
Jesse Davis
@jessejdavis1
Followers
2K
Following
2K
Media
17
Statuses
631
Professor @KU_Leuven researching AI, machine learning, data mining and sports. Co-founder of @Activ84Health & @runeasi_ai
Leuven, Belgium
Joined August 2019
Using xG overperformance to measure finishing skill rests on the claim that xG models represent an average player. We show that biases in the data skew what average means. This bias has a surprisingly large effect & skews our understanding. @p_robberechts.
dtai.cs.kuleuven.be
Expected Goals (xG) has emerged as a popular tool for evaluating finishing skill in soccer analytics. Intuitively, the idea is dead simple.…
13
82
536
I submitted a paper to @arxiv that’s been on hold for > 3 months. This seems atypical. Any suggestions?.
1
0
0
RT @l_cascioli: 🚨The third and final blog post in our series on possession value models design decisions🔍: Can the features chosen to repre….
dtai.cs.kuleuven.be
Conceptually, possession value approaches such as VAEP, PV, OBV, and g+ are all identical: they estimate the chances of scoring (and…
0
6
0
RT @l_cascioli: Part 2 in our series on possession value models design decisions🔍: How the definition of "near future" has interesting effe….
dtai.cs.kuleuven.be
Conceptually, possession value approaches such as VAEP, PV, OBV, and g+ are all identical: they estimate the chances of scoring (and…
0
9
0
RT @l_cascioli: Has soccer gone too far in its obsession with keeping possession?⚽️. A few weeks ago I presented our latest research paper….
0
4
0
RT @Vinc_Guyot: Testing Day in Saint-Moritz 🇨🇭📊. HR🫀| Lactate🩸| RPE 💭.⚖️ : @runeasi_ai .👟 : @strydrunning
0
5
0
RT @Bapt_hnr: 👨💻 Internship - Recruitment Data Scientist ⚽ .RSC Anderlecht is looking for a student who combines knowledge of programming….
0
1
0
RT @andrepugni: The joint work with @LorenzoPerini95 @jessejdavis1 and @ruggieris is finally out in the @DMLRJournal. A huge thanks to my….
0
2
0
RT @azimmerm_dm: Doing something new this year for MLSA @ #ECMLPKDD24: panel (instead of the usual invited talk) with people doing sports a….
0
5
0
RT @andrepugni: Hello there, if you are around in Vilnius for @ECMLPKDD, today I am presenting A causal framework for evaluating deferring….
0
3
0
Steven has been a wonderful collaborator! I’d highly recommend applying!.
I am looking for a postdoctoral research associate to work on neurosymbolic methods for story understanding, as part of an EPSRC project on “Reasoning about Structured Story Representations”. Details: @Cardiff_NLP @cardiff_krr.#NLProc #ML #neurosymbolic.
0
0
1
RT @StevenSchockae2: I am looking for a postdoctoral research associate to work on neurosymbolic methods for story understanding, as part o….
0
15
0
RT @GregorydSam: ASIS registration is now open! . Look forward to details around speakers and a research competition in the near future 👀….
0
4
0
RT @hugoriosneto: Happy to announce that the paper @bruno_martins29, @resaReis, @joaollg, @CoolHandJakeGS and I submitted to the Machine Le….
0
3
0
RT @LiuPatla: The proceedings for the research track of LINHAC 2024 are now available at The book with additional….
0
1
0
RT @JanVanHaaren: I'm excited to announce that I've been promoted to Head of Football Insights & Innovation at @ClubBrugge! I'll be driving….
0
7
0
Each Group E team #EURo2024 earned 4 points. What if we used chance quality to estimate a team’s chance of winning, losing or drawing a match? This allows building an alternative table by deriving the number of points a team is expected to earn from each match @p_robberechts
0
1
3
Based on chance quality, here is the distribution over possible scorelines from last night's Belgium-Ukraine match #EURo2024
0
0
6