efStats

📊 Why Robust Z-Score Matters - A standardization method designed to reduce the impact of skewed distributions and outliers - It can be seen as a middle ground between Percentile and Z-Score 🧵Thread.

@academicismo @PER90s Exactly! New sources always bring surprises.

@ef_Stats @PER90s Lovely! you are keepin' on with new sources of data. per90 is awesome.

🏴󠁧󠁢󠁥󠁮󠁧󠁿 Callum Doyle (Wrexham / CB / 22) Excels in progression through both passing and carrying. An Overlapping Centre-Back who takes up advanced positions in possession.

Defensively intriguing too. SkillCorner data (small sample as of December) shows strong Stop or Reduce Danger % and Force Backward %. A dual-class profile — high output in possession while functioning as a Shield. scoutedftbl.com/france-ligue-1…

✅ This is the result of my data analysis on the specific play-styles teams employ in the PL this season! Plenty of interesting talking points up for discussion off this 🤔 If you are interested in my methodology for this study, check out the link in the follow-up tweet 👇

16-YEAR-OLD CAVAN SULLIVAN IS SPECIAL ✨ What a performance in Philadelphia Union’s 7-0 win in the CONCACAF Champions Cup 👏 90’ played 61 touches 2 goals 2 assists 3 chances created 4 shots 2 tackles 1 interception 9 recoveries 3 duels won You love to see it 🇺🇸

Missed the green horizontal bars? I’ve added 'scouting reports' for each player on markstats.club. Click the small icon next to a player’s name to view them. Everything is still free

📊🇯🇵 Keisuke Goto, data report + physical specimen, the kind that is trending to be prepared for the rigors of the Premier League; 6'3" (191 cm), and attempts/wins aerials at a fantastic rate for his age + high defensive work rate + used primarily (an is quite proficient) as a "f it, Keisuke down there somewhere" long ball outlet + finds insanely high quality shots despite 'limited service' in the form of shot/attacking penalty area touch volume - despite electric shot quality, his shot on target rate this season has been relatively abysmal, also worth noting he's managed just 6.71 xGOT on 10.34 xG from 40 shots this season... just far too early to make any assertions on his finishing ability long term, but something to monitor - definitely more of a poacher at this point (or at least used as one) considering shots per danger area touch, low xA and general creative output ➡️ Filter by aerial duels and npxG considering age and he'll be at the top of most lists. To be fully transparent, I haven't watched a second of him this season, so this is purely a data read... if I'm wrong about things or further context is needed, please let me know. The recent reports make sense that many mid-level PL clubs are tracking him... everyone wants power forwards for strikers, and w/ the recent data point of Igor Thiago probably over-indexed in decision maker's minds, he'll be an appealing prospect to keep on shortlists for the coming year or so.

xPts & xTD Rankings | 25/26 Championship xPts follows the quoted source, while xTD is simply "xGD that accounts for non-shot threats". Factoring in current Pts, Ipswich Town, Coventry City, and Middlesbrough look like the promotion candidates.

Adam Wharton attempts a progressive pass on 23% of his total attempts, by far the highest rate of any midfielder in the Premier League. The synthesis of the player's aggression and the manager's pursuit of verticality. No wonder Palace has easily the highest direct speed.

🚨Introducing Tactical Power Ratings (TPR) Measures any manager containing Opta-era data using two sophisticated metrics; • Shot Domination % = (Shots For / Shots For + Shots Against) * 100. Weighed based on game-states to eliminate desperate attempts or the leading team 'backing off' when maintaining a comfortable lead (100% if the score is within 2-goal deficit, 45% for =3, 15% for >=4). Benefits teams that can kill the game off early and removes stat-padding. The number is then normalized against the league average of the specific season as well as the strength of the opponent, ensuring it is NOT due to the strength of the players, but by how much the manager tactically outperforms his opponent relative to their strength. For instance, the league average against Burnley this season for Raw Shot Share is ≈ 67.8%, and in GW12 Chelsea came up against them under Maresca only managing to bolster a 65.2% raw shot share, meaning his team underperformed by -2.6% to the rest of the league (normalizing to league), and given Maresca's average career shot dominance measures at +5.3%, it means he underperformed by -7.9% in that particular match compared to what was expected. Ensuring managers get benefitted irrespective of their teams 'natural' quality. • NET Shot Quality % = (W. Avg xG/Shot - W. Avg xG/Shot Conc) * 100. This is the efficiency metric, and it is similarly weighed using the same game-state percentiles for the same reasons mentioned earlier to prevent blowouts etc, and benefits managers that can obviously generate higher quality chances, and only concede lower quality ones. So, using our example of Chelsea's 2-0 victory away at Burnley earlier in the season, Maresca's team generated an average 0.12xG per shot, and only conceded one's worth 0.05xG, meaning a +0.07 margin or 7%. Finally, the actual TPR, which you can see by the color graph. This isn't calculated using a standard 1:1 ratio of the two X/Y variables, instead by a 1:1.55 (Shot Domination: Shot Quality), as research proves efficiency is 0.23 times more predictive of PPG than volume. Net Shot Quality measuring the efficiency holds a 0.65 correlation with PPG, with Shot Domination only showing a 0.42 correlation with PPG. So, TPR = (Shot Domination + (Net Shot Quality * 1.55)) / 2.55)... Essentially the holy grail of a manager's capability. I've inserted a graph of permanent PL managers since the 20/21 season, reminding you that it samples their entire careers outside the PL, meaning not completely indicative of their actual success in the PL that has its own (dominant) tactical niche.

I'am excited to share my first web app: Mean-variance analysis of passes received Right now you can find matches available for the last weekend in Serie A and the Premier League. Any feedback, suggestions or requests → drop them in the comments! 👉🏼…s-received-meanvariance.streamlit.app

For Avdullahu, I’d also recommend reading Spencer Mossman’s analysis: x.com/fc_mossman/sta…

Profile Comparison I also created scouting reports for Sangaré and Barco. As mentioned previously, while they each possess distinct characteristics, they share the common ability to combine high pass completion rates and ball recoveries. (Source: scoutingstats.ai)

🇽🇰 Leon Avdullahu (21) A typical DLP who excels in ball retention. He prefers passing over carrying and is more influential in possession than out of possession.

Source: datascout.pro

Prog-P accuracy (%) vs Progressive passes - 25/26 T5 U-23 DM & CM - 900+ mins played Leon Avdullahu (21, Hoffenheim) is one of the few players who combines a high volume of progressive passes with a high completion rate.

Been working on a new framework to measure how two-footed a player really is called the Ambipedal Contribution Index. My first long-format article explains the ideas and logic behind it. Let me know your thoughts and guesses for who ranks at the top. @sahilgdwn/quantifying-ambipedal-ability-in-modern-football-3f2852fbfcb2" target="_blank" rel="nofollow noopener">medium.com/@sahilgdwn/qua…