Cape Verde’s World Cup fairytale has become one of the competition’s most celebrated storylines, especially for the Paddy Power customer who backed the rank outsiders at 1000/1 pre-tournament. Inside Flutter, the archipelagic nation’s rise shone the spotlight on a key operational challenge: pricing a 48‑team World Cup where many squads arrived with almost no reliable data.
The expansion from 32 to 48 teams pushed the firm’s trading operation into uncharted territory, explains Jason Murphy, principal football trader at Flutter UK and Ireland (UKI). With the likes of Uzbekistan, Curaçao and Haiti all making it to the finals, and their respective 26-man squads including the opposite of household names, it meant work had to be done.
“The additional 16 were really low-profile teams,” he says. “We had over 1,200 players to price, with up to 15 different inputs for every player to power our pricing and player markets.”
Flutter responded by mobilising more than 80 colleagues across its global trading network, assigning every World Cup nation to a dedicated trader across the group’s global portfolio of brands, from Paddy Power to FanDuel and Georgia’s Adjarabet to Sportsbet in Australia. This effort led to the creation of a central database to help power pricing across the world, which ultimately resulted in 50,000 player selections being priced ahead of kick-off.
Everything there is to know
Karen Lockwood, football trading director at Flutter UKI, says the model relied on “leaning on expertise from all of the trading network across Flutter globally”, with each trader responsible for knowing “everything there is to know about that team and its key players”.
The project began with Flutter’s own mock World Cup draw, complete with decorations and a tongue‑in‑cheek recreation of Rod Stewart’s chaotic draw for fifth round of the Scottish Cup in 2017.
Some traders were allocated teams at random; others volunteered based on heritage or cultural ties, giving Flutter access to local knowledge that would otherwise be difficult to source.
Murphy says the scale of the challenge made the previous approach to pricing up tournaments impossible. “Five or six experienced traders covering those teams just wouldn’t have been feasible,” he explains. “We needed more hands. We needed more eyes.”
Scouting network
Each player required up to 15 different inputs to underpin pricing models, with traders researching likely starting XIs, formations, set‑piece takers, club form and tactical roles. Scouring for any available statistical or video analysis to bolster the offering was also undertaken. Lockwood says the process was akin to a “scouting network”. A tough task up against limited match footage and “almost non-existent historical data”.
There were, to be expected, gaps. Cape Verde’s final warm-up game before the World Cup was a 3-0 win over Bermuda. Highlights are hard to come by, and Google’s overview of the game doesn’t even include the goalscorers. Of course, one can dig slightly deeper, but the point stands. Sometimes, the model might have an aperture.
“We always acknowledge we don’t know what we don’t know,” Murphy admits. “People made every effort to find as much data as possible, but sometimes it just wasn’t there and, in those incidences, our experienced football traders with specialised skillsets still manage to provide prices for customers.”
Jason Murphy, Flutter
However, the result was a data bank comprising detailed profiles of each team that provided a reference point for Flutter’s Dublin-based trading team, which then finalised prices. The trading desk in the Irish capital were described by Flutter as “experienced football traders with specialised skillsets”.
Pretty sparse
For FanDuel senior sports trader Barney Spooner, who was tasked with pricing Cape Verde markets, the information gap was stark. “There was very little footage of their matches,” he notes. “Even the friendlies before the World Cup only had a couple of minutes of FIFA highlights. When you go back to African World Cup qualifying there was almost no footage, and even the stats themselves were pretty sparse.”
Compare that to, for example, the England national team and wider UEFA qualifying, where Spooner points out “you can see every pass, every shot, every foul”, the equivalent in Africa doesn’t have the same “depth of information”.
To this end, Spooner relied on the limited data, some from 18 months ago, and built player pricing “from scratch”. And when Cape Verde’s first World Cup lineup was announced, nine of his projected XI matched. “That was a pretty rewarding moment,” he says.
Cape Verde’s compact, defensive style has since reinforced those early assessments. Murphy says the team “were better than just seeing the name on paper,” feeding directly into Flutter’s ratings.
Lockwood argues the broader benefit of the exercise is customer-facing, as deeper research allows Flutter to offer “a much broader range of markets with a high degree of pricing confidence”.
“Competitors may offer these markets, but they may not have the same level of pricing confidence or the same depth,” she continues.
Cape Verde take on one of the tournament favourites Argentina in a last 32 clash on Friday, 3 July. Spooner’s prediction? “They’ll try to keep it tight, take it to extra time and maybe penalties. A player like Messi unlocks those low blocks.”
The post “We needed more hands. We needed more eyes” – how Flutter priced up the biggest World Cup ever first appeared on EGR Intel.
Ahead of 1000/1-shots Cape Verde’s knockout clash with Argentina on Friday, senior traders at Flutter UKI reveal how the operator took on the challenge of pricing a 48-team tournament
The post “We needed more hands. We needed more eyes” – how Flutter priced up the biggest World Cup ever first appeared on EGR Intel.