This is an extract from a University of Portsmouth study, with key information to support our effort to establish a system to measure diving.
Before we can address the the idea of measuring the overall prevalence of diving in football, we need to find a robust framework for identifying what is and what isn’t a dive. How are we to go about this? Many people would say it’s impossible to know for sure. All diving judgements, they would argue, are subjective interpretations, and therefore inherently fallible.
Fortunately, a non-verbal behavioural study published in 2009 by Paul H. Morris and David Lewis from the University of Portsmouth, UK entitled “Tackling Diving: The Perception of Deceptive Intentions in Association Football (Soccer)” has provided us with tools with the potential to take us beyond the realm of subjectivity.
Results from the study show that science can be applied to the challenge of identifying dives. The main conclusion is that deceptive intentions in diving players can be correctly observed by spectators. Two of their findings are especially important for our effort:
The results of the study are summarised below:
The participants in the study were remarkably consistent in their judgments about the deceptive intentions or otherwise of tackled players. Furthermore, there was general agreement about tackles where it was difficult to make a clear judgement about the intent of the tackled player. Consequently, it was found that non-expert viewers can consistently and reliably sort the responses of players on the receiving end to all tackles into one of three categories: dives, non-dives and inconclusive incidents.
The validity of a judgement about diving can only be determined if the intentions of the alleged perpetrator are known. The researchers therefore devised an experiment in which this was the case. Experienced players were asked to perform ‘natural’ passages of play in which the attacker attempted to dribble the ball past the defender. Both were instructed to perform their roles ‘authentically’: the attacker was to seek to evade being tackled while the defender was asked to attempt to perform a fair tackle if they could. But crucially, if a tackle did take place, the attacker was instructed to exaggerate the effects as convincingly as possible.
This experiment showed that there was a powerful and consistent relationship between the intentions of the tackled players and the judgments of those intentions by the participants in the study. This indicated that the evaluations of tackled players’ intentions by non-expert viewers consistently have a high level of validity.
According to this study, amateurs with different levels of expertise did not differ significantly in their evaluations. However, other studies have shown that experts (world class players) are liable to place a lower percentage of alleged dives in the ‘inconclusive’ category than non-experts. To mitigate this difference in performance in our study, inconclusive incidents were simply not taken into account. Conversely, where a non-expert deems an incident to be a definite dive or non-dive, an expert is overwhelmingly likely to agree.
The study revealed a scientifically verifiable relationship between behaviour and intentions. This means that deceptive intentions can be observed in players’ behaviour. Four categories of behaviour were identified that were consistently associated with deception:
See more details about these behaviours here.
After analysing the Portsmouth study we learned that a team of non-expert viewers can be trained to identify the specific behaviours that are tied to deceptive intentions in players, and that their judgements of whether a dive has taken place or not can be consistent (different viewers would agree on the outcome) and correctly associated with the intentions of the tackled player.
And so we will embark on a journey to make it happen. Stay tuned!