Every provider of expected points I've seen, the models are basically like "lol what the fuck Legia why are you so shit" π
15.12.2025 11:13 β π 2 π 0 π¬ 1 π 0Every provider of expected points I've seen, the models are basically like "lol what the fuck Legia why are you so shit" π
15.12.2025 11:13 β π 2 π 0 π¬ 1 π 0This is gonna be a huge mess
25.11.2025 15:54 β π 2 π 0 π¬ 0 π 0
However I think I've shown more than enough to back up how good KFUM Oslo have been out of possession this season. Nothing too fancy, but working hard as a team and committing to calculated risks with the support of teammates
Thanks again to Twelve for the incredible data! π
And from the same game, 4 minutes later. Mentality is crucial in football, because when something goes wrong you risk others going wrong
I talked about KFUM players trusting teammates. This can happen when as the GK you don't trust your teammate to win the ball
Of course, over the course of a season there are bound to be issues. Even against the worst Eliteserien team 2 weeks ago
This clip shows the problems that can happen when a defender is a straggler, or another over-commits, and also the risks of big gaps for opponents to run into
This final clip highlights their drive to cut out balls played in behind. As we've seen, KFUM show high risk tolerance & willingness to leave gaps in their back line if it allows them to cut out the most likely danger... but a moment of individual brilliance could unlock them!
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0
These next 2 clips show KFUM's ability to cut out passes in behind that do come. They deter them, ofc, but also focus on making sure their positioning is solid enough to pounce when these passes are played
This clip also shows KFUM's ability to keep their line high & disciplined
This clip against Brann illustrates the back line's ability to communicate & recognize when to switch. This isn't a novel concept in the slightest bit, of course, but highlights how KFUM are able to use their individuals to create a strong whole. Highlight of a switched-on OOP
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0
When players overcommit without support behind them, major gaps/spaces/disconnections occur & can be easily exploited
Conversely, when players aren't feeling supported they may undercommit to their actions (caught in 2 minds) which can have the same problems
KFUM work hard here
Basically, we see commitment without unnecessary *over*commitment from players off the ball. There's a very solid base behind players that make "aggressive" decisions, allowing them to comfortably make those decisions and commit to them
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0We then see that same drive to commit to actions & risk having gaps to put out fires with 3 players trying to win the ball back. But 1 KFUM defender keeps calm and guards the 2 opponents out wide & the space in behind evenly, trusting his teammates to get their job done
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0As I said above, KFUM seem to trust their teammates to do their jobs. They also trust their teammates (and themselves) to commit to their actions. Here we see #42 pressing from midfield. He commits. He's beaten. But he trusts #8 behind him to do his job & push the receiver wide
17.11.2025 15:35 β π 1 π 0 π¬ 1 π 0Of course, the tradeoff that KFUM are taking is that a well-placed pass could easily break them in these situations. KFUM do what they can with marking/positioning to deter these (and with GK able to come out if passes played), but they take on some risk with these decisions ofc
17.11.2025 15:35 β π 1 π 0 π¬ 1 π 0At the end of the clip, we see their line at the top of the box, compactness once more, and the back line's calmness to not get too sucked into an obvious run in behind. Tracking that run could open more space for the player on the ball or keep the runner onside
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0This shows their line at the top of the box, compactness, & limited run-following; but also something more risky. KFUM are willing to leave gaps in back line to put out fires. This requires teamwork, communication, and work rate from the others around where the ball ends up
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0
This clip vs Brann showcases their tendency to keep defensive line near the edge of the box. They don't want to drop back too early, or get sucked into 1 defender following a runner in behind & disjoining back line
The midfield & forward lines like to be quite compact as well
So... how might KFUM achieve this? My keys before we get to videos:
1) Keep defensive line from dropping back too much, too quickly
2) Commit to actions
3) Trust teammates to do their jobs, support them, & to commit in turn
4) Communication to turn individuals into a whole
This naturally translates to their very low non-penalty goals allowed, ranking 4th with just over 1 per game
Sandefjord are a fun one too, they've conceded a league-topping 11 penalties... only concede about 1 goal per game excluding them, about 0.4 per game
It's all quite impressive. Despite having average possession unlike BodΓΈ/Glimt or Brann (>60%) or Viking (league leaders, almost 60% field tilt)... KFUM are almost an outlier for how few high opportunity shots they allow per game
Just 1.7, while BodΓΈ & Viking are 2nd with 2.3
These graphs illustrate that final point perfectly
57.7% of opponent possessions reach KFUM's third, ranking 11th
But just 50.2% of possessions that reach third third go on to enter KFUM's box, the best in the league
Tradeoffs are key for them. They have a plan & execute well
Deter:
- limit opp. ability to enter box after entering final third
- limit opp. xT & box touches
Defend:
- limit shots allowed after opp. enters the box
- limit high opportunity shots (>0.15 xG)
While KFUM allow opponents to enter the final third, they limit progress afterward
However, as we saw in the first image, KFUM rank #1 in "opposition chance creation"
This is the ability to both DETER and DEFEND AGAINST opponent attacks
That shows in their style of play too
Very average defensive style, mix of high pressing and low blocks. And it can vary in a single OOP phase too. Might sit low, then press on a trigger, then get beaten and sit back to defend their box once more
Same with field tilt... KFUM are very average here too. 50.4% possession and 50.3% field tilt. They are actually a fun case study as well, since they defend about half the game, and also aren't suffocating opponents in the final third or being suffocated themselves
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0To set the stage, it's important to know that KFUM have just over 50% possession. They aren't just dropping into a low block, but they aren't only defending for short stretches either
17.11.2025 15:35 β π 0 π 0 π¬ 1 π 0
KFUM Oslo π³π΄ has been one of the most effective Eliteserien teams out of possession this season
Mid table. 50% possession & field tilt. Not great at scoring... but solid at keeping opponents at bay; they rank 5th for goals against
π & πΊ thread
Data via Twelve Football π
β½οΈ You love football and have a fascination for data π?
Then footballytics Learn is made for you.
Discover 30+ audios and videos on #analytics, #tactics and #innovation β turning experience and insights into short, practical lessons you can apply right away.
π www.footballytics.ch/learn
Realistically having more than a single game of tracking data is unheard of, some providers might have a single game of masked tracking data as samples. This is next-level and nothing I've ever seen before. Amazing to have this for the hackathon!!
14.10.2025 15:39 β π 5 π 0 π¬ 0 π 0Yeah, I was in IA when I researched this (unlimited ability to self-distribute), lived in OH right before (almost unlimited), then moved to VA. And it was wild to see the worse selection in VA, mainly larger brands or "craft" breweries owned by AB InBev or something lol, just a few small local ones
06.10.2025 18:17 β π 1 π 0 π¬ 0 π 0I read every state's laws in 2019, so it's very possible more states have changed their laws/removed 3-tier laws. When I collected this data, no states that had repealed the 3-tier system had then re-enacted it, I'd assume that's still the case
06.10.2025 18:00 β π 1 π 0 π¬ 0 π 0