In a recent Reddit post, redditor TapeTen presents evidence allegedly rejecting artificial momentum swings at 2-0. Although the post is based on a fairly small set of his own matches, we thought it deserved some attention.
TapeTen recorded data on matches where either he or his opponent were 2-0 up. Below, we see the entire sample consisting of 10 matches. What you should notice below is that the team leading by 2-0 managed to win in all cases but one.
To cater for the obvious fact that the sample above is fairly small, TapeTen continued recording data. At 21 matches where either player had been leading 2-0, TapeTen still hadn’t recorded any other examples of “momentum shifts” than match #1 above.
Some fair objections
What is absolutely clear here is that TapeTen found very few cases where a 2-0 lead clearly didn’t lead to any sort of momentum shift. However, the applied method has some shortcomings.
If we ignore the obvious sample size problem, TapeTen’s data rejects any notion that the game has a build in mechanic to reverse 2-0 leads. One could however argue that this argument is a bit of a straw man. After all, most momentum proponents actually don’t believe that all 2-0 leads are overthrown through EA’s divine intervention. A far more widespread idea is that the game has a “bias” towards the losing team at 2-0.
TapeTen’s results probably won’t convince those who believe in the invocation of a “losing team bias”. A “losing team bias” might, if it exists, make matches more even. But it won’t necessarily overthrow all 2-0 leads. And we can’t rule out that all matches in TapeTen’s sample theoretically would be matches which accidentally were extremely unbalanced from the start. In that event, the alleged “trailing team bias” perhaps just wasn’t powerful enough to overthrow any matches.
Hence, the results presented here don’t rule out the option that a trailing team bias could exist.
Evidence that stands
Yet, the fact remains that TapeTen only recorded one single 2-0 lead being overthrown in 21 matches (5 %). If we ignore the statistical uncertain, an obvious question to ask would be what that percentage should look like in order to convince momentum believers that they are wrong. It is a given that some 2-0 wins should be overthrown under natural circumstances, but exactly how common is that phenomenon?
The first place to look for comparison would be real football where EA has no influence other than offering sponsorships.
As most football fans know, many real footballers, coaches and pundits subscribe to the claim that “a 2-0 lead is the most dangerous lead”. So, according to folklore, 2-0 comebacks are pretty common in real football!
However, the statistics tell a different story.
SKY sports fact checked the claim that 2-0 is a dangerous lead. They found that in 90 % of the 2766 cases in their sample, the leading team would hold on to the lead. Only in 2 % of the cases, a 2-0 lead ended up as a win for the trailing team. As one might have expected, the bigger the lead, the lower the risk of losing the grip.
Adding more data to the party
TapeTen’s data set is too small for any meaningful comparison, but we are in luck here: Our old acquaintance Arlington69 made a much larger data set with a similar content. Using Arlington69’s data, we analyzed 274 matches where either player was 2-0 up at some point. We found that the team that was 2-0 up, ended up winning in 79 % of the cases and losing in 11 % of the cases.
The result confirms that it is more likely to win than not when you lead by 2-0. But it also reveals that it is more difficult to hold on to a 2-0 lead in FUT than in the EPL. At a glance this would seem to fit well into the momentum narrative. But before we jump on to conclude anything, we however need to consider a tiny detail:
Goals are significantly more frequent in FUT than in real football matches. According to earlier estimates, the average FUT-match contains around twice as many goals as real football matches.
If we think about sports like handball or basketball, where both teams easily score +20 points / goals a match, we clearly wouldn’t consider a 2-0 lead safe or significant. In handball, it is common that both teams lead by 2 multiple times throughout the match. This goes to show that the threshold for, what constitutes a safe lead, grows with the number of goals being scored in a sport.
Give these considerations, it shouldn’t come as a surprise that it is more difficult to hold on to a 2-0 lead in FUT than in real football. At least, the mere fact that it is shouldn’t lead us to conclude that momentum is at play.
Although TapeTen’s experiment is based on a fairly small sample, it remains another entry to a growing list of evidence suggesting that momentum is a myth. What is true is that momentum shifts do occur, albeit most likely not due to EA’s intervention.
The claim that 2-0 leads are particularly dangerous may not be entirely true in real football, but it is slightly more accurate when it comes to FUT…