Transfer deal optimisation

Using our scientific approach, where we do not assign any weights or importance to individual variables but rather let the mathematics behind our Machine Learning model decide, we calculate the most accurate transfer values available in the market.

The model uses more than 200 variables, covering player performance, contracts, age, social media statistics, and club financials. By subsequently utilising real-time data and feeding in the required variables for current players, the real-time transfer value of any player in the top-80 leagues is calculated.

Our databases are constantly updated and the model is retrained after every significant number of transfers to make sure our insights stay up-to-date with recent developments in the market.

We generate objective assessments of players' relative strengths and weaknesses by leveraging an extensive dataset comprising over 40,000 players active across 80+ leagues globally. This enables us to compare and rank players using more than 50,000,000 data points. Our player profiles are meticulously crafted by synthesising data from diverse sources to ensure the accuracy, validity, and predictive capability of our analyses.

For each position, we've developed distinct profile types that provide insights into player archetypes. These profiles incorporate 8 to 12 composite variables, allowing for comparisons of player performance across the top 80 leagues worldwide. Additionally, we can forecast a player's profile in a different competition by leveraging historical transfer data. Through cohort analysis, we predict how a player would fare in another competition, considering factors such as the specific dynamics and playstyles within leagues and teams.

One of the most frequently asked questions when considering a player acquisition is whether the player will seamlessly integrate with your team and if their playing style aligns with the vision of your current coach. To address this potential concern, we provide insights into how the prospective player is anticipated to perform within your club's setup.

Drawing from historical transfer data, we conduct cohort analysis on players. By projecting players into teams and lineup environments, we can identify and predict how players with similar playstyles have fared at specific clubs or competitions under particular coaches.

We aim to support the transition in player selling strategies, as there's a noticeable shift from reactive selling to proactive engagement. While clubs previously waited for bids or offers on their players, the current trend among Technical Directors is to take a more active approach, pitching their players to potential destinations to optimise transfer outcomes for all parties involved.

Our model plays a pivotal role in this process by identifying the favourable context for the player you wish to sell in terms of formation, playstyle and teammates, resulting in a list of suitable destination clubs.

Subsequently, the highest potential destinations are identified by taking into account (1) the clubs' behaviour and budgets on the transfer market, (2) the quality, contract duration, age, and likely sale of players on the same position, and (3) contextual variables such as current team performance that might increase or decrease the likelihood of a player purchase. This allows Technical Directors to actively pitch their players to the most promising destination clubs to optimise their financial returns.