to the Falcons and Panthers. The rest of the league posted a 58.5 QBR against those same teams on the whole throughout the season. The only quarterback who threw more than 100 passes against those teams and posted a better QBR than Maye is Dak Prescott, who was 0.8 points higher than Maye over a much smaller sample. If Maye was really a product of his schedule, his performance would be much more in line with what average quarterbacks did against the same opponents in 2025. Instead, not only have we found that Maye's schedule of pass opponents wasn't much easier than Stafford's slate, we've also found that Maye grossly outplayed Stafford and the rest of the NFL against the same opponents.

New column: My awards for the 2025 season, including why Drake Maye is my pick for MVP over Matthew Stafford and the major flaw in the schedule argument: measure the *pass defenses* they faced and their quality of opposition was mostly the same www.espn.com/nfl/story/_/...

3 months ago 89 9 17 9
Preview
'Analytics history' - there is so much to write about and share The history of football analytics often focuses on expected goals and the 21st century, but decade after decade through the 20th century brought insight

There are far too many interesting football analytics papers that I want to read. From the 1980s. And from before that, if they were actually accessible.

This is the blog I've rewritten the most, and is still so short of doing the subject justice.

www.getgoalsideanalytics.com/analytics-hi...

4 months ago 17 8 4 2
Post image

Nikola Jokić has started the 2025-2026 NBA season in historic fashion.

Currently at 127%, his true shooting from all over the floor is hitting unprecedented heights. An all-time great, box office player.

His league-adjusted true shooting % from season-to-season per @basketball-reference.com

5 months ago 0 0 0 0
Post image

Liverpool's squad outlook by looking at peak ages for each position. There will be some turnover in the next 2 years (Alisson, Salah, possibly Van Dijk from the starting XI), but the age profile looks very healthy.

Exciting times for Liverpool FC.

Thank you for the help, Dobi Gergő!

6 months ago 0 0 0 0
Advertisement
Post image Post image Post image Post image

Liverpool confirms the signing of Alexander Isak on #DeadlineDay
He is great finisher, shooter, but is also a very good facilitator, which will further enhance Liverpool's attacking prowess. He can scale up massively in this team, which has the potential to be incredible.

7 months ago 2 0 0 0
Preview
Szezonfelvezető 9x05 PoolBarátok Podcast · Episode

Szezonértékelő a PoolBarátokban, valamint a jövőbe révedés. Köszönöm a lehetőséget! / Season review podcast on PoolBarátok, where we also talked about the new iteration of Liverpool. Thank you for the opportunity! (HU)
open.spotify.com/episode/2ZA2...

8 months ago 1 0 0 0
Post image

Overall, Zubimendi can be a good signing for €65M, but he still won't solve Arsenal's issues in the final third in open-play. Arsenal needs to add players for that and Arteta has to provide more freedom, let his team take more risks in order to score enough goals.

End of thread

9 months ago 0 0 0 0
Post image Post image

A) Expected Threat zonemap
B) Typical Key Passes
He creates threat on the ball from the central areas close to the centre circle. His ball retention will be key for Arsenal to keep the other team under pressure.
His presence isn't as prevalent in the final third.

9 months ago 0 1 1 0
Post image Post image

Interceptions are usually proactive, ball recoveries are more reactive parts of defending. He produces these in the opponent's half as well, which bodes well for Arsenal's pressing.

9 months ago 0 0 1 0
Post image Post image

A) Typical passes into the final third
B) Typical progressive passes

He operates from the half-spaces, delivering diagonal passes towards the attackers. Sometimes he comes over to the right side as well.

9 months ago 0 0 1 0
Post image

His pizza chart shows his qualities in attack, in-possession and in defending. He has no real weaknesses in defending. In-possession he contributes in deeper parts of the pitch. His performance vs England in the Euros final (replacing Rodri) was a key factor in Spain's triumph.

9 months ago 0 0 1 0
Post image

Martín Zubimendi has signed for Arsenal. The North London club has signed a player, who is a metronome midfielder, whose all-round game will help Mikel Arteta's side in multiple phases of the game.
Visualisations are by @xfbsays.bsky.social
Thread🧵

9 months ago 1 0 1 0
Post image

Overall, Liverpool sealed a very expensive transfer, but it is not a deviation from Fenway Sports Group's modus operandi. If the data and the scouts confirm that a player is world-class, they see the overall financial package as value. Wirtz is a hugely exciting player.

End

10 months ago 1 0 0 0
Advertisement
Post image

He is very active off-the ball as well, when his team has to defend. His two-way game will increase his added value to the team, his impact will be immense.

10 months ago 0 0 1 0
Post image

He touches and receives the ball a lot in the final third, which is also incredibly valuable. Despite his age, he is among the very best of elite players, when combining these metrics.

10 months ago 0 0 1 0
Post image

He is among the very best dribblers and through ball passers. He is close to Salah in these metrics, which is definitely something to keep an eye on.

10 months ago 0 0 1 0
Post image

Compared to other great U23 players, this is how he performed last season. Numbers are adjusted for league-quality, using VAEP method by the @comeonman on Twitter.

10 months ago 0 0 1 0