Why 90th minute goals are frequent in FIFA

Share

One of the absolute classics in regard to scripting, handicapping and momentum is related to goals happening in the 45th or the 90th minute. As illustrated by the quotes below, a lot of people believe that such goals are part of a sinister plan to change the outcome of our matches:

“Why is it there are so many dramatic 45th and 90th min goals? Yes one could argue that you have injury time and are therefore stuck on that minute for a while longer. But its plain too see that whilst you get ever closer to this minute the pressure and momentum ramps up.”
(user Angry_Gooner on reddit.com)

“But in the 45th minute. i think there’s a script to enable bullshit in ,added time and against silver/bronze teams.”
(user justlur on reddit.com)

“The scripting is completely insane in this game. The opponent ALWAYS scores in the 90th minute. Some games I’m winning 3-0 so I mean it’s alright but when it’s 1-0 or 1-1, it is infuriating.”
(user Vampire_Prince10 on Reddit.com)

One thing is for sure: We fully recognize the frustrations of conceding a late equalizer. But as it happens, there may be a less sinister explanation than scripting.

How frequent are 90th minute goals?

It would nice to know how often these late goals actually occur. EA doesn’t publish any official statistics, but we have a variety of data sources on our hands, and this allows us to come up with an educated guess.

The first data source is our own 2014 survey. In that survey, we found that approximately two-thirds of the players experienced decisive stoppage time goals in minimum every other match.

Q: How often are matches decided by 90th minute goals?

Although this method isn’t highly accurate, and probably doesn’t allow us to establish the exact frequency of late goals, it does seem to support the claim that a lot of people experience late goals on a very frequent basis.

In addition to the survey, a redditor called Arlington69 shared a data set covering 761 of his own matches in total. In that data set, he recorded the timing of all goals. According to Arlington69’s data set, 45 % of his matches contained goals scored in the 45th or the 90th minute. Arlington69’s data is not representative to everyone, but together with the earlier survey, it seems reasonable to assume that most players will experience goals scored in the 45th or 90th minute in every other match give and take.

But the mere fact that late goals happen frequently doesn’t lead us to conclude that EA makes them happen deliberately.

EA’s incentive – if there is one

Before one gets too deeply entangled in speculations about a conspiracy, it seems fair to consider what EA’s possible motive possibly could be. And despite the fact that we have asked that question many times, we so far haven’t seen anyone present a credible motive for EA to make goals happen in those particular time slots.

We have seen people claim that those goals are put in as part of a plan to help bad players out, but we have a hard time seeing how goals scored in the 45th minute would support that aspiration to any larger degree than goals scored in the 52nd minute. Furthermore, it seems somewhat counter intuitive to describe a player trailing by a single goal as particularly bad relative to his opponent.

The absence of a credible motive is definitely a reason to remain skeptical, although it doesn’t rule anything out per se.

A long minute

The next question one should ask is whether there perhaps is a natural explanation. And as it happens, there is.

A notable detail about the mysterious “45th minute” / “90th minute” is that those terms in fact often don’t refer to a minute as such, i.e. 60 seconds.

Often when people complain about “45th / 90th minute goals”, they either mean (a) goals recorded as 45th or 90th minute in the post match report or (b) goals scored while the clock was stopped at 45:00 / 90:00, i.e. during stoppage time.

To illustrate why it’s a bit of a misunderstanding to refer to the above goals as either 45th or 90th minute goals, take the example where the fourth official shows a “5” sign as below.

4th Official Phill Dowd shows 5 minutes extra time to be played

As most people probably know, the sign tells that a minimum of 5 minutes will be added. This implies that the referee may add extra time if new stoppages occur during stoppage time itself.

But if the referee adds 5 minutes, it goes without saying that only 40 effective minutes were played during ordinary time while another 5 effective minutes will be played during stoppage time.

Under such circumstances, the average minute of the ordinary match only contained 40 / 45 * 60 = 53 seconds of effective playing time meaning time where players potentially could score goals.

To complicate things even further, FUT’s post match reports list goals scored between 89:00 and 89:59 as 90th minute goals because 89:00 marks the beginning of the actual 90th minute of the match.

So, if 5 minutes are added, a total of 5:53 minutes of effective playing time is played during the so-called “90th minute”. And logic dictates that the probability of scoring a goal in a time span with a duration of 5:53 minutes is approximately 7 times larger than the chance of scoring a goal during a time span lasting only 53 seconds.

However, the lengthiness of the 90th “minute” isn’t the only contributor to the excessive number of late goals.

How the scoring chances increase towards the end of a half

The average number of goals per minute in a match as a function of the time (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0047678)

FIFA is a football simulation, and that inevitably means that it carries many of the same traits as real football.

In real football, goals aren’t evenly distributed across the match. Statistical analyses demonstrate that the goal ratio increases throughout the match and particularly so towards the end of either half. The closer you are to the end of either half, the bigger the chance of scoring during the next minute.

In the EPL, the goal ratio between the 30th and the 45th minute is 70 % higher than the goal ratio between the first and the 15th minute, and during the first rounds of EURO 2016, nearly 1/3 of all goals were scored in the dying minutes of the match.

Image (c) Telegraph.co.uk (http://www.telegraph.co.uk/football/2016/06/17/euro-2016-on-course-to-break-record-for-late-goals-scored-but-wh/)
Image (c) Telegraph.co.uk

A number of factors contribute to late goals being more frequent in real football: Psychology and fatigue imply that players make more mistakes. Many teams apply different tactics during the last minutes of a match.

There is no immediate reason to assume that FUT is fundamentally different. After all, all the needed ingredients are present:

  • Humans, who are able to become nervous or mentally fatigued.
  • Virtual footballers, who will perform worse and make more mistakes due to fatigue.
  • The ability to make tactical changes, where you take a bigger risk in the dying moments.

With regard to the psychological and tactical aspects, it’s worth noticing that not only FUT Draft and tournaments but also FUT Seasons contains a large proportion of knock-out matches.

Knock out matches affects the psychology

As explained in our article on the difficulty of promotion and relegation matches, the combination of a very brief season and three different point thresholds implies that anything between 2 and 9 matches per season are deciding promotion, relegation or winning the title. In our article on decisive matches, we assess that most players in fact will experience that significantly more than 50 % of their FUT seasons matches are potentially decisive.

A notable fact to consider in this context is that even if player 1’s match may be indecisive, player 2 may be having his final chance of securing promotion. Hence, if 50 % of player 1’s matches are decisive and 50 % of player 2’s matches are decisive, an estimated 75 %  of all matches will be decisive to minimum one of the involved players, meaning that minimum one of them has a reason to go all out attacking to claim that final, decisive point – or – become nervous towards the end and concede a stupid goal.

Many goals = many goals in stoppage time

A third factor which influences the occurrence of late goals in FUT is the high overall goal ratio.

Liverpool4-3

In a recent 8-0 win, I managed to score twice in “the 90th minute”. Was this scripting?

According to some scripting believers, EA scripts matches to make them more even. But clearly, my 7-0 and 8-0 goals didn’t make the match the least bit more even. Considering that I was playing against an opponent who lost 9 out of his 13 FUT seasons matches this year, these goals hardly were an attempt to help out a noob against a better player.

But my example illustrates an inevitable fact: If there are many goals in general, there will be many goals in stoppage time – and in any other random range of time you may define.

A real life example of this is Liverpool’s 4-5 win against Norwich on January 23rd 2016. The match contained two, potentially decisive 90th minute goals. According to BBC’s detailed commentary, Bassong equalized at 90 + 1:57, whereas Lallana decided the game at 90 + 4:19. The match finished at 90 + 5:33. A notable detail is that the two goals scored in the 54th and 56 minute actually happened closer to one another (1:40 in between) than the two goals scored in “the 90th minute” (2:24 in between).

FUT’s extreme goal ratio

In an earlier article, we counted the average number of goals across 400 FUT 15 matches, which is a fairly large sample. We found that the average was 5.5 goals per match. An average of 5.5 goals per match is extremely high when compared to real football. Among the five big leagues, the Bundesliga has the highest goal ratio with 2.82 goals per match.

At 5.5 goals per match and 1/9 of the goals scored during stoppage time, we should expect to see .61 goals scored in stoppage time per match, assuming that the scoring probability remains constant throughout the match.

Of course, the ratio of goals scored in stoppage time will vary depending on the number of goals and the number of added minutes. Both parameters will depend on your playing style.

The table below shows the expected frequency of stoppage time goals with different combinations of added minutes and goal per match ratios. What you should notice about the table is that even if we look at the more conservative section of the table, stoppage time goals still happen very frequently.

Minutes added per half
+2 +3 +4 +5
Goals/match 3.5 0.16 0.23 0.31 0.39
4.5 0.20 0.30 0.40 0.50
5.5 0.24 0.37 0.49 0.61
6.5 0.29 0.43 0.58 0.72

Expected number of goals per match scored in stoppage time

The point we want to make with this table is that, unless EA intervened to prevent stoppage time goals from happening, these are the stoppage time goal frequencies we should expect to see. If EA – for some unknown reason – indeed does like us to score many stoppage time goals, they really don’t need to do anything to make it happen.

Under what circumstances do those goals happen?

Stoppage time goals are frequent, but is it true that they mostly come in the shape of late equalizers? With Arlington69’s aforementioned, detailed data set, we are able to look into this from a fact-based perspective.

This is what we found:

Occurrences
Matches total 761
  Matches with stoppage time goals 45 % (339)
  Matches with stoppage time goals, which cause turnover  18 % (136)
Goals in total 4576
  Equalizing goals 17 % (776)
Stoppage time goals in total 409
  Equalizing stoppage time goals 14 % (59)
  Stoppage time goals which cause 1-goal lead 22 % (92)
  Stoppage time goals which increase a lead 63 % (258)

There are many important takeaways from the table above, but there are two particular highlights:

Firstly, most stoppage time goals aren’t equalizers. In fact, 2/3 of all stoppage time goals increase an existing lead and another 1/5 stoppage time goals cause someone to take a 1-goal lead.

Second, our sample shows that equalizing stoppage time goals in fact occurred slightly less often than equalizing goals in general (the difference is hardly statistically significant, though).

Hence, Arlington69’s data effectively rules out that stoppage time goals play any particular role in making matches more even.

Why 90th minute scripting is a myth but bullshit goals aren’t

We have seen above that there is a perfectly natural explanation for the large number of 45th / 90th minute goals. And absolutely no evidence suggests that there is a residual of goals which cannot be explained by the natural reasons above.

Further, we effectively can rule out that stoppage time goals play any particular role in regard to making matches more even.

Given the above, it is fair to conclude that the claim about 45th / 90th minute scripting is a myth.

Having said that, the facts listed above tell us something else about the goals we score and concede in FIFA:

As stated, FIFA has an extreme goal frequency. With 5.5 goals (FIFA 14) in a match which lasts just 12 minutes, we end up with 0.45 goals per minute, whereas the average Bundesliga match has a frequency of 0.02 goals per minute. We are not even close to realism here, and that is hardly a coincidence.

If we got the chance to ask them, EA most likely would confirm that the goal ratio is intentional. FIFA probably wouldn’t have sold 12 million copies a year, if the goal ratio was at a more realistic level.

So, while rejecting the claim that EA has a particular preference for 45/90th minute goals, it is beyond dispute that EA are “at blame” for making it relatively easy to score and hence inevitably also to concede goals. After all, they are in this business to entertain us – not to bore us to death.

So, “bullshit” goals are real, bullshit goals happen during stoppage time – but 90th minute scripting doesn’t exist.

%d bloggers like this: