2010 was Paul the Octopus’s year. Known as the animal oracle, Paul correctly predicted the winners of 12 out of 14 games in the 2010 FIFA World Cup. Paul’s death left a void in the prediction world, and when the next World Cup occurred in 2014, a search commenced for ‘the next Paul’. Headlines surfaced of a turtle that correctly predicted Brazil would win their World Cup opening match against Croatia; equally successful was China’s physic panda Ying Mei. Flopsy, a kangaroo from Australia, was renamed Predictaroo in the hope of becoming Paul’s successor.

Predictions are a big part of the fun of sport. Everyone loves to guess a score, to preview a match, to bet on a result, and the added competition that feeds the natural competitiveness of sport fans. There is always something to predict: not only can you guess who will win, but by how much and how many times, through whose goals and whose assists. The game isn’t only played on the pitch, as countless betting commercials constantly remind us.

What about sport makes people want to keep guessing? Perhaps it’s because sport is often unpredictable and surprising. The beauty of sport is that it rarely has a certain answer, but it usually has a likely one. It draws hopeful betting individuals in, but without too much sense of security. Even after countless calculations and statistics, on the day something can always go the other way. The difficulty of getting them right makes fulfilled predictions even more satisfying.

Many predictions go horribly wrong; in the 2014 Football World Cup, all bets were on Spain to take the title. Yet they were smashed 5-1 by the Netherlands in their opening match, and the next game saw them defeated by Chile. They didn’t even make it out of the group stages, previously unthinkable as reigning European and World champions. The very same tournament saw Brazil, the home nation and favourites to win, lose 7-1 to Germany. It was the largest margin of victory ever recorded in a World Cup semi final, and for it to happen to the home nation – a team with unquestionable talent – was something else. More recently, Japan shocked the rugby world, beating South Africa 34-32. South Africa, a team that has won two out of six possible World Cup titles, a team that is currently ranked third in the world, inferior only to New Zealand and Australia. Shock results like these blow predictions out of the water.

But what if the shock results aren’t one-off occurrences? Chelsea’s Premier League fall from grace, Team GB winning the Davis Cup and Tyson Fury defeating Vladimir Klitschko. These are teams and individuals who for many matches, many rounds, have defied expectations. Should these trajectories define their performance for the following year? Perhaps they can be used to predict the next year’s results. The question is, does anyone really believe that Chelsea’s place is now forever near the bottom of the league? Unlikely. Compared with all the historical data, the current results are still odd: Chelsea are a world-class team and GB have a long way to go before they become the all-round most successful tennis nation. So when can results like this start to define the team or individual, and change our perception of them?

In this case, as in many others, the predictions are down to the journalists’ discretion. But if you think about the content, there is no way sports predictions can be a fully truthful account of what journalists think. Regardless of personal opinion, there must be shocks in there. Reporters’ words are worth nothing to the reader if they predict that something that happened last year will happen again. This isn’t putting their expertise to good use; this is copying history, so even if it will happen, it’s probably not worth putting it in the predictions. Prediction articles are not only designed to predict sporting outcomes, but they are also a key way of highlighting upcoming talent, the key sporting events of the year, and crises that have yet to be resolved, all in one. So think about the content formula before believing all the words of wisdom.

Of course, not all predictions rely on individual judgement; a growing number of journalists are turning to mathematics. All bookmakers use statistics in order to create their odds. At The Times, Henry Stott and Alex Morton have been developing The Times’ ‘Fink Tank’, using statistics to predict football match outcomes. This takes beating the bookmakers to another level. Stott claims, “wherever you look you see people that could make better decisions if they were assisted by statistics.” It has been suggested that these statistics and computer-generated predictions should be used by managers as a key source of information in deciding which team to field. But do these statistics really provide a more reliable, accurate prediction, or are there some things that number crunching can’t solve? Many managers are still reluctant to denigrate ‘the beautiful game’ as being solely defined by statistics.

Despite journalists’ idiosyncrasies, the shock results and Paul the Octopus’s complete lack of sporting knowledge, people love to read the predictions. They may be wrong but at the end of the day, this time next year no one will be looking back at last year’s predictions, checking how many were right. They will already have moved on. Their gaze will be on the future, predicting the next year.