The rebuild season is underway in the PSL. AmaZulu, Chiefs, Marumo Gallants and Chippa United have all recently released multiple players, and recruitment teams now face the real work: finding the right replacements, within budget, that fit the club’s long-term plan. But how can data help? At Laduma Analytics, we work with clubs, scouts and analysts to turn information into better decisions - from player reports and opposition analysis to custom tools built for each team’s needs.
Introducing our Player Similarity Model
Among the many visuals and tools available on our platform is the Player Similarity Model - a valuable starting point in the scouting process. The idea is simple: select a player, and the algorithm generates a ranked list of players with statistically similar profiles, scored from 0 to 100 (with 100 representing an exact match, which rarely occurs). For example:
You admire a well-known player like Themba Zwane, but his age raises long-term planning considerations. The model can identify similar profiles as potential successors.
You’ve identified a transfer target, but negotiations stall due to contract length or cost - the model can suggest alternatives in the league with comparable profiles.
Or, one of your current players is ageing, and you want to identify a suitable long-term replacement with a similar playing style.
By narrowing down potential candidates, the Player Similarity Model provides an effective starting point for deeper evaluation, providing data insights that can be blended with traditional scouting.
The Process
Once a position profile is defined, different filters can be applied to refine the search and target specific needs. That could mean setting a minimum minutes threshold (e.g. at least 1000 minutes played in that position), narrowing by age (for example, players under 25), or comparing performance across multiple seasons.
The output creates an effective shortlist - a starting point for further scouting and analysis. Our data tools help clubs both in shortlisting players who fit a desired profile and in adding greater depth and context when evaluating those players during the recruitment process.
For example, a club might search for young midfielders who:
Can progress play through tight central zones
Can create chances from open play
Can break lines with their passing
The weighting of each metric can also be adjusted depending on the profile you're trying to build. If progressive passing is the key trait for a deep-lying playmaker, or aerial dominance is critical for a centre-back, the model can be tuned to reflect those priorities.

The metrics used can be tailored to match the specific role a coach is looking to fill and/or to accommodate the team’s playing style, ensuring the shortlist reflects the qualities most important for that position.
Of course, matching statistical output doesn’t automatically mean two players share the same ability, style, or long-term potential. Clubs will still assess the full picture - technical, tactical, physical and psychological - combining both traditional scouting and data-driven insights to build a more complete evaluation.
You may not act on any of the recommendations, and that’s still part of the process. The difference is that here, it’s informed by data. We know recruitment teams are busy, especially at this stage of the season. Data won’t replace traditional methods, but it helps focus limited time on profiles worth deeper attention and provides a valuable head start before live evaluation.
Let’s look at an example: Makhehlene Makhaula
The Squad Analysis section on our demo platform enables a club to quickly identify which players have played the most in a season, relative to their age. For Orlando Pirates, two players immediately stand out as being heavily relied upon, but also older: Deon Hotto and Makhehlene Makhaula. This visual can be generated in seconds on our platform.

Makhaula was one of the best performers on the pitch for Pirates, but they will have started succession planning for life after the 35-year-old. If Pirates were in the market for competition and a potential long-term replacement, where should they start their search? Based on general defensive midfielder metrics we have chosen (a club can choose its own), our Player Similarity tool points to Moses Mthembu, Sihle Nduli and Samuel Darpoh as the top three players with a similar statistical profile.

Sihle Nduli is reportedly on his way to Orlando Pirates. How that decision was made is open to interpretation - a mix of traditional scouting, data analysis, and internal discussions. But for a club like Pirates, who have increasingly embraced data-driven processes in recent seasons, and with whom we’ve recently engaged as part of their end-of-season review, it’s unlikely to be a coincidence. Identifying the right profiles through both scouting and data has become a key part of their recruitment strategy.
Paid subscribers can see similar replacements for Gaston Sirino and Grant Kekana, or request up to five players of interest for free.*
Gaston Sirino

Gomolemo Grant Kekana

*credit to StatsBomb for the methodology that helped inspire the approach behind this model.
