I check whether a signing is likely to hold up in your league, and I tell you where the data can't see. Everything I'd build for you, I've already built in public.
Editors and brands: analysis pieces and graphics.
One player, six questions, one verdict. This is what you get, in the order you get it.
Position labels lie. I group players by where and how they act on the ball, so he's compared with players who do his job, not with everyone who wears his number.
His profile against every regular in his role and league, from chances and shooting to link play, carrying and defending. Then my finishing and passing models, which separate skill from luck, and how he's trended over his recent seasons. Where he stands out, and where he doesn't.
Every league move has a track record. For strikers leaving the Championship, 45% of 240 held their level in their first season and 35% never won a regular place. That's the base rate: a prior, not a verdict on the player.
Event data can't see runs in behind or how hard a striker presses. The report says plainly where the numbers stop, so you know what they leave open.
A fair price with a range, learned from what clubs actually paid for players like him. Price any player yourself in the price app.
One word up front, a one-page answer behind it, then the detail if you want it, with the limits stated.
Ask for a Second opinion, €200One defined question, a fixed price, the method handed over with the answer. Each service is something I've already done in public: read the free version before you pay for any of it.
A short report on one player: the role he actually plays, how good he is in it (his profile against his role in his league and in yours, finishing and passing models, his trend), the base rate for his move, his fair price, what the data can't see, and a sign / monitor / pass verdict.
Built on: which leagues produce successful transfers and the price model (try it).
First, what your squad is actually short of, measured against teams that play like you. Then a target profile, candidates filtered and ranked for your league, fees and contracts checked. A one-page answer up front.
Built on: what Juventus are short of and finding their missing midfielder, the whole process done in public.
An agreed number of reports or shortlist updates each month, scoped together. Offered after a first project, so we both know how the other works.
My models cover 18 leagues: the English top four, Serie A, La Liga, Bundesliga and 2. Bundesliga, Ligue 1, Eredivisie, Belgium, Portugal, Turkey, Russia, Scotland, MLS, Brazil and Argentina.
On paid work I run the same pipeline on the data you already license (Wyscout, StatsBomb or Opta). That covers any league, and academy football too.
I graded about 72 of my own transfer calls in public. More right than wrong, and the misses are in there too.
Jonathan David, whom I misprofiled. Højlund, Modrić, Grønbæk. I've never worked inside a club, so this is how you can check my judgement before hiring it.
Read “Was I Right?”A model that prices players from real transfer fees, with a range, and shows how often the range was right (about 80% of fees land inside an 80% range). Free to use.
Open the price app13,765 transfers judged against players who stayed, so a "success" means more than a good season. Version two retracted one of my own earlier claims.
Read the pieceI'd like to work on recruitment models and the visual side of how clubs read them. In October 2026 I rebuilt my whole event-data pipeline after finding bugs in the coordinate frame, the match clock and the VAEP formula, and re-ran everything on top of it. My transfer-corridor piece retracted one of its own March claims. That's how I'd work on your data too.
Six seasons of moves against an intra-league control group: which leagues' players hold their level, and why most failures are about minutes.
ReadWorth learned from market values, then calibrated on 3,593 real fees: selling league, age, window, football inflation. Backtested window by window.
Use the appRoles from where and how players act, not from team sheets: the role behind every comparison I make.
Which teams play like yours, from possession and pressing maps, so needs are measured against real peers.
Team strength from my own xG model, comparable across leagues: how I translate a player's numbers from one league to another.
What a defender's zone concedes, from convex hulls and VAEP. Event data struggles with defending, and the method says where.
Two proxies for beating a man from a standing start. The simple one was the more repeatable.
ReadAn acceleration proxy from event data, compared against Gradient's tracking data: the La Liga outliers largely match.
ReadA Bayesian model of how much better than expected a player passes, with shooters and keepers in the same framework.
Peak velocity, and why one number isn't enough; off-ball runs three seconds before the pass. Both on Gradient's free World Cup 2022 data.
Stack: Python, pandas, scikit-learn, Streamlit, VAEP and xG modelling, data visualisation. Hudl StatsBomb certifications in recruitment and video analysis. Italian and English.
My experience on LinkedInOne person who can find the story in the data, write it and draw it, in Italian or English.
Commissioned analysis piece, from €150. Graphics, €25 each.
Ask for a quote
Most clubs buy around transfer windows. In between, The Cutback publishes recruitment work in full: methods, models, and the calls I got wrong.
Get the next piece in your inboxI'm Davide, a football data analyst in Turin. Data visualisation is what I do best; recruitment is where I'm pointing it. I work freelance with Math&Sport on event and aggregated data, and I have an MA in Cultural Anthropology from the University of Turin, which still shapes how I read collective behaviour on the pitch.
I've never worked inside a club, so I publish everything. I'm a Juventus fan and I say so. I use AI to help me build and edit, and I'm open about that too.
If you're not sure which one you need, send me the problem in two lines. I'll tell you honestly whether I can help, and if I can't.
Sending opens your email app with your answers filled in. I'll reply personally, usually with a couple of questions before we talk.