https://soccerment.com/expected-threat/ Expected threat (xT) is essentially a machine-learning model that uses ball-location data to measure how much a player’s actions increase (or decrease) the chances of a goal being scored by the team in the next 5 actions. Thus, it is, in a sense, measuring how much a player’s actions progress the ball into more dangerous locations. So it is a complex measure of a player’s value in ball progression and playmaking. Not surprisingly, by this measure, Messi is elite both as a passer and a dribbler: The chart shows Messi increasing attacks’ threat with passing more than anyone else and increasing attacks’ threat with dribbling more than anyone, in the 2017-2018 season. There’s also data in the link on the 2020-2021 season, where Messi has the highest combined expected threat total per 90 minutes, but it looks from the charts like there may have been 2 or 3 people that season with higher xT from passing specifically (but Messi overcome them in total due to having higher expected threat from dribbling). This once again shows Messi being an absolutely elite playmaker, as it is squarely a playmaker’s job description to move the ball into more threatening positions, and Messi is clearly the best in the world at it. _____________________________ Relatedly, that same link includes the following chart as well, which relates to a similar measure of how much a player’s actions (including passes, dribbling, as well as losing the ball) adds (or subtracts) to the danger of attacks: Specifically, on the x-axis, this chart shows Expected Offensive Value Added (xOVA) for the 2017-2018 season. Expected Offensive Value Added is a machine-learning-driven model that basically takes any pass or shot a player makes and measures the expected goal value of the shots and the expected assist value of the passes (i.e. measures the chance that a pass to a player in that exact location/scenario would end up being an assist), and then it subtracts the expected assist value that existed each time the player received the ball. This therefore is a slightly different way of measuring the amount that the player increased (or decreased) the team’s chances of scoring from where it was when they received the ball. Since it uses xG and not goals when a player ultimately shoots, this measure is unaffected by the player’s actual finishing if they take a shot, but instead just measures the value added by the offensive actions made by the player leading up to that shot. It is therefore essentially a measure of playmaking (including for ones’ self). In any event, the individual data points on the chart aren’t labeled with their names, but we can tell that the upper-right-hand corner one is Messi because the chart maps xT on the y-axis and we know from the prior chart that Messi is the one with just above 0.35 in combined passing and dribbling xT. Thus, we can tell that Messi was #1 in this measure as well in that season, since the data point that is him is also the right-most point on the chart (i.e. it has the highest xOVA per 90 mins too). As such, Messi is the best playmaker by another complex measure.