Category Archives: Football

FIFA World Cup Predictions

With the week getting away from me, the analysis will be short and sweet!

In fixed odds for the tournament win, it is difficult to look beyond the four of Brazil, Argentina, Spain and Germany (ranging from 3.75 to 7.00, and 23.00 for France who are fifth favourites). But I don’t buy into the ‘European teams can’t win in South America’. There have only been four South American tournaments, and none since 1970, whilst football in terms of conditioning, preparation and ‘internationalisation’ has come on leaps and bounds. Whilst the gut thinks that Argentina have the best chance, I would favour Germany for their preparation, good team ethic, relatively kind draw and big tournament experience, not to mention the strength in depth that I don’t think Argentina and Spain have, whilst Brazil are at a premium as hosts.

The other option is the play the exchanges, or spread betting if that is preferred. Belgium are a team I highly rate, but at 26.00 for fixed odds and 28.00 on Betfair to win, there just isn’t much value. Without a major team in their half of the draw having a nightmare and not making it into the second round, it is unlikely that they will come in that much before the quarter final (which is where I would want to be getting out). But they may be worth a risk.

Chile, at 55.00 on Betfair, appear to me to offer good value. I can see them surprising some teams, and those odds should come in as the tournament goes on, enabling some profit to be taken before the quarters.

So, short and sweet – and with a pinch of salt!


EPL: The Palace Experiment – Betting on a Promoted Team

Betting is really a matter of finding value. Bookies aren’t mugs, and the odds that they offer are generally a fair representation of the underlying probability (slightly less usually, to give them their margin), and so the aim is to find those cases where the odds are not quite right. When multiple opportunities are available, such as in a horse race, there is a much higher chance of finding such value than in a football scenario with only three options – win, lose or draw. That is one reason why accumulators combining the results of multiple matches, goal supremacy, Asian handicaps etc. have risen in popularity in order to find some more markets where an edge is possible. But bookmakers also generally react to the market itself – they may think the appropriate odds are something in the region of 3.50, but if the money is flowing in on that selection then it will come in (and usually the other side goes out).

Finding value is often either a matter of a gut call, or the application of statistical analysis, or – as is more often the case – a combination of the two. Knowledge of the sport is essential, knowledge of what has happened in previous games/seasons/tournaments is vital. Whilst I have tried this approach with posts in this blog (particularly on F1 this season), there are many others that are worth reviewing. There are many covering exchange betting, but you could do a lot worse than checking out Dav Aulak’s blog ( which covers football, tennis and a number of other sports on both sides of the pond. His picks are always accompanied by a clear analysis, and someone looking at taking their betting forward won’t go far wrong by checking the blog out (and this is from a Man City fan, the relevance of which will come clear if you visit).

So for the 2013-14 season I wanted to look at a season long market that could be checked and tested. I had read (and I apologise for the lack of an original source) that the promoted teams are often under-rated when it comes to their home matches. In this case I thought that Crystal Palace would be the team that could offer the best value, and so backed them for the win for each match at Selhurst Park (fixed odds) and laid the opposition on the exchanges (to cover both a Palace win and a draw).

With the fixed odds, the 19 home games came in at an average of 4.75 for a Palace win. That only goes down to 3.26 when the final top 4 are excluded. There were 8 wins from the 19 matches, for a +9.05 return on level stakes of 1 unit. Of course, the big result here was a win at 10.00 at home to Chelsea, but this is exactly what was expected in the theory – the odds offered would be higher than they should be, and at least once the team concerned would provide the upset.

On the exchange side of things the level stakes were at 2 units. This market returned a profit of +5.94 units. There were 11 winning matches here, and the average lay price was 2.48.

But let us not get too excited and think that this will result in 9.05x(stake unit) for every season! Palace finished with a home record of W8 D3 L8. The other promoted teams were Hull (7/4/8) who may have had a positive return (but I believe the odds would have reflected higher expectations of Hull – another thing to check!), and Cardiff (5/5/9). That said, Cardiff did beat Man City at home early in the season, and may have broken even but I doubt they would have hit the highs of Palace.

So we can restrict the conclusions to the play-off winners. Palace went up as the play-off winners having finished in 5th place in the Championship. What of this season? Leicester City won the title, and have been playing at the sharp end of the Championship for a while. I can see them doing rather well next season in the Premiership, and I am sure the bookies will as well. Burnley accompanied them as the other automatic promotion. This time the play-off winners were QPR, back up in the Premier League from a short stay in the Championship. Will they provide the same returns? QPR’s last stay in the Premier League was a disaster, notably at home where they won only 2 games, but a year out will have changed their approach and allowed them to get rid of a number of freeloaders. I don’t think the same value will be on offer this season, but it will be worth repeating the experiment to see if lightning strikes twice.


EPL Prediction Review

Well last time I made my first set of predictions based on the use of goal expectancy. I was concerned, and mentioned it at the time, about the sparse nature of the data underpinning the predictions, and I was right.

Nine predictions were made, with the top three scorelines expected. Of these, only one top prediction matched the actual result: Arsenal’s 2-1 victory away at Swansea – and yet this was the one that had the lowest % of all top predictions! The only other one of the top three probably scorelines that hit was Hull’s 1-0 victory over West Ham which was the second most likely score.

So the scorelines weren’t good, but what about the top probabilities in relation to the actual 1X2 result? Even worse, with only Arsenal’s game a hit. That only leaves the 2.5 goals market of the ones that I look at. Four out of nine matched the right over/under prediction.

The conclusion? As it stands the model isn’t very good at all – certainly wouldn’t provide a profit to level stakes in the real market (but would if implied odds were exceeded). The 1X2 results were equally poor, and although the over/under 2.5 results appear better, this is an ‘easier’ market to predict and so four out of nine has to be marked as disappointing. There are a number of reasons for this, but my belief is that the primary concern is the lack of underlying data. Predictions based on just four or five matches just isn’t solid enough, and so I don’t intend to repeat the exercise until teams have 10 games under their belt (in general – postponements and TV schedules truly mess up the number of games played). Then we’ll see how much needs to be changed – and I fear it will be a lot.

EPL: Goal Expectancy Predictions

This year (well, since July anyway) I have been looking at the various statistical and mathematical models relating to sporting events. The idea is simple, use statistics to identify probabilities and patterns and see whether the accuracy of such models can exceed the ‘gut call’ approach. Now I have a fair amount of the gut call data to hand, and I know it isn’t that good! The main problem is that there are a lot of models out there, and even the most straightforward concepts have been refined by people with far more experience and statistical knowledge than me. It can be very daunting, and so rather than trying to start with a refined model I thought I would start with a base model and see how well that performs, and how to amend it.

I am working on a model based on the concept of ELO Rankings, but that will wait until later in the year as far more games are required before an effective rating emerges for each team. Consequently I will start with a basic prediction model based on ‘Goal Expectancy’. This model predicts the number of goals a team will score, and using Poisson distribution can be used to predict the likelihood of various scorelines as well as (naturally) the likely winner of a game. In addition the data can be used to indicate the status of a game in the typical Over/Under markets.

As indicated above, I am not trying to take credit for the base formulas and calculations. In the case of goal expectancy and Poisson distribution I have taken the work from an excellent blog by SoccerDude. I heartily recommend his blog to anyone interested in football statistics, whilst acknowledging any errors or omissions as my own!

The principle underlying this method uses the number of goals a team has scored thus far, either in total or home/away, and combining that with a Poisson distribution to get a probability for the number of goals that will be scored by a specific team. By repeating this for both the home and away team a list of probabilities for the number of goals scored and multiplying them together gives the probability for that scoreline. Simple really!

Goal expectancy has three values: the overall rate (total goals scored/matches played), the home rate (goals scored at home/goals scored away) and the away rate (goals scored away/goals scored at home). The goals expected from any team when playing at home is calculated as ‘overall rate * home rate’, and for a team playing away it is (if you can’t guess…) ‘overall rate * away rate’.

So an example shows that, as of this weekend, Fulham have scored three times in five matches (overall rate 0.60), with two goals at home and one goal away (home rate 2.00). This gives an expectancy of 0.60*2.00 or 1.20. Cardiff have scored four in five games (overall rate 0.80), with three at home and one goal away (away rate 0.33). Their expectancy is 0.80*0.33 or 0.27. By plugging the number of goals we are interested in, along with the respective rate, we can get a probability for the number of goals the team might score, as shown below. (The reason these are pasted in as images is that I really haven’t got a good handle on tables in WordPress yet!)

Home Away

The probability of any scoreline is therefore calculated by multiplying one by the other, e.g. 1-2 is 0.36143*0.02723 or 0.00984. This equates to 1%, or odds of 101.60 (using the 1/x formula). The grid is shown below.

Fulham Cardiff

So what does this mean for the weekend’s fixtures? Well the three most probable results (a dubious definition!) are shown below along with the probability and implied odds for each. The Spurs-Chelsea games has no data simply because Chelsea haven’t scored away from home yet!

EPL 260913

But, before we all go and put the mortgage payments on some nice accumulators, let’s be clear that there are some significant issues here. Firstly, and in my personal opinion with the level of understanding I have of statistics at the moment the most important, is that this approach doesn’t account for the quality of the opposition that the goals were scored against. If two teams have played at home twice, and both have scored 10 goals whilst conceding 2, there is something of a difference if one team was playing Chelsea and Man Utd and the other was playing two teams at the bottom of the table. There is also – I am reliably informed – a problem with Poisson distributions that they don’t marry up well at their default values (0-0 and 1-1 is under forecast whilst 1-0 and 0-1 are over forecasted – that worries me as 6/9 of my predictions are 0-0!). And finally the most obvious problem – basing goal expectancy after five games of the season isn’t going to provide a very large data pool to work with.

But this is a first shot into this world, so let’s keep it simple for now and run this base model a few times to see how it performs before looking at amendments.

Premier League ’13-14 Final Prediction

Having looked at the mathematical models, and the gut calls, it is time to combine them to provide a final prediction of the table as it will appear at the end of the 2013-14 season.

Pos Team Pts
1 Chelsea 80
2 Manchester City 78
3 Manchester United 74
4 Arsenal 74
5 Liverpool 70
6 Tottenham Hotspur 64
7 Everton 60
8 Swansea City 51
9 West Bromwich Albion 49
10 West Ham United 45
11 Fulham 43
12 Southampton 43
13 Cardiff City 40
14 Aston Villa 40
15 Newcastle 38
16 Sunderland 38
17 Stoke 37
18 Norwich 34
19 Hull City 25
20 Crystal Palace 22

(Apologies for the table formatting, not completely used to WP yet)

I guess we’ll see at the end of the season just how close – or far – I was!


Premier League ’13-14 Gut Calls: 1st to 10th

After the last post looked at the bottom half of the table, it is time to look at the top half and the European qualifying places not to mention the Champions!

  • 10th – West Ham United
    • Same old same old for The Hammers – they have fallen into the same predicted place for each model and that is the place that they finished last season. Perhaps a Cup run is due before they move to the Olympic Stadium.
  • 9th – West Bromwich Albion
    • The Baggies have surprised many since coming back up to the EPL, and appear to have put the ‘yo-yo’ club moniker to bed. They don’t yet have the quality or strength in depth to break into the European places but they should still be the ‘best of the rest’ and could be closer to the others if Anelka still has his goal scoring boots.
  • 8th – Swansea City
    • Another team that has surprised many, and actually improved during their time in the Premiership. The League Cup win last season was nothing less than they deserved, but it remains to be seen how they will manage a challenging league campaign along with European competition. Michu will need another standout season, but the constant improvements seen thus far justify an 8th place.
  • 7th – Everton
    • A lot here depends upon the team keeping its ‘star’ players. If Baines or Fellaini head over to Old Trafford it is unlikely their replacements will be of the same quality. And whilst Martinez is certainly a scrapper, his experience at the sharp end of the Premiership is limited meaning he is something of an unknown quantity. Assuming the players do say another top half finish should be straightforward, and perhaps a little closer to those above, but a challenge for the top is still beyond a team that doesn’t have the resources – or the desire to get into debt – to attract stellar names.
  • 6th – Tottenham Hotspur
    • As with Everton, the position Spurs will occupy depends on the remainder of the transfer market. Yes, Bale is that important to the team. Spurs appeared to overperform last season, and although the defence was impressive at times it is unlikely to be as easy this year. AVB does appear to fit the team well, so I don’t see any tremendous fall from grace, but if Bale goes that will not be an easy gap to fill.
  • 5th – Liverpool
    • Whilst I remain unconvinced about Rodgers ability as a manager of a top flight team (and I am sure that gives him sleepless nights!), Liverpool were a far better team last season than their final position would indicate. The defence has solidified, although the loss of Carragher will be keenly felt, and Mignolet will be under a lot of pressure replacing a well loved Reina. The midfield has relied too heavily on an aging Gerrard (still extremely talented, but every game?), but the forward line is where the real danger lies. Too reliant upon Suarez the transfer shenanigans could have terminally damaged that relationship. Without Suarez (either gone in a transfer or just not as good without the motivation) Liverpool could struggle for goals, although I think the other attacking options will step up.
  • 4th – Arsenal
    • Fourth again, but I expect them to be closer to the top three this time. The defence at the Emirates is solid as we know, but last season showed a need for more forward work. A quality ‘out and out’ striker (someone like an RVP perhaps?!) could see Arsenal mount a serious challenge, but such a player costs significant money and Arsenal just never seem prepared to spend.
  • 3rd – Manchester United
    • What? Manchester United third? Really? Yes, really. After what can only be described as an outstanding season in 2012-13, things may be trickier this time around. Last year RVP was sensational, but it is rare that a player can produce such a phenomenal season two years in a row. Defensively there is still a question mark, not on the quality of the players but on the ability of Ferdinand and Vidic to last a season. Midfield still seems somewhat light (for a team of such stature, not when compared to many others in the league), and the change in manager is likely to have some effect. This won’t be the dramatic fall from grace that some have predicted, but it was always likely to be a tough season, made worse by an expected improvement from their chief rivals.
  • 2nd – Manchester City
    • City’s defence of their EPL title was disappointing to many. The players were still there, in the main, but they weren’t brought together as a strong unit. Reports, if they are to be believed, would indicate that Mancini was a major factor in this, and similarly reports indicate that Pellegrini has addressed this. Reports, as we know, are notoriously unreliable though… There have been significant purchases over the Summer to offer alternative attacking options and to strengthen the midfield, and these were completed early in the transfer window allowing the players time to get to know the rest of the squad. There is still a concern over the defence, and the reliance on Kompany to marshal everything not to mention Hart’s variable form where he has never recaptured the stellar performances of the title winning season.
  • 1st – Chelsea
    • Yes, the ‘Special One’ will win the title on his return to the EPL. Or at least, that is what I think will happen. The man certainly has managerial talent, and clubs appear to get the best out of him in his first couple of seasons with them; but is this to be counted as his first couple of seasons with Chelsea? The squad available is also very strong, and their performance last season puts them into a good place to challenge from the start. There are questions – the defence isn’t as imposing as it was (Luiz is variable, Terry overrated and ageing) and the midfield maestro that is Lampard is on the downward slope of his career, but certainly still capable of driving a team throughout a season. This may well be the last chance for a number of players to command regular starts for a team such as Chelsea, and I expect Mourinho to feed off that to produce a special season that will end with the title.

So there we have the full table based on gut predictions. Not a scientific analysis in this case, and one that I am sure many would disagree with – I claim no particular expertise as a pundit but I am interested to see how well my efforts stack up with the professionals or even the statistics!

Next, and finally on the pre-season predictions, will be a summary predicted table that combines the Pythagorean approach with the gut call to see where there are differences.

Premier League ’13-14 Gut Calls: 11th to 20th

We have had a look at a mathematical model relating to the potential EPL finish, but now it is time to temper the stats with a ‘gut’ assessment. In this post we’ll look at the bottom half of the table including the relegation candidates:

  • 11th – Aston Villa
    • I am expecting an improvement from Villa this season after a difficult outing last time. The younger players have had a chance to bed in and gain some experience, and having managed to retain their primary goal threat they should start reaping the rewards.
  • 12th – Fulham
    • The perennial mid-table team I don’t see any reason for that to change this season!
  • 13th – Southampton
    • A bit of an unknown quantity, Southampton had a mediocre season punctured with some good wins. A poor start didn’t help them, but despite the predictions of doom they certainly had a better second half that helped keep them clear of the main relegation dogfight. Nonetheless, I don’t see the quality to propel them further up the table, although I have a suspicion that they might put something together in one of the cup competition.s
  • 14th – Cardiff City
    • One of the promoted teams usually does well on its entry into the EPL, and Cardiff are well placed to be that team this year. Having the money behind them will help, as well as some experienced players. Cardiff could be in for the long haul.
  • 15th – Newcastle
    • I am concerned about Newcastle. Things really fell apart last season, and the role that Kinnear will play could well lead to more of the same for the Geordies. Their performance last season was woeful at times, and they were lucky not to get involved in a more serious relegation battle.
  • 16th – Sunderland
    • Well at least some in the NE will still like me… Sunderland could be in for a sticky patch, and I am not sure of Di Canio’s temperament if things go against him. He certainly revitalised a team heading for disaster last season, but this time I fear the long haul may be too much. Although I have them 16th, I do predict something of a gap to 15th and very much part of the relegation battle.
  • 17th – Stoke City
    • Stoke have been an EPL constant for many years, but last season’s 13th place flattered a team that seemed to have run out of steam. It appeared that a dramatic change in personnel was needed to inject some energy into the team, and the buys don’t seem to have it. Hughes has a big job on his hands.
  • 18th – Norwich City
    • The fairy tale ends. Although The Canaries will keep it close, probably to the last game, I fear that their luck will run out. That said, the club has made some astute buys in the off-season that may well prove me wrong, but I fear just a couple of injuries (e.g. Bassong) could consign the club back to the Championship.
  • 19th – Hull City
    • Some experience of the EPL will help, but not enough to bridge the gap from the Championship. Despite expressing some interest, the club hasn’t really been strengthened in the off-season; in fact it has lost a couple of solid players that won’t help them make a good start in the new season.
  • 20th – Crystal Palace
    • Although play-off winners can often upset the apple cart, there doesn’t appear to be much chance of this for Palace. The Eagles route into the EPL was fraught to say the least, and the loss of their most potent weapon will be keenly felt. The signing of Chamakh on loan won’t replace the loss of Zaha, and more threat from the forwards would be needed for the team to secure the goals, and therefore the wins, that they would need. Unless, of course, Phillips can continue churning out the goals! At least Holloway will provide some good post-match entertainment!

So there is the gut call for the bottom half of the table. I am sure that many will disagree with the conclusions, and that is part of the joy of the pre-season prediction. Top half to follow, and then the final prediction combining both the stats and the ‘feelings’.