In this article, I will explain how we (as punters or traders) can identify a potential fixed football match. If we can identify match-fixing (and I believe we can), we have a very strong method of making money from the market.
Why? Because if you are right about the match being fixed, you are no longer dealing with probabilities. This is because, if a match is fixed and you’ve read the market correctly, you have a bet that is close to a dead cert. This is because a fixed match is, well, fixed and therefore, probabilities don’t enter the equation.
This article is about match fixing, how to read the market and how to find value.
So, let’s get started.
A. Table of Contents
I will cover 3 topics in this article.
Firstly, I will explain the difference between standard value betting and betting on the match-fix. I will explain that, when betting on the fix, you can be wrong a lot of the time, but still come out with a long-term profit.
The media often try and make out that match-fixing mainly occurs in unimportant matches, such as friendlies. Is this true or is the media trying to distract us from the truth?
So, the second topic is a discussion of whether there is match fixing at the top level of football. When I say the top level, I mean the top 5 leagues rated by the UEFA coefficient. These include the English Premier League, Spanish La Liga Primera, German Bundesliga 1, Italian Serie A and French Ligue 1.
I will also mention other leagues and incidents in the UEFA Champions League and UEFA Europa League.
Thirdly, I will explain how to read the market for suspicious betting activity. I will also explain, what I think are good ways to implement trades.
So, let’s get started.
B. How You Can Get Into Profit When You Are Wrong 98% of the Time
Everyone should be interested in match-fixing because, when you see value, it might not always be value. So, even if you’re not interested in betting with the fix, it would still be beneficial to identify and screen out the fix when you are doing standard value betting.
In the first column of this table are odds. In the second column there are probabilities that represent a team’s chance of winning.
The probabilities are based on the middle point between the back and lay odds. For example, if the back odds are 2.0 at Betfair, the lay odds are 2.02, I base the probability on odds, half-way between these odds.
So, I base the probability on odds of 2.01.
The third column is the amount of money that you would lose on average, if you closed your eyes placed 100 different bets at £10 at these odds. The 2% commission is taken into account.
Let’s say that you are basing all 100 bets on a belief that the match is fixed. In addition, you believe that you know which way it is fixed.
The 4th column shows, how often would you have to be right about the match-fixing to get into profit. The claim (in the heading) that you can be wrong 98% of the time, relates to the odds between 1.5 and 3.0. So, here the 2 certs are replacing 2 of your 100 bets. So, you would have 2 certs and 98 bets running randomly. In other words, you won’t necessarily lose just because you thought a match was fixed and you were wrong.
The 5th column shows how much profit you would make if you were right about the match-fixing when you get exactly the number of required certs, that are listed in column 4.
The key to understanding this is, if you are right about the match fix, you have something approximating a dead cert. There are no longer probabilities involved because the match is fixed.
You don’t need to be right often.
You won’t ever know for sure, which bets were based on the match being fixed and which were based on random wins. Even if you win, you could have been wrong about the match-fix, and your win could have come from a random bet.
However, as you have a lot of room for error, this is a topic of major interest.
Value-Betting versus Betting on the Fix
So, I want to make clear, the difference between standard value-betting and betting on the fix. If you saw odds of 1.76 to back and 1.77 to lay and you think the real odds for the win are 1.6, this is a standard value-betting situation.
If you bet at 1.76, you are getting value and (if your real odds of 1.6 estimate is correct) you would win the bet 62.5% of the time. If you are wrong about the real odds being 1.6 and the Betfair odds represent the true probabilities, you would win 56.7% of the time.
If you were betting because you think that the match is fixed, you would win 100% of the time that you are correct. When you are wrong about the match-fixing, you would win 56.7% of the time assuming the Betfair odds represent true probabilities.
The decision on how you bet isn’t necessarily a choice of one or the other. You can bet on both. However, there is an interaction between value and the fix. Occasionally, you can bet for value and the fix. Sometimes, you can get big value (in-play) after the fix has played out.
You might not understand exactly what I mean by this now. However, I will explain this properly in the 3rd section of the video.
C. Is Match-Fixing Just Hype?
Approximately 44% of referees believe that at least 1 in 10 Belgian football games is fixed (Visschers, Jet al., 2020).
So, now I will present the case that there is likely to be match-fixing at the highest levels of football.
We hear about match-fixing in the media a lot these days. The BBC has been putting articles out on match-fixing for about 10 years.
I first got interested in the idea of identifying match-fixing from the pre-match betting, during the first lockdown in 2020. I was reading media articles that were saying that match-fixing is increasing because of what was going on in the world. My understanding from headlines saying that match-fixing is on the increase, is that match-fixing is always there, but it is just increasing now.
This is a website called Researchgate. It’s a database of academic articles. If I put “match-fixing football” into the search box, it gives 100 articles on match-fixing. There’s 10 on each page. I think 100 is the is the maximum that they show in a search. Therefore, there could be tons more.
So, it does appear that match-fixing is real. Therefore, whatever your betting strategy, you need to read the market for potential fixed matches. This isn’t any different from reading horse racing markets, where you are looking out for gambles and non-triers. In fact, it is easier to read football markets because there are only 2 teams and you don’t have complications of bad form horses dropping down in the weights to a winnable mark or horses being off the track for a long time.
However, there is something doesn’t make sense about the media frenzy about match-fixing. A lot of media articles on match-fixing seem to be attempting to cause a panic about match-fixing on small games, such as friendlies. Not that many people bet on friendlies and where’s the liquidity? I would assume that match fixers need liquidity .
If you read between the lines of both media and science articles, they appear to want us to believe that the top leagues are clean and the match-fixing is primarily the work of rogue and possibly, illegal bookmakers and other criminals from other countries.
I’m not going to speculate as to, who might be behind the match-fixing. However, we’ve seen a lot of corruption at home and in other European countries in recent years.
I will discuss whether there may be match-fixing in the top leagues.
1. Bayern Munich versus Leeds United
So, I will start with a 1975 match in Paris in which, Leeds United may have got badly stitched up in a match against Bayern Munich. The match was the European Cup Final (which is equivalent to the Champion’s League final).
The controversy was associated with 2 main refereeing decisions. In fact there were 3 incidents and all of them involved the West German Captain, Franz Beckenbauer.
The first was a disallowed Leeds goal that was scored by Peter Lorimer. The second was Leeds, Allan Clarke being brought down in the penalty area by Bayern Munich’s, Franz Beckenbauer. In addition, there was a claim for a handball in the penalty area against Franz Beckenbauer again.
However, the way Lorimer’s goal was disallowed was bizarre.
After Lorimer scored, the Referee Michel Kitabdjian pointed to the centre circle (indicating a goal). The linesman didn’t raise his flag and ran back to the half-way line.
This is the bizarre bit. Franz Beckenbauer spoke to the referee, who then, spoke to the linesman. Then, the referee indicated offside against Bremner.
Remember, this was the final of the top competition in Europe. There was obviously no VAR back then. The question is what did Beckenbauer say to the referee?
Other than this match, I don’t think that I’ve ever seen a player get a referee to reverse a goal decision before or since, although I have seen some get booked for trying.
There is no proof that the match was fixed. However, one of the main ways in which, matches get fixed is through referees.
By the way, being a great sportsman doesn’t mean that a person is beyond dodginess. For example, no-one would have believed that the late South African cricket captain, Cronje, would have taken match-fixing-related bribes, before he got caught.
The 1970s Leeds side was a brilliant side. The players were strong individually and they played really well together as a team. However, in a Daily Mirror article in 1978, former goalkeeper Gary Sprake accused former Leeds manager, Don Revie and former Leeds captain, Billy Bremner of routinely bribing players from other teams to throw matches. There was a specific match against Wolves, in which Leeds only needed to draw to win the league. The allegations were that Bremner tried to bribe a Wolves player to allow Leeds to win. Leeds lost the match.
Nothing was proven. However, think about this. If Gary Sprake was telling the truth, it has implications that go way beyond just, dirty Leeds. For a team to be routinely bribing other teams, it requires other teams to be routinely taking bribes and those kinds of teams, would probably be bribing each other as well. That casts a shadow over the whole of the first division of that era.
As I said nothing was proven, but let’s turn to a proven example of how a bit of bribery between teams leads to most teams in the league being involved.
2. German Bundesliga Scandal 1971
Four years before the match between Leeds and Bayern Munich, in 1971, bribery did occur in the top tier of West German football. This is the first example of a proven bribery scandal.
Two clubs, Kickers Offenbach and Arminia Bielefeld, who were in relegation trouble, bribed players from other clubs to throw games.
The end result resulted in more than 50 players, managers, and officials receiving suspensions and bans. In addition, Offenbach was relegated to a lower league in spite of the fixing efforts.
This is a list of “manipulated games”. The number of teams involved adds up to 11. There were 18 teams in Bundesliga 1 in 1971. That means that 61% of teams in the league were involved in the bribery scandal.
So, let’s get to modern times.
3. The European Match-Fixing Scandal 2009
I’ll just mention that before, the 2009 scandal, there was a 2005 German match-fixing scandal, which involved €2 million and a referee, Robert Hoyzer, who confessed to fixing and betting on matches in the Bundesliga 2, German Cup, and the then third division.
The 2009 match-fixing scandal involved 9 European countries, including Germany, Belgium, Turkey and Austria.
Anyway, in the 2009 match-fixing scandal, 200 matches were investigated, including 12 qualifying matches in the UEFA Europa League and 3 in the UEFA Champions League.
So, we have Europa and the Champions League, which are prestigious competitions.
4. The Calciopoli Match-Fixing Scandal (Italy, 2006)
The Calciopoli Match Fixing Scandal is an interesting one because it shows how far the match fixers might go to get results.
This scandal is associated with Italy’s Serie A and specifically with Juventus.
Phone taps and other methods uncovered evidence of match fixing.
Club officials threatened referees with negative effects on their future earnings, to benefit Juventus. So, referees weren’t necessarily bribed – they were threatened.
Referees gave red cards to top players in matches immediately before a game against Juventus (which meant the red carded player couldn’t play in the next match). This is an unusual and a more complex form of match fixing. In this case, a earlier match was fixed to influence the result of a future match.
Referees used other actions (eg incorrect offside decisions) to benefit Juventus
So, that is Serie A and a top club in Serie A as well.
5. The French Match-Fixing Scandal
So, let’s move over to France.
The French football bribery scandal occurred during a 1992–93 French Division 1 match between Valenciennes and Marseille. The president of Marseille, Bernard Tapie and the general manager Jean-Pierre Bernès got one of their own players to ask 3 Valenciennes players to underperform in the match. They said that they did this so that Marseille could stay fresher for their 1993 UEFA Champions League Final match against A.C. Milan six days later.
He’s hardly going to say that he did it because he wanted to make sure that he got enough points to win the league.
2 of the Valenciennes players took the bribe. However, the third player, Glassman, didn’t. He grassed them up instead and was, consequently, awarded the 1995 FIFA Fair Play Award.
The scandal led to the league title being taken away from Marseille and 5 people getting convicted of corruption; 2 of whom, went to jail and 3 others getting suspended sentences.
6. The Spanish Match-Fixing Scandal
So, Spanish La Liga has had the same kind of thing. In this case, a team was trying to buy it’s way out of relegation. The result was 2 former players getting 1 year jail sentences and the Osasuna director getting 8 years.
The Sun writes, “It is believed that end of season payments to other clubs for beating rivals have been commonplace in the past, making this first matchfixing prosecution a landmark moment in Spanish football”. The Sun 24 April 2020
This is an important point. For every fixed match, that ends up with convictions, there are probably tons that of fixed matches that don’t. It’s like any repeatable crime for money. By the time that the criminal gets caught, he’s probably committed the same crime numerous times.
In addition, it’s unlikely that teams just take bribes to enable other teams to win leagues or avoid relegation. They take bribes for money. As the betting industry as big as it’s ever been, this means that match-fixing is potentially profitable for those, who partake in this activity throughout the season.
Therefore, match-fixing might not necessarily just occur at the end of the season. It can occur throughout the season.
7. The FIFA Corruption Scandal
In 2011, FIFA announced a £17.5 million plan to crack down on match-fixing. They said that the money would be spent on a 10 year anti-corruption programme to educate players, referees and match officials.
This might sound like an honourable and generous act.
However, it sounds a bit odd to me. The way I read that is that FIFA pledged to spend £17.5 million to educate players, referees and match officials, who already know that they’re not supposed to be match-fixing.
It’s already over 10 years, since the FIFA’s 10 year anti-corruption programme and the media are still going on about match-fixing. So, I guess that FIFA’s £17.5 million plan wasn’t successful.
Anyway, 4 years after FIFA’s £17.5 million plan to crack down on match-fixing, FIFA, themselves, were found to be involved in a major corruption scandal. The FIFA corruption scandal wasn’t match-fixing but it involved a lot of money and a lot of people.
When the organisation that is supposed to be overseeing football, gets found to have large scale corruption within its own ranks, it’s unlikely that match fixing is going anywhere soon.
In addition, it’s not unreasonable to wonder if match fixers might only need to grease the right palms to get permission to continue.
8. Match Fixing Summary
What do we know? We know that match fixing has occurred, at least, since the 1970s. The cases that have been discovered in football have mainly been on the inside of the football industry. So, most fixed matches (that have ended up with arrests) have involved teams bribing each other, although the 2005 Bundesliga 2 involved a referee, who was bribed by a dodgy gambling syndicate.
There have been match-fixing cases in 4 out of the top 5 European Leagues, including La Liga, Bundesliga 1, Serie A and French Ligue 1. In addition, match-fixing has occurred in UEFA Europa and Champions League qualifiers.
I’m not interested in the morality of match-fixing. I am just interested in identifying match-fixing and differentiating normal value from match-fixing.
There’s no solid evidence of match-fixing in the Premier League. However, there has been match-fixing and corruption almost everywhere else. The question is, do you think it is possible that the Premier League could be clean, when there has been corruption in so many of the top leagues and in FIFA itself?
Then, there are bookmakers. If you google gambling industry fines, you will see that they have been fined for “money laundering failures”, accepting stolen money and social responsibility failures of course. So, the bookmakers don’t really have, what you might call a faultless resume.
So, I will leave you decide whether the Premier League could be clean, within these types of surroundings.
D. Reading the Market for Suspicious Betting
In this section, I will explain how I identify suspicious betting.
I will start with this paper by Feustel & Rodenberg (2015). Their ideas regarding identifying suspicious market behaviour are similar to what I have been doing. However, this paper was published several years before I even thought about working on suspicious market behaviour in football. In addition, I’ve gone deeper than they have.
I’ll tell you what they did first.
They looked at matches from 4 leagues
(i) EPL
(ii) USA MLS
(iii) Italian Serie B
(iv) French Ligue 2
1. They used the Poisson distribution model to calculate percentages or odds for each match
They measured suspicious market behaviour by percentage difference between the Poisson results and the market closing price. The more a betting market’s closing price differs from the model price, the more suspicious it is.
2. If corrupt money was bet, the authors expected to see:
Closing prices that were significantly different from odds generated by the Poisson model
The corrupt money winning at a significantly higher rate than the market’s odds
Note that if you were blinding using Poisson, you would be seeing value where these authors would be seeing suspicious market behaviour. This is why being able to differentiate between value and suspicious market behaviour is important.
Obviously, the authors couldn’t say that matches were definitely fixed. However, they implied that fixing may have been occurring in French Ligue 2 and Italian Serie B, but not in the English Premier League or USA’s MLS.
They found matches with closing prices that were 19% or more different from the Poisson results in terms of implied odds in all 4 leagues. So, the first condition of the analysis was met in all 4 leagues.
However, the second condition, of corrupt money winning at a significantly higher rate was met in 2 of the leagues, namely French Ligue 2 and Italian Serie B.
In both French Ligue 2 and Italian Serie B, the 8 most suspicious matches were played near the end of the season and the result had a significant impact on a team’s chance of getting promoted or risk of relegation.
E. My Method
So, the study, that I just described, supports the idea that corruption might occur in situations in which teams are at risk of relegation or they need points to gain promotion. As I’ve said, I think match-fixing can occur at any time during the season, because, according to the media, match-fixing is also instigated by gambling organisations (who apparently don’t seem to get caught very often). Such gambling organisations are not interested in promotions and relegations. They are just interested in the money.
However, in contrast to the study, I use a few more ideas in my analysis.
The first point about the paper, is that the authors didn’t look at the team sheets. So, I do look at the team sheets. If I see market odds that differ significantly from the Poisson results, to view this as potentially match-fixed, I want to exclude the team sheets as being the reason for the odds difference.
The second point is that the authors just looked at Poisson and the closing prices. I look at the whole timeline. I will explain in a minute how that can help us to read the market.
The third point is that the authors of the study just looked at match results. In other words, they looked at whether corrupt money could win on the whole match result.
Again, I do more than that. I also count a match as potentially corrupt if the corrupt money could have hedged for a profit at any time during the match. This is especially the case when the long-shot just happens to score a goal early in the first half.
Let’s say that I see an odds on favourite drift and a longshot getting heavily backed. So, this is a graph of the betting. Let’s say that this looks like suspicious betting. In other words, I think the favourite should be significantly shorter in the betting while the longshot should be significantly longer.
Let’s say the match goes in-play. The odds start by drifting and then, the outsider scores early in the match. Now, the corrupt money can hedge for a profit. Even if the outsider goes on to lose, I would view this as a potentially match fixed.
I’ll go into a more detail on how I differentiate between a potentially fixed match and value in a bit.
1. The Time Line
So, this is how I see the football timeline.
At each point across the timeline, there will be a mixture of types of money in the market. The mix of players will increase as you get closer to the start of the match. When you have different types of players all doing different things, this is often referred to as “market noise”.
The first part is I day before the match. I’ve put one day before the match. However, this could be even longer, like 2 days or a week.
There will be punters and traders all through the timeline. I’ve just put names, such as form-readers, early team sheet money and inside information money, to emphasise, who is likely to be active in the market or possibly, who might be influencing the market most at each point in time.
In addition, these aren’t exact times that the different types of players enter the market. So, the moves might occur earlier or later and this might depend on how fast the liquidity builds up in the market.
The types of people, involved in the market at 1 day or before, are likely to be form-reading punters and traders. They’ll be looking for value. They won’t be thrill-seeking or instant gratification punters. Such punters will usually get involved closer to the start of the match.
Because these are form readers and liquidity is low, if anything dodgy is planned, it won’t usually show up yet.
From 5 hours to the start of the match, there will be relatively more activity. At around 5 hours before the match, you might start to see money that comes in from people, who have got the team sheets early.
The team sheets are usually published online 1 hour before the start of the match.
I’ve heard that journalists get the team sheets early so that they can print them out immediately, after the clock hits one hour pre-match. Although they are probably not supposed to do this, some journalists sell the team sheet information early.
This means that there will be some punters, who know who’s on the team sheets before the 1 hour pre-match mark. In addition, the bookmakers will almost definitely know who’s on the team sheets. If punters have got enough money to obtain the team sheets early, I’m sure the bookmakers can afford to buy this information as well.
So, the effects of the team sheets can sometimes show up in the betting before the 1 hour mark.
After 1 hour pre-match, I’ve put the emphasis on insider money. This is because insider money is usually big money and these players need liquidity.
Don’t forget that some of this activity can overlap.
2. Looking for a Betting Pattern that Doesn’t Make Sense
So, how can we identify potential match-fixing? Basically, if the betting pattern of the market doesn’t make sense, there may be something going on.
So, you would look at graphs, against a background of having already estimated the odds using the Poisson model and looking at the odds, when the teams have played each other in the past.
You also want to exclude recent form and weak team sheets as a reason for odds movements.
I’ve gone through the top 3 on the checklist in the previous 2 videos.
Graph 1
So, in this graph, the Poisson result is represented by the blue horizontal line. The odds of a short-priced favourite open higher than the line and they don’t do much until 1 hour pre-match. Then, the odds drift.
So, the last part of the pattern doesn’t make sense. You would expect the odds to stay where they are or move towards the expected odds.
When value looks too good to be true, it might be.
Another point, that I want to make is if we move the expected odds up and keep the shape of the graph the same, the graph now makes sense.
This is because, in the first part of the graph, the odds were lower than expected. This isn’t that unusual for a short-priced favourite. Then, in the last hour, the odds moved towards their expected price.
This is why, you need to have an idea of what the expected odds should be, in order to interpret the graph.
So, when you see a late sharp movement away from the expected odds, you should be suspicious. It’s not so suspicious when a short-priced favourite is overbacked below the line. The market will sometimes overback a favourite (or the bookies might hold the price of a favourite down).
However, when the longshot is heavily backed and the short-priced favourite drifts, that is suspicious. That’s because most longshots don’t have the players to control the game. Generally, a weak team is responding to the better team, rather than being proactive and making the better team respond to them.
3. How do we make money from this?
As the strong team controls the game, it is the strong team that I follow. If I’ve never seen the strong team in combination with this betting pattern, I watch how the match plays out. If I can see a way of getting value, I get involved the next time I see the same betting pattern involving the same team. So, basically, when I’ve seen how the match plays out, I will try to plan put how I will get involved next time I see the same team and the same betting pattern.
You might think that the same betting pattern won’t be used. However, you will be surprised. I’ve often seen the same pattern used in the next match.
Towards the end of last season, I saw this betting on a match between a weak team and a very strong team. This is a graph of the short-priced favourite. It hadn’t done anything significant in the betting coming up to the 1 hour pre-match mark. However, it’s odds were above the expected odds. After 1 hour pre-match, the odds drifted, while the outsider was getting heavily backed.
The strong team was doing well in the league but couldn’t afford to lose points, if the team wanted to stay in contention to get what it was aiming for.
So, the match started and the heavily backed outsider scored a goal, within 15 minutes into the first half. So, at this point, pre-match longshot backers could hedge for a profit.
The favourite went on to win 2-1. The favourite was the away team in that match.
In the favourite’s next away match, the pre-match betting was, more or less, exactly the same. I decided to wait to see if there would be an early goal from the long shot. Again, the long shot scored within 15 minutes.
So, now, I’d seen 2 parts of the previous match repeat – the betting and the early goal. At this point, I still couldn’t be sure about the details of how the match would be fixed after the first goal had been scored. I didn’t think that the favourite would simply give 3 points away. So, that left 2 possible scenarios.
The first would be an agreement that the longshot will score first and then allow the favourite to win. The second is that there would be an agreement to score one goal each, which would make the match 1-1 and then, play a proper football match from that point on.
In any case, I backed the favourite after the goal and the favourite went on to win 2-1 again.
Yes, it may have been a coincidence that, 2 matches in a row, longshot backers could have hedged for a profit after the early goal in 2 matches. You’ve still got to wonder why weak teams were getting backed so heavily. As I said, earlier in the video, you never really know if you have identified a fixed match or if you just got a random win.
Ok, let’s say this was a coincidence. However, there is another way to think about this. At the start of the match, the favourite is at value odds because it is above the Poisson line. When the goal goes in, it is likely that the favourite is still value.
I don’t think it’s unreasonable to assume that, after a goal is conceded, the odds of the favourite will increase relative to the odds that it started the match at. Therefore, a team, that was value odds before the match, should be value odds after they have conceded a goal.
Therefore, even if the match wasn’t fixed, I should still be getting value. Earlier in the video, I said that, occasionally, you can bet on the fix and value at the same time. This is an example. So, even if you think a match is fixed, don’t ignore value. It helps if you can get both.
This is about betting on history to repeat, in combination with getting value.
D. Conclusions
The idea of this video isn’t to explain a single trade, but rather a way of using the pre-match betting to think through trades.
1. Corruption has (and probably still does) occur at the top level of football
Don’t expect to see match fixing around every corner. Although match-fixing can occur throughout the season, there may be a larger percentage of fixed matches close to the end of the season.
2. Reading the pre-match betting helps to identify implied value when you are betting/trading in-play
Even if I’m betting at half-time, I’m still conscious of how the pre-match betting went and figuring out whether the bet that I’m thinking of placing is implied value. In an example in this video, I explained how you might be able to deduce whether you are getting implied value in-play, based on the closing pre-match odds.
3. At the start of the video, I explained that, when you are looking at a match, that you think might be fixed, you don’t have to get it right that often to be in profit in the long-term.
So, now you know about match-fixing, don’t forget about normal value.
Anyway, that’s all for this video. Take care and Good Luck.
References
Feustel, Elihu & Rodenberg, Ryan. (2015). Sports (Betting) Integrity: Detecting Match-Fixing in Soccer. Gaming Law Review and Economics. 19. 689-694. 10.1089/glre.2015.19105.
Visschers, Jonas & Paoli, Letizia & Deshpande, Abhishek. (2020). Match-fixing: Football referees’ attitudes and experiences. Crime, Law and Social Change. 74. 10.1007/s10611-019-09880-3.