Weekend Review – 29/11/16

Highlights

English Premier League
Chelsea 2-1 Tottenham Hotspur (Saturday 5.30pm)
Our feeling on the biggest game of the weekend was that the market had priced it perfectly, which was no great surprise. However, the lines of -0.75 on Chelsea and 2.75 goals proved to be almost perfect with the Fair Score matching the actual score of 2-1 and justifying our decision to steer clear. Tottenham were handed their first defeat of the season by their bogey team, although the absences of Alderweireld, Lamela and Rose must be acknowledged, and Chelsea finally conceded but further strengthened their claims for the title with another relentless display full of determination.

Conte is moulding the team in his own image and nobody has benefitted more than Moses or Pedro, who were both on the scoresheet here. The right side of the team is functioning beautifully and was able to take full advantage of Spurs’ weakness down that flank. The hosts fought back from an early concession to Eriksen and managed to limit their visitors to nothing better than a Fairly Good chance (~8% conversion) in the game, while at the other end they just about created enough to be worth their two goals, with the wing-backs again getting some notable chances in the box.

Recommendation: N/A
Actual Outcome: 0
Fair Outcome: 0

French Ligue 1
Olympique Lyonnais 1-2 PSG (Sunday 7.45pm)
The marquee fixture in France did not disappoint and pre-match we had suggested that PSG -0.5 looked fair enough on ability, which proved to be the case. However, our real liking was the Over 2.75 on the goals line due to the attacking intent of both sides, which was just about covered thanks to Cavani’s 81st minute winner. On the balance of play there were enough chances to justify a half win here, though PSG were fairly unconvincing once more.

The champions seem to have regressed under Emery and are undoubtedly missing Ibrahimovic’s presence, while Di Maria was also a loss here too, but they were worthy of the win after excelling in the 0 game state but getting sloppy at +1. This has happened more than once this season and marks a sea change from 2015/16, when they would add more goals before easing off. For Lyon it was same old, same old, with their defence not good enough to carry the load in games where they do not create enough to score twice.

Recommendation: Over 2.75
Actual Outcome: +0.5
Fair Outcome: +0.5

German Bundesliga 1
Bayern Munchen 2-1 Bayer Leverkusen (Saturday 5.30pm)
Bayern took a fully deserved 2-1 win here, and such was their overall control of the game in the second half they could even have an argument for winning by two. In the end our preference for the Over 3 instead of Leverkusen +1.5 proved to be a little misguided as we got a push rather than a full win, but the fact that it could quite easily have gone 3-1 meant that we were happy to be out of the handicap trading in-play.

Ultimately both of these teams are hard to trust on recent form, but with Bayern now able to focus on the league until the break then we expect their performances to improve in terms of consistency. Leverkusen will continue to be Leverkusen and though it seems a long way back from here, we still would not rule them out for another top three finish just yet.

Recommendation: Over 3
Actual Outcome: 0
Fair Outcome: 0

Italian Serie A
Internazionale 4-2 Fiorentina (Monday 8pm)
A superb game punctuated by poor refereeing decisions that often went against the visitors was our biggest miss of the weekend, with Fiorentina never looking likely to beat the +0.5 handicap. Inter came out of the blocks firing and were amazingly 3-0 up after just 19 minutes, making a mockery of the 2.5 goal line that we thought had looked a “little low” in the early preview. Thankfully some late team news meant we traded the overs ourselves, but this was definitely one that got away on Friday afternoon. Inter still have issues at the back but the new coach is making a big impact and they could be an interesting outside trade for top three in Serie A at this stage of the campaign.

Recommendation: Fiorentina +0.5
Actual Outcome: -1
Fair Outcome: -1

Dutch Eredivisie
Heerenveen 0-1 Ajax (Sunday 1.30pm)
In the end the market got this goal line spot on, with the Fair Score being somewhere between 0-2 and 0-3 for Ajax. Schaars proved to be a huge miss for Heerenveen, as expected, who never managed to create anything better than a Fairly Good chance. The visitors, on the other hand, had five Great opportunities and many more openings besides, being worth more than their single goal victory. Heerenveen were thankful to goalkeeper Mulder for a virtuoso performance, though Ajax still drew level with Feyenoord at the top after they dropped points in a tricky away game at Utrecht.

Recommendation: Over 2.5
Actual Outcome: -1
Fair Outcome: -1

Ones to Watch

Turkish Super Lig
Besiktas 1-1 Istanbul Basaksehir (Saturday 4pm)
This was the most incorrect line of the weekend and Basaksehir comfortably beat it, the Fair Score giving 1-1 verging on 1-2. They had a goal softly ruled out at the end and were much the better side in the 0 game state, though invited too much pressure when ahead.

Recommendation: Istanbul Basaksehir +1
Actual Outcome: +1
Fair Outcome: +1

Spanish Primera Division
Real Sociedad 1-1 Barcelona (Sunday 7.45pm)
This did indeed prove to be a nightmare away fixture for Barcelona, who were actually a little fortunate to escape with a draw. Sociedad had the better chances and attacked more frequently, doing another big favour to Real Madrid and continuing their hex over the Catalan giants at this venue.

Recommendation: Real Sociedad +1.5
Actual Outcome: +1
Fair Outcome: +1

French Ligue 1
Monaco 4-0 Olympique Marseille (Saturday 4.10pm)
The home side made the absolute most of their shooting opportunities, managing to hit four goals despite only registering one Superb, one Great and one Very Good chance. Marseille did enough to score, with a Fair Score of 3-1 meaning that our suggestion of an Under 3 game was off.

Recommendation: Under 3
Actual Outcome: -1
Fair Outcome: -1

Swiss Super League
Sion 1-2 Basel (Sunday 3pm)
This was one of those unusual games where our Fair Score was the inverse of the actual result, with Basel’s late winner being offside and Sion creating much the better opportunities over the course of the contest. Either way, 2.75 goals did prove to be a touch short in reality.

Recommendation: Over 2.75
Actual Outcome: +0.5
Fair Outcome: +0.5

English Premier League
Southampton 1-0 Everton (Sunday 4.30pm)
The biggest surprise here was that there was just one goal scored, though we can have no complaints with a loss on Everton +0.5. Koeman’s side were second best in every department, and while they deserved a goal it could easily be argued that the hosts deserved two or three.

Recommendation: Everton +0.5
Actual Outcome: -1
Fair Outcome: -1

9 Recommendations
Total Actual Outcome: -1
Total Fair Outcome: -1

Investigating Total Goals Lines (Part I)

While the Asian Handicap market is known as the smart place for traders to operate, there is a significant sway towards Asian Handicap lines rather than Total Goals lines. Natural bias may be one reason for this, as traders will often have a preference towards one team over another, and if the line is lower than what they might expect they use that apparent edge to execute a trade.

In general the Total Goals lines are often seen as harder to call and more open to interventions of “luck”. The lines themselves are usually set between 2.25 and 2.75 in the major European leagues, and while some teams are seen as being prone to being involved in “overs” games or “unders” games, the slightest thing can throw pre-match expectations into disarray.

To give an example, the Queens Park Rangers versus Norwich City game from the 19th of November positioned QPR as +0.25 underdogs on the Asian Handicap pre-match, mainly due to Ian Holloway’s return as manager and Norwich’s poor form, while the Total Goals line was 2.5. At the time this actually seemed slightly higher than expected, because Rangers were also out of form and were playing a “better” team, so may have tried to keep it tight. In addition to this Norwich were notable for scoring many of their goals from Fairly Good and Poor Chances, something that can only continue for so long.

The game began and within 45 seconds Martin Olsson of Norwich had handled the ball on the line, given away a penalty and been sent off. Had the trader known this was going to happen before the game, they would certainly have changed their view of whether to go over or under the Total Goals line.

While unforeseen circumstances such as these are part of the nature of sports trading and will rarely be avoidable, I wanted to use StrataData to look at how Total Goals lines were set, whether there was any pattern to them being Over or Under and how the chance data we collect could be used to determine whether an advantage could be gained. My belief is that even the smallest advantage could provide the edge most traders are looking for.

To begin, I took all of this season’s data up until Thursday 24th November from the “big five” leagues, as well as the English Championship, Portuguese Primeira Liga and Dutch Eredivisie. The first thing to note is the average goals per game so far in the leagues mentioned:

graphic1

This shows Spain and Germany having the most goals per game, with almost 3, while as expected the Portuguese Primeira Liga brings up the rear with only 2.35. The one mildly surprising thing to note is that the Dutch Eredivisie is fifth of the eight leagues investigated, largely because it is generally thought of as one of the more high scoring competitions in Europe.

Looking at the Total Goals lines for each competition throws up some more interesting data, but using the average Total Goals line would not be useful in this situation. Teams like Barcelona, Real Madrid and Bayern Munich all have high Total Goals lines for almost every game they play, which would skew the data quite heavily. Instead I looked at the median and mode to determine which is more useful to use. While the results were similar in most cases (only three leagues had a difference between their median and mode) I chose to focus on the most common line, as this gives a better representation of how the league is thought of by the market in terms of goals:

graphic2

Despite having the highest average goals per game, the most common line in Spain has been a measly 2.25. This shows just how much Barcelona and Real Madrid can skew the average number of goals with their averages of 3.75 and 3.67 per game respectively. I noted my surprise at the Dutch Eredivisie, which was largely because it tends to have a relatively high common Total Goals line of 2.75 on a week-to-week basis. While there is again some skew due to Ajax, Feyenoord and PSV it is still generally thought of as a high scoring league. Interestingly, five of the eight leagues had their most common Total Goals line as just 2.25.

The market will often pitch these lines quite nicely, but just how often do the actual scores beat the Total Goals line? To look into this I have counted any score over the Total Goals line as a full win, so even if there were 3 goals scored in a game with a 2.75 line it just counts as a win (rather than a half). The same applies to losses Under the Total Goals line (so if there were 2 goals in a game with a 2.25 line it just counts as a loss, rather then a half):

graphic3

Only two of the eight competitions are over the Total Goals line more than 50% of the time. This is a clear indication that despite higher average goals than the pre-stated Total Goals lines there is perhaps more value in picking unders rather than overs in general. Teams in French Ligue 1, Dutch Eredivisie and the Portuguese Primeira Liga are only hitting overs around 40% of the time; which is in spite of two of these competitions having their most common Total Goals line at 2.25!

This again shows just how much the top teams can skew the average number of goals in each league. The better teams will always score more goals, but when this happens it will almost always then see an increase in their individual Total Goals line the following week. The inverse can be true for poorer teams at the bottom end, but there is rarely such a significant decrease in the Total Goals lines for those teams. Using the Barcelona and Real Madrid example again, their most common Total Goals lines were 3.75 and 3.5 respectively, which you would think would be reflected in some very low Total Goals lines for a bottom of the table team like Granada, for instance. However, while their most common line is 2.25 this also matches the most common line for Spain, meaning it is not significantly below the mode.

For the second part of this piece I am going to investigate some of the factors that drive Total Goals lines and look into how the granular chance data we collect reflects against these. In case you missed it, the methodology behind how we collect chances was previously covered in great detail here.

I believe it will be interesting to see if games with higher Total Goals lines are reflected in both a higher number of overall chances and also a higher number of better quality chances. Ultimately, by using StrataData I hope to show that we can better judge whether a team is more or less likely to beat the Total Goals Line in any given game.

Dave Willoughby (@donceno)

Weekend Preview – 25/11/16

Highlights

English Premier League
Chelsea vs. Tottenham Hotspur (Saturday 5.30pm)
Chelsea have a fully fit squad for the visit of bitter rivals Tottenham this Saturday, with the home side looking to extend their incredible streak of six successive wins and six successive clean sheets in the league off a handicap of -0.75. The visitors were eliminated from the Champions League on Tuesday night with a 2-1 loss in Monaco, while they are still without Alderweireld and Lamela, plus Rose is suspended. The fact that Conte’s men did not have a midweek game also plays into the home side’s favour, and goes some way to explaining the high looking line.

Tottenham remain the only unbeaten team in the Premier League but performances have taken a notable dip recently, with the 3-2 comeback win over West Ham last weekend not at all convincing. Indeed, our Analyst Fair Score believes Pochettino’s men should have drawn all five of their last games, which does at least show how hard to beat they are. We would be more tempted by Tottenham +1, while we think the 2.75 Total Goals line is pitched about right due to both teams’ intent to win and the firepower on the pitch. Ultimately this one looks to be efficiently priced.

Recommendation: N/A

French Ligue 1
Olympique Lyonnais vs. PSG (Sunday 7.45pm)
Lyon have been far from convincing this season but have begun to turn a corner, being unbeaten in five across all competitions and having relatively good team news. They are only without Jallet and Cornet, the latter being missed more than the former, while visitors PSG have similar availability with just Kurzawa and Rabiot still sidelined through injury. Di Maria and Aurier return in a boost, though Lyon have nobody coming back that can make a big difference.

The hosts have had an extra day to recover since their win over Zagreb in the UCL, but PSG were only in London on Wednesday and had an easy trip home. Taking everything into account, especially pure ability, then PSG at -0.5 looks fair enough, but Over 2.75 total goals looks nice due to the attacking power and general offensive intent of both sides. Lyon will believe they can beat their rivals to reduce the gap to them in the table, while PSG always play to win and do not stop attacking in a +1 game state.

Recommendation: Over 2.75

German Bundesliga 1
Bayern Munchen vs. Bayer Leverkusen (Saturday 5.30pm)
These two are not operating anywhere near the top of their respective powers, which means we have a relatively flat and basic market structure ahead of the contest. Bayern being -1.5 is a standard home line for them against middling-to-good opponents like Leverkusen, while a goal mark of 3 is a natural fit for a handicap of that size.

Ultimately there is not much to knock this out of line, with team news for both the hosts and the visitors relatively uninteresting. For Bayern Neuer should return in a boost alongside Alaba, who was rested for the shock midweek defeat to Rostov, while Leverkusen have Wendell returning to play left-back and otherwise have their usual team ready. The away side have had an extra day since their 1-1 draw with CSKA in Moscow and are almost tempting at such a wide mark, but we find more comfort in the goals.

Recommendation: Over 3

Italian Serie A
Internazionale vs. Fiorentina (Monday 8pm)
Inter have become interesting again since the appointment of new coach Pioli, grabbing a 2-2 draw at Milan last time and being unlucky not to win. They may be without holding midfielder Medel here, which weakens them a bit defensively, but they should provide solid opposition to a Fiorentina side who have not lost since an opening day defeat away to Juventus. Both teams had Europa League commitments on Thursday, with Inter losing in Israel and being eliminated from the competition, while Fiorentina lost at home to PAOK from a commanding position at the top of their group. A goal line of 2.5 looks a little low but quite well judged, though we find it hard to accept Fiorentina being +0.5 underdogs in the circumstances.

Recommendation: Fiorentina +0.5 

Dutch Eredivisie
Heerenveen vs. Ajax (Sunday 1.30pm)
The big team news ahead of this one revolves around home captain Schaars, who remains doubtful and would be a big loss to his side if absent again. Ajax look set to be at full strength following midweek rotation in Europe, with this one shaping up to be an exciting contest between two teams who like to attack. Heerenveen tend to be involved in high scoring games (54% of their total this season have been over the natural line), while Ajax always have the offensive power to find the net multiple times and key striker Dolberg is in superb form.

Recommendation: Over 2.5

Ones to Watch

Turkish Super Lig
Besiktas vs. Istanbul Basaksehir (Saturday 4pm)
Basaksehir still lead the league and have deservedly won their last four matches in a row, including a great 1-2 victory away at Galatasaray. This makes a line of -0.75 to second-placed Besiktas look jarring, especially as the hosts are missing two (possibly three) automatic starters and played in midweek.

Recommendation: Istanbul Basaksehir +1

Spanish Primera Division
Real Sociedad vs. Barcelona (Sunday 7.45pm)
Barcelona made light work of Celtic in midweek and are now only without Iniesta, but this looks like a nightmare away fixture for them. Sociedad have deservedly won four in succession, including a 2-0 home victory over Atletico, and have had the hex over Barca for a number of years.

Recommendation: Real Sociedad +1.5

French Ligue 1
Monaco vs. Olympique Marseille (Saturday 4.10pm)
Monaco have really hit their stride of late and this would have been earmarked as a comfortable home victory just last month, but Marseille have new owners and a new coach and team news is improving every week. They will seek to replicate their performance from the 0-0 in PSG.

Recommendation: Under 3

Swiss Super League
Sion vs. Basel (Sunday 3pm)
Sion have been involved in some hugely entertaining games recently, with 26 goals in their last five in the league. Basel have been their usual selves and have already amassed a commanding lead at the top of the table. Both will play to win, so 2.75 looks a touch short.

Recommendation: Over 2.75

English Premier League
Southampton vs. Everton (Sunday 4.30pm)
Koeman returns to St. Mary’s for the first time in a very interesting clash between two even teams who are in strange form. Both should be fighting for top eight places this season, with Southampton better at the back and Everton exciting up front. It looks impossible to split them.

Recommendation: Everton +0.5

Weekend Review – 22/11/16

Highlights

Spanish Primera Division
Atletico Madrid 0-3 Real Madrid (Saturday 7.45pm)
The biggest game of the weekend saw Real heading to city rivals Atletico on a plus 0.25 Asian Handicap, which they easily beat with a 3-0 win courtesy of a Ronaldo hat-trick. The Ballon d’Or favourite grabbed all of the headlines with his deflected free-kick, penalty and finally counter-attacking treble, with Real’s relatively early goal meaning that the entire game was played to their strengths. As soon as Atletico fell behind it looked a long way back for them, despite Griezmann passing fit in time to feature from the start.

Our recommendation of taking the visitors at the 0.25 mark paid out without too much fuss in the end, as Atletico’s only chance of note came in stoppage time when the game was already over. Indeed, Ronaldo and company are looking imperious at this relatively early stage of the season and with the Fair Score registering this as a 0-2 for Real it means they are now boasting a perfect “Fair” record over the last six outings. Barcelona continue to stutter, while it seems their big rivals have managed to successfully navigate a tricky period of injuries and unavailability, so there could be merit at backing them for the title.

Recommendation: Real Madrid +0.25
Outcome: +1

German Bundesliga 1
Borussia Dortmund 1-0 Bayern Munchen (Saturday 5.30pm)
Dortmund reduced the gap between themselves and their visitors to three points with a completely deserved 1-0 win, which was built on incredible defensive efficiency. Another plus 0.25 recommendation in a big derby game proved successful thanks to Tuchel’s side allowing an enormous 48 key entries but never conceding a chance with a higher than 15% conversion rate. At the other end of the pitch the made the better openings throughout and were almost worth a 2-0 on the Fair Score, which shows just how good a performance this was.

This fixture’s propensity for an unders game also continued, as we had noted pre-match that seven of the last ten meetings between the two had resulted in less than 2.5 goals. This extended that run and was also a good indicator for how the corners market links with inefficiency in attack, as Bayern racked up nine corners from their 48 key entries to Dortmund’s three. Remember that if a team are not scoring but applying a lot of pressure, they are usually generating flag kicks.

Recommendation: Borussia Dortmund +0.25
Outcome: +1

English Premier League
Manchester United 1-1 Arsenal (Saturday 12.30pm)
The hosts were left feeling hard done by after Giroud’s late equaliser off the bench here, with United having edged the contest throughout in terms of momentum. A closer look at the numbers shows that this was actually a much more balanced game than it appeared to be on first sight, however, with Arsenal making the two best opportunities of the contest from significantly fewer key entries.

We had liked the away team on the DNB line pre-match but were a little bit fortunate to escape with a push, because the balance of the side looked wrong from the outset. In future we are going to be watching carefully for a lack of a creative element in Wenger’s engine room and particularly for when El Nenny is paired with Coquelin. The better selection here would have been Ramsey in midfield with Coquelin, Sanchez wide left and Giroud leading the line.

Recommendation: Arsenal DNB
Outcome: 0

Italian Serie A
Milan 2-2 Internazionale (Sunday 7.45pm)
This was another big derby game where the line was incredibly tight pre-match and our suggestion of taking on Milan at DNB looked to be an unlucky push in the circumstances of Inter’s last minute equaliser. However, when evaluating the deeper metrics in the game the case was actually that Inter were fully deserving of the draw, if not the win!

A disclaimer here is that they were forced to be aggressive due to being in a -1 game state for a large portion of the play, but the fact remains that they created twice as many Great Chances (~40% conversion) as Milan and four Very Good Chances (~25%) to the hosts’ zero. Really, anything less than a point would have been an injustice.

Recommendation: Milan DNB
Outcome: 0

Turkish Super Lig
Fenerbahce 2-0 Galatasaray (Sunday 5pm)
Galatasaray have now gone an enormous 18 years without victory in this fixture and they can have few complaints at this result, with Fenerbahce deserved winners. 2-0 may have flattered the home team a little bit, but the fact remains that Gala managed just 11 key entries and a paltry three Poor Chances (~2% conversion) in the entire game. Fener were so strong as a team that they gave next to nothing away, only conceding a solitary corner, while managing to create just enough at the other end to get the three points and the bragging rights for another year.

Recommendation: None
Outcome: 0

Ones to Watch

English Premier League
Southampton 0-0 Liverpool (Saturday 3pm)
The -0.25 on Liverpool looked attractive, but Southampton’s ability to grind out results meant we stopped short of making any kind of recommendation here. In the end the visitors were worthy of a 0-1, but a 0-0 was fair enough.

Recommendation: None
Outcome: 0

German Bundesliga 1
Bayer Leverkusen 2-3 RB Leipzig (Friday 7.30pm)
Our feeling for an over 2.5 game paid off handsomely and deservedly here, but the real story is that Leipzig again confounded the critics with a huge victory. However, it can’t be forgotten that Leverkusen missed a penalty at 2-1.

Recommendation: Over 2.5
Outcome: +1

English Championship
Leeds United 0-2 Newcastle United (Sunday 1.15pm)
A gift from Green helped Newcastle to a relatively comfortably win, with the suggestion to back them at -0.25 a very solid one. Leeds had them worried at 1-0 but the game was not as close as Garry Monk claimed.

Recommendation: Newcastle United -0.25
Outcome: +1

German Bundesliga 2
Union Berlin 1-1 Stuttgart (Sunday 12.30pm)
This was a pulsating game to watch and could have gone either way, though Union took it 2-1 on Fair Score thanks to their better chance creation over the course of the game. In actuality we got a fortunate push.

Recommendation: Stuttgart DNB
Outcome: 0

Austrian Bundesliga
Salzburg 2-1 Rapid Wien (Sunday 3.30pm)
Another fortunate push for our recommendations here, with Rapid Wien pulling a goal back in added time after being very much second best throughout against the title favourites. Team news made little difference as Salzburg dominated en route to victory.

Recommendation: Rapid Wien +1
Outcome: 0

Total Outcome: +4 (8 recommendations)

Weekend Preview – 18/11/16

Highlights

Spanish Primera Division
Atletico Madrid vs. Real Madrid (Saturday 7.45pm)
The first Madrid Derby of the season sees stuttering Atletico hosting Real, with the hosts having hit a rough patch and losing two of their last three. Our Analyst Fair Score believed that they were worthy of a point at Sevilla, but we had no arguments with the 2-0 loss to Sociedad last time out, which makes the fact that you can back Real at 0.25 on the Asian Handicap look enticing. The visitors are still unbeaten this season and have taken 13 points from the last 15, although Fair Score believes they should have a perfect record during that time.

Atletico have been excellent at home, but Griezmann remains a slight doubt, while the away side are now just without Casemiro and Kroos. Ramos and Benzema are back and expected to start, while Modric has achieved full match fitness to enhance the midfield. It’s tough to take a view on the 2.5 Total Goals line in circumstances such as these, though we are completely ignoring the 4.3 per game average between the two sides in the campaign to date. Both have crucial Champions League games on the horizon, but they should concern much for a game of this magnitude.

German Bundesliga 1
Borussia Dortmund vs. Bayern Munich (Saturday 5.30pm)
Another huge clash on the continent sees the German giants meeting in Dortmund, with Bayern currently holding a six point advantage over their big rivals. The home side have no first teamers missing and should edge the motivation, while Ancelotti has Ribery back to full fitness and only Coman on the unavailable list. Team news should not dictate too much here, with primary focus coming down on each teams’ tactical and psychological approach.

Dortmund’s form isn’t great with just six points taken from the last five games, though Bayern haven’t been imperious either with nine in the same time. Both have been hard done by according to Fair Score, however, while both also have winnable UCL clashes on the horizon as well. Overall there is very little to split them aside from form, which makes the home team tempting on a 0.25 handicap in a fixture of this size. 2.5 goals looks low but surprisingly seven of the last ten meetings have ended up Under this mark, which is enough for us to steer clear.

English Premier League
Manchester United vs. Arsenal (Saturday 12.30pm)
One of the most famous clashes of the Premier League era returns, though the gloss has been taken off this fixture recently. For this particular meeting the market has set the handicap at DNB, which almost looks fair until you see the respective form and team news. Mourinho is without Ibrahimovic in addition to Bailly and Smalling, which makes his team look weak. On the other side Wenger is without Bellerin, a big loss on ability and when considering he has completed 47 of the last 49 league matches.

United are undoubtedly under more pressure here and while Arsenal were unconvincing in the 1-1 draw with Spurs they look like a good pick to at least match the handicap. However, Mourinho has had a couple of weeks to prepare, which is always dangerous, while the visitors also have a more pressing upcoming European assignment to be distracted by as well.

Italian Serie A
Milan vs. Internazionale (Sunday 7.45pm)
The fourth highlight of a barnstorming weekend of club football comes from Italy, where a rekindled Milan host the struggling Inter led by a brand new manager in Pioli. At first sight the DNB line is jarring given the recent trajectories of the two clubs, though the fact that Milan look set to be without lynchpin Romagnoli and Inter are at full strength could be playing a part in the price action.

Still, Milan have surprised so far and their form remains very good, while Inter got an expectedly comfortable win over Crotone a week after De Boer’s sacking. They have been consistently very poor on the road, and even though this game is technically in their home stadium we’re still backing the hosts DNB.

Turkish Super Lig
Fenerbahce vs. Galatasaray (Sunday 5pm)
Turkey is our final stop on “derby weekend”, with old rivals Fenerbahce and Galatasaray meeting as the visitors seek a first win at this venue in 17 years. Their atrocious record goes a long way to explaining the -0.75 line, which looks very punchy at first sight, because team news would dictate that Gala are value off that mark. The visitors are full strength but have the weaker recent form, with Fener starting to turn a corner after a slow start and deservedly unbeaten in their last five league games. This is probably just one to sit back and enjoy.

Ones to Watch

English Premier League
Southampton vs. Liverpool (Saturday 3pm)
Long is the only key absentee for Southampton, while Liverpool are almost certainly without Lallana and Coutinho has emerged as a doubt. The price is reflecting the news and Liverpool -0.25 looks attractive, but it may be best left alone.

German Bundesliga 1
Bayer Leverkusen vs. RB Leipzig (Friday 7.30pm)
Leipzig continue to confound expectations and remain unbeaten, but this is a rare occasion where they find themselves as underdogs. Leverkusen have a number of starters returning from injury and look strong, though the over 2.5 looks like the trade.

English Championship
Leeds United vs. Newcastle United (Sunday 1.15pm)
This is the first meeting of the two Northern powerhouses since 2004 and Elland Road is sold out for the first time since 2011. The form of both is great, but at just -0.25 Newcastle seem to offer good value.

German Bundesliga 2
Union Berlin vs. Stuttgart (Sunday 12.30pm)
Ahead of the season we predicted that Stuttgart would finish second with Union Berlin in third, which still looks a fair trade. The visitors are in much the better form and look a good back at DNB in the circumstances.

Austrian Bundesliga
Salzburg vs. Rapid Wien (Sunday 3.30pm)
Salzburg have started slowly despite being heavy favourites for the title, with Rapid weakened over the summer due to transfer business. The visitors have a new manager and the hosts have very bad team news, which makes +1 look attractive.

Who should win the Ballon d’Or? A goalkeeper? (Part 2)

Thanks for joining me for my follow up Ballon d’Or piece!

Last time I focused on the attacking metrics of the game to rank each player, which admittedly left little room to praise or highlight the four goalkeepers in contention for the 2016 edition of the award. As a result of that I am now going to provide an in-depth analysis of these goalkeepers, with the aim of showing that there is more to football than the people putting the ball into the back of the net and making chances.

The last (and only) time a goalkeeper won the Ballon d’Or was way back in 1963, when Lev Yashin lifted the coveted trophy. So who are the four goalkeepers vying to do the same this time around? We have Bayern Munich and Germany’s Manuel Neuer (100/1), Juventus and Italy’s Gianluigi Buffon (66/1), Sporting Lisbon and Portugal’s Rui Patricio (500/1), and finally Tottenham and France’s Hugo Lloris (150/1).

The bookmakers aren’t giving any of them much chance at all, but can anything in our metrics show that they are perhaps more deserving of further recognition than the market thinks? Also, just which of the four goalkeepers deserves to have the shortest odds based on his output in 2016? I will uncover this and much, much more, but let’s start with the basics:

Minutes Played

Like last time, let’s level the playing field by measuring how many minutes each goalkeeper has played. We’ll use this to normalise each player’s metric set so that we’re making sure we’re being fair to all four nominees.

Handily StrataData covers each player’s domestic league (English Premier League, German Bundesliga 1, Italian Serie A and Portuguese Primeira Liga) as well as the Champions League, Europa League and Euro 2016. So here’s the amount of time each player has spent on the pitch across those competitions in 2016:

graphic1_minsplayed_v3

Neuer has spent most time on the pitch with 3651 minutes under his belt for Germany and Bayern Munich. Patricio, Lloris and Buffon, with 3629, 3545 and 3376 minutes respectively, follow him.

Unadjusted Metrics

Now let’s take a look at key goalkeeper metrics in the form of actual saves made versus expected saves, with the ratio of these two metrics indicating overall performance. My expected saves in this instance are simply calculated by totalling the actual number of saves made and the goals conceded before adjusting for minutes played.

Remember that we’re focusing only on saves made here, which basically boils down to shots on target that required goalkeeping intervention to stop a goal (as opposed to shots that were blocked or hit the woodwork.) These metrics are outlined in the graphic below:

graphic2_unweighted_v3

The blue bars represent actual saves made per 90 by each goalkeeper, while the green bars represent expected saves made per 90, as a function of saves made and goals conceded. To highlight goalkeeper performance, I’ve also plotted the ratio between these two metrics in yellow. I’ll be using this ratio as the key performance indicator for the goalkeepers. I’ve calculated this by dividing the saves made by the expected number of saves made to give us a ratio from 0-1; with a larger number indicating better performance (i.e. that the goalkeepers are saving as many shots that they are expected to save), whereas a smaller number (closer to 0) indicates that the goalkeeper is not saving as many shots as he is expected to save.

Neuer again tops the rankings for saves made, with 1.73 saves per 90 minutes, followed by Lloris, Buffon and Euro 2016 winner Patricio. Neuer also appears to be performing the closest to what is expected from him, with the largest ratio (0.74) between what he actually saves and what he is expected to save. In this light, Patricio is the worst performer, with a ratio of 0.59 saves per 90 below expectation.

There is a nice spread between all four goalkeepers, clearly showing that at the most basic level Neuer appears to be the most reliable goalkeeper, closely followed by Hugo Lloris. However, to go a bit deeper let’s now consider the type of chance the four goalkeepers are saving.

In the following analyses, a weight is given to each type of save a goalkeeper makes to help us differentiate between routine saves and exceptional, match winning saves. Weighted adjustment is a commonly applied technique and in simple terms it allows us to assign an adjustment weight to the metric of our choice (in this case, saves and expected saves) to get a richer picture of goalkeeper performance.

Saves (Adjusted for Chance Type)

In a similar way to how I evaluated the attacking players in my previous post, I’ve calculated the expected number of saves per 90 as the number of actual saves on the number of chances (i.e. shots) faced (both saved and conceded).

Combining chances saved with those that are not saved allows us to work out what a goalkeeper is expected to stop given the overall number of chances they face. I have worked this out using the average conversion rates of our six chance categories, labelled Superb (~75% conversion), Great (~40% conversion), Very Good (~25% conversion), Good (~15% conversion), Fairly Good (~8% conversion) and Poor (~2% conversion), summing a goalkeeper’s total chances faced per 90 by total conversion rate rather than by individual category.

This has allowed me to weight each goalkeeper’s saves and goals conceded by the quality of chance to provide a firm expected save number. This is an important metric to highlight the ability of our nominees to keep the ball out of their own net, though it is key to note that it assumes that every chance on target a goalkeeper faces can be saved. Once again, for ease of interpretation, I’ve plotted actual saves and expected saves, as well as the ratio of the two as the key performance indicator:

graphic3_weighted_chancerating_v3

After adjusting for chance type, Neuer still tops the list for saves made (0.34 per 90), but this time has the second best save ratio of the group (0.52). Indeed, it seems that Lloris has been the most efficient goalkeeper in 2016 so far, just edging Neuer with the greatest ratio between number of saves made and the expected number of saves (0.53). Patricio is again the worst performing goalkeeper with a ratio of 0.41, indicating that he is not saving as many shots as he might be expected to and probably should not be in this discussion in the first place.

Saves (Adjusted for Chance Type and Shot Quality)

The depth of StrataData means that I can also take this analysis further and use a multi-element weighting method. To provide a much richer picture of all four goalkeepers, I’ve added the Shot Quality rating as captured by our analysts. This rating ranges from 0-5, where 0 illustrates when a shot was unintentional and 5 indicates that the shot was nigh on unstoppable. Therefore, a chance with a rating of Superb and a shot quality of 5 wouldn’t be expected to be saved, whereas a chance with a rating of Poor and a shot quality of say 2, for instance, would be expected to be saved.

Creating and applying a weighting based on chance rating and shot quality provides a much deeper picture of goalkeeping performance. I calculated the relative weighting for each shot quality and chance rating using all goals and chances conceded and faced, and applied these weights to our four goalkeeper’s saves and expected saves. This adjusted data can be seen in the graphic below:

graphic4_weighted_chancerating_sq_v3

Once again, Neuer tops the list for saves made, even after adjusting for chance rating and shot quality, with 0.4 saves made per 90, closely followed by Lloris with 0.39, Patricio with 0.25 and Buffon with 0.23. Despite that fact Buffon has the lowest save rate amongst the goalkeepers, what I find most interesting here is that he has the best expected save rate (0.76 saves per 90), giving a better save ratio (0.30) than Patricio (0.29), who again drops down being the worst performer of the group.

In terms of the best performing goalkeeper in this category, we see that Lloris still stands at the top with his save-per-chance ratio of 0.41, though Neuer is again right behind him and is ultimately the “busier” of the two.

At this point in the piece it seems as though if there were a goalkeeper seriously in contention for the 2016 Ballon d’Or then it would be a straight shootout between Germany’s number one and France’s number one, with Patricio not in the same class and Buffon falling away badly when efficiency is brought into the equation.

Still, let’s round things off by looking at something not always capture in mainstream statistics – individual mistakes made leading to chances and/or goals.

Mistakes

You may remember we looked at red and yellow cards for outfield players in the previous blog, so this time I decided to look at mistakes made by our four goalkeepers. Mistakes are an unwanted aspect of any player’s game and can be useful to look at as a final key differentiator when deciding between two players for scouting purposes, or if hypothetically discussing if a goalkeeper might win the Ballon d’Or…

With that in mind I counted the number of mistakes attributed to each player from the 1st of January to the 31st of October 2016 and once again normalised for minutes played:

Graphic5_Mistakes_v3.png

Although mistakes are not at all common for goalkeepers of this quality, a few still appear and have to be taken as a red mark against them. Lloris undoes some of his prior good work by topping the list, making 0.20 mistakes a game that lead directly to chances or goals. Neuer and Patricio aren’t too far behind, with 0.17 and 0.15 mistakes a game respectively, but Buffon recovers somewhat to show his enduring class here, with a mere 0.05 mistakes per 90.

Ultimately this might be enough to tilt some people in the direction of Neuer, but again the mistakes are so few between he and Lloris that it really becomes a game of fine margins.

Conclusions

I hope that I have shown how calculating expected saves alongside saves made and then adjusting for key performance metrics is important to see the true value of a goalkeeper. We saw in the first analysis that Neuer appeared to be comfortably the “best” goalkeeper based on his number of saves per 90 and the overall save ratio for each goalkeeper.

However, to provide a richer picture of the data we adjusted the goalkeeper data to account for the quality of chance he faces and the quality of the shot he faces. After carrying out the first adjustment Lloris stood out above the rest and was once again at the top when shot quality was taken into account, though Neuer was right behind him in both instances. The only red mark against Tottenham’s captain were the elevated number of mistakes he has made versus Neuer in 2016, though the difference really was negligible once again.

Indeed, if a goalkeeper were to be in the running for the 2016 Ballon d’Or then the evidence would point towards the Frenchman being the “value” bet at 150/1 versus Neuer’s 100/1. Still, Buffon is the surprising favourite amongst this quartet at 66/1, while Patricio rightly does not seem to feature in the discussion at all at 500/1! If it were me choosing then Lloris would be the man and I’m pleased that the numbers I have uncovered have shown him as a peer to Neuer, who is already widely accepted as being a great in this era of the game.

Until next time,

Sagar Jilka (@DrSagarJilka)

PS: To more easily compare the players evaluated in detail in my last two posts I will soon be releasing some radar plots showing their performances in each key metric category, so please keep your eye out for those! Credit for the inspiration behind them must go to www.ramimo.com for their NBA originals in 2013 and a post on www.statsbomb.com from Ted Knutson (@MixedKnuts) uncovering these back in April of this year.

Are Chelsea and Liverpool true title contenders?

Having moved beyond the opening quarter of the season Liverpool sit atop the Premier League table, closely followed by Chelsea. With both playing an exciting brand of football that has seen them comfortably outscore opponents at times, I wanted to take a look to see if the two clubs are going to be involved in a straight shoot-out for the title, or if they are simply enjoying a good patch of form before their rivals eventually overhaul them.

At the start of the campaign, Liverpool were at 9/1 to win the Premier League, putting them behind Tottenham (8/1), Chelsea (13/2), Arsenal (6/1), Man Utd (5/1) and favourites Man City (9/4). However, given their strong start they are now seen as second favourites to Man City, who remain well backed at 2/1. Liverpool’s odds have shortened by almost half to 5/2, with the market reacting to their performances as might be expected. Comparatively, Chelsea have also seen their odds of lifting the trophy coming in to 3/1, marking quite a substantial shift when considering that both Man Utd and Tottenham have moved out to 20/1.

But is there any data to suggest that either Liverpool or Chelsea will be at the sharp end of the table come the end of the season? And are there any factors that could work in their favour?

One thing that immediately jumps off the page when looking at the league table is just how effective both teams are in front of goal. Liverpool are averaging just shy of three goals per game, having netted 30 in their opening 11 matches, while Chelsea are a little further back, but have still found the net more than any of the other competitors (26 times in all). However, is this high rate of scoring down to poor opposition? Or a ruthless attacking unit that is being unusually clinical when chances arrive?

In Liverpool’s case it would be fairly easy to point to a relatively soft schedule, with high scoring wins over Hull, Watford and Crystal Palace possibly not major reasons to get excited. Scoring four times against both Arsenal and Leicester, though the latter are admittedly less defensively sound than they were last season, does point to a very potent Liverpool attack. What is notable about their scoring is that the goals aren’t coming from one single source. Indeed, so far 10 players have found the net in the league, though perhaps unsurprisingly most of the net finding has been done by Saido Mane, Philippe Coutinho and Roberto Firmino, who have 16 goals between them in what has proven to be a very effective front trio.

In contrast, Chelsea may have scored a few goals less, but their opponents have been a little more reputable. Everton and Southampton both came into their games with Chelsea boasting impressive defensive records, but neither were able to shut them out and ultimately conceded seven goals between them, though Everton’s heavy 5-0 defeat does skew that statistic somewhat! Against top opposition, Chelsea have been rather mixed in their results to date, as an impressive 4-0 win over Man Utd was preceded by a 3-0 loss to Arsenal and a 2-1 home defeat to Liverpool. While Chelsea themselves have also seen nine players find the back of the net, their goals are far less evenly distributed, with Eden Hazard and Diego Costa responsible for 16 of the 26, suggesting that should either pick up an injury for any length of time, the team would find it a little harder to cope, whereas Liverpool might be more covered in such a circumstance.

What this does suggest is that both teams are doing a good job of creating and converting chances in the first place, something that is backed up by StrataData, which sees Chelsea and Liverpool in the top three in terms of chances created. However, what is interesting is that under half of the goals scored by either is viewed as a Great Chance, with both teams sitting toward the middle of the pile on that statistic, suggesting that they are in fact able to still find the net without opening up really promising opportunities. If compared to Man City, who are the most potent team from Great Chances, it leads to two possible explanations. Either Man City are doing a good job of converting when they should be, or Chelsea and Liverpool are under-performing with their great chances and their current rate of scoring from other means will be that much harder to sustain. Either way you choose to interpret it, the fact that they are creating so often in the first place is very encouraging and usually much more repeatable:

alecblog_151116

Some further reasons for optimism if you are considering Liverpool and/or Chelsea as potential title winners are that both have found a settled and effective playing style and a formation that suits them. Liverpool have stuck with their high energy 4-3-3 shape since the start of the season, while Chelsea have changed to 3-4-3 from 4-1-4-1 in the last few weeks, and the difference has been astounding. Since the change of shape they have won five straight league games without conceding a single goal. The new system is one Antonio Conte is comfortable with from his days coaching both Juventus and the Italian national team, and having seen the potential to make the transition at Chelsea they are certainly reaping the benefits. They may not have the same relentless attacking outlook of Liverpool, but they have much better balance and the incredibly high work rate demanded (and often shown by the coach himself on the sidelines!) is working wonders for team spirit and morale.

Comparing this to Manchester City once again, Pep Guardiola has so far struggled to settle on a system that he is happy with and provides the best solution to get the most out of the plethora of attacking options at his disposal. So far he has trialled 4-2-3-1, 4-1-4-1 and 3-4-2-1, each without ever truly being convinced. The uncertainty over the system the coach wants his players to operate in hasn’t had a major impact on results thus far, as their attacking capabilities often pull them through, but in recent games where changes in shape have become more frequent, they have fared a little worse, failing to win any of their last three home league encounters. City may arguably have the higher quality squad with greater depth (a topic for another day), but if Guardiola cannot find a way to get the best out of them, he risks leaving the door open for both Chelsea and Liverpool, who are prospering with a settled and effective style.

So Chelsea and Liverpool are both creating chances at a similar level and enjoying success brought about from a settled and effective style of play, but are there more factors that could prove beneficial to their respective title challenges in the long run?

One thing that most certainly shouldn’t be overlooked is that neither club have any European commitments this season. We need only look at Leicester last year as example that not having the additional burden of midweek games on the continent can in fact be a blessing, with rivals needing to manage and rotate their squad to compete effectively on multiple fronts. Chelsea and Liverpool are both able to name their strongest team week in and week out, having had a complete break between games and having spent plenty of time on the training ground to work on approach play and shape. If either Chelsea or Liverpool do go on to lift the trophy this season, the argument could be made that those clubs battling for the final European spots may actually be better suited to finishing outside of those positions, relieving the burden to give themselves a better chance of success in the next campaign. That period of rest and preparation time from one game to the next is proving to be invaluable as the schedule for rivals becomes increasingly congested, particularly during the Group Stages of both the Champions League and Europa League when there are also early rounds of domestic cup competitions to consider as well.

Everything seems to be going in favour of the two clubs at present, but is there anything that could stop them in their tracks? And who of the pairing is best placed to mount the more serious challenge?

For Liverpool, their Achilles heel this season has been their defensive output, with their attacking philosophy winning many admirers, but also seeing them left very exposed in transition. Even though they are scoring at a rate of nearly three a game, they are almost required to, as they are shipping over a goal a game at the other end. Now, the approach to simply outscore opponents by playing offensively is undoubtedly effective when it works, but what happens on the days where their attacking play is stifled and ineffective? Burnley highlighted this flaw superbly at the start of the season when they pulled off a shock 2-0 win over Klopp’s men. Liverpool dominated territory for almost that entire game, but up against a defensive team with something to hold onto they struggled badly and over-committed offensively to concede a second on the counter. Even Middlesbrough, down in 15th place, have conceded fewer than the 14 goals allowed by Liverpool this season, raising questions over whether their approach is truly sustainable.

Chelsea on the other hand have adopted an approach based around being defensively organised, boasting the second best defensive record in the league, while also maintaining a very impressive attacking output. In the five games since moving to a back three they are yet to concede, while their goals per game ratio has actually gone up, even against tougher opponents. In a more reserved 4-1-4-1 system they were averaging 1.8 a game, but the change to 3-4-3 has yielded 3.2! The fact that they have managed to increase output and decrease the number of goals conceded is a massive point in Chelsea’s favour, as though the high rate of scoring may well come down as the season progresses, maintaining a healthy defensive record will see them standing a far better chance of sustaining a title push. As alluded to before the big concern for Chelsea will be keeping their key players fit, particularly Costa and Hazard, who have been rejuvenated this season under the tutelage of Conte. Still, Costa’s availability may have less to do with fitness and more to do with discipline at times, with the Spaniard renowned for walking a tightrope with officials and only slightly calmer this season compared to last.

Even the imperious looking Man City attack cannot compete with the numbers being put up by Chelsea and Liverpool, as although they made a very brisk start to the season, they have arguably picked up results that would be expected of them when beating Stoke, Bournemouth and West Brom by scoring four goals, and West Ham and Swansea by scoring three. A glancing comparison shows that Chelsea defeated Southampton and Everton without conceding, while Man City laboured to a pair of 1-1 draws at home to the same opponents. The fact that Man City faced Southampton just days after playing Barcelona in the Champions League cannot even be used as a case in their defence, as Southampton themselves had even less preparation time after a difficult trip to Inter Milan in the Europa League. It’s key to note here that Liverpool are yet to face either opponent, so a direct comparison cannot be made.

alecblog_151116_2

The table above tells us that even though they may be less potent than Liverpool, Chelsea are actually increasing their goal difference by a slightly greater margin every game, largely through having a more organised defensive set up. Man City on the other hand actually lag behind both, with their attacking threat having diminished a little of late.

In terms of the upcoming schedules for the three sides, Liverpool will certainly feel they can continue their recent purple patch of form over the next five games, facing Southampton, Sunderland, Bournemouth, West Ham and Middlesbrough before the Merseyside derby with Everton the week before Christmas. Chelsea and Man City actually meet each other in three weeks time, and have a slightly tougher run of games, with the outcome of that match likely to have some bearing on how both are viewed as title contenders moving forward. If Chelsea can come out on top at The Etihad, they will almost certainly consider themselves as one of the favourites for the title and the market should well agree. Even if they were to lose, it would be unlikely have a major impact on their prospects and could indeed make them a more attractive option in the long term.

One thing that is for sure is that both Chelsea and Liverpool are clubs with genuine title ambitions because they have each found real balance and effectiveness in their performances. Chelsea’s better defensive organisation could see them a little better placed to be the real contenders to a fairly neurotic Man City than the goalscoring juggernaut that is Liverpool, but in a league where unpredictability is king, nothing should be discounted. At this point in time buying in on Liverpool looks the best approach to take, once it has been confirmed that all of their internationals have returned from duty unscathed, of course!

Alec Payne (@Payney3)

Who should win the Ballon d’Or? (Part 1)

France Football recently unveiled their 30-man shortlist for the prestigious Ballon d’Or award. The Ballon d’Or is given to world football’s best men’s player, as voted for by football managers and journalists alike.

Now I’m sure you’ll already have an opinion on who you think should win it, and my colleague Dave Willoughby’s recent blog may have influenced your decision a little, but I’d like to extend that by giving you a completely statistical perspective on the nominees. I’m going to be using StrataBet’s unique data to compare all 30 shortlisted nominees across a variety of metrics, including goals scored, primary assists and secondary assists, while looking at their expected values for each of these things as well.

Before we get into that, let’s start with levelling the playing field by measuring how many minutes each player has played. We’ll use this data to normalise each metric so that we’re making sure we’re being fair to all 30 nominees.

StrataData handily covers the domestic league competitions of all of the nominees (English Premier League, German Bundesliga 1, Spanish La Liga, Italian Seria A and the Portuguese Primeira Liga), as well as the UEFA Champions League and Europa League. So here’s the amount of time each player has spent on the pitch in 2016 so far:

minutesplayed

Toni Kroos, Antoine Griezmann and Manuel Neuer have had a busy year, playing 3732, 3679 and 3651 minutes respectively, while the current Ballon d’Or holder Messi is 17th on the list, having only played 3050 minutes in a somewhat injury hit year.

Goals

Now that we know how long each nominee has spent on the pitch, let’s use goals as the initial metric to rank and compare them.

To understand the performance of a player, I’ve calculated the expected number of goals per 90. This is a fairly common metric these days but my number contains a fair degree of inimitability, as StrataData allows us to calculate expected goals on the actual number of goals scored in addition to the number of chances each player has had.

At the most elementary level it is key to understand that every goal is a chance, but not every chance is a goal.

Combining chances not scored with those that are scored allows us to work out what a player is expected to score given the overall number of chances they get. I have worked this out using the average conversion rates of our six chance categories, labelled Superb (~75% conversion), Great (~40% conversion), Very Good (~25% conversion), Good (~15% conversion), Fairly Good (~8% conversion) and Poor (~2% conversion), summing a player’s total chances per 90 by conversion rate rather than by category.

This has allowed me to weight each player’s chances and goals by the quality of chance to provide a firm expected goals number. This is an important metric to uncover the finishing efficiency of our nominees and is easy to recognise when plotted alongside actual goals scored.

As such, the following graphic outlines the (normalised) number of goals scored per minutes played, with the green bars representing the number of goals per 90 and the yellow bar the expected number of goals per 90 for each player:

Goals_ExpGoals_per90.png

What strikes me here first and foremost is just how effective Lionel Messi remains, which I appreciate is news to nobody. Considering he was 14th in the minutes played ranking, he jumps to number one for goals scored per 90 and is followed by Barcelona teammate Luis Suarez. Gonzalo Higuain is third with just under a goal per 90, while Cristiano Ronaldo, despite playing the fifth most number of minutes out of the nominees, is down in fourth after normalising his goals scored by minutes played.

The relationship between the green (goals) and yellow bars (expected goals based on the chances they have) is also worth noting. For the majority of the nominees, the green bars are taller than the yellow bars. This highlights the attacking and finishing prowess these players have on the pitch, as they are essentially over-performing to a point where they are converting more than would be expected. This is not true for all of the nominees however, with Pierre-Emerick Aubameyang, Neymar, Jamie Vardy, Thomas Muller and Zlatan Ibrahimovic scoring fewer goals than the chances presented to them should have produced. Aubameyang and Ibrahimovic are the biggest culprits here, but keep getting into the positions to score and there is still plenty to be said for that.

Primary Assists

Let’s look at primary assists next, and for this we are investigating the overall ability of the nominees to provide scoring opportunities for their teammates. This is captured by assists (if their teammates score) and chances created (if they don’t).

For clarity, it’s important to note here that while assists are taken from goals only, expected assists are taken from both goals and chances. Ultimately we want to ensure that we are rewarding the nominee for his creative ability and not punishing him for a teammates’ poor finishing, and with that in mind I have sorted the data according to expected assists rather than actual assists.

Remember that StrataData captures chances in six categories, labelled Superb (~75% conversion), Great (~40% conversion), Very Good (~25% conversion), Good (~15% conversion), Fairly Good (~8% conversion) and Poor (~2% conversion). I will be using the sum of chances created by their average conversion rates to rank the 30 nominees, again normalising for minutes played:

Assists_ExpAssists_per90.png

Lionel Messi again tops the ranking for expected assists and is again assisting more than he “should be”, with Barcelona teammate Neymar joining him to make up the top two. West Ham’s Dimitri Payet and Juventus’ Paulo Dybala deserve great praise here, as they are 21st and 27th out of 30 for minutes played but third and sixth on the list for assists, respectively.

It is also worth noting that the top three players for assists are Barcelona’s fearsome strikeforce of Messi, Suarez and Neymar. This trio also convert the chances created by one another more than they miss them, which in turn helps their teammates to perform well in this ranking category.

This section is probably where you would expect to see attacking midfielders coming to the fore, and this is highlighted by the fall of out-and-out strikers such as Sergio Agüero, Robert Lewandowski and Higuaín, who drop down the ranking for expected assists. Unsurprisingly midfielders such as Kevin De Bruyne, Paul Pogba and Kroos replace them near the top, demonstrating how important it is to consider metrics beyond the scoring of chances to properly evaluate a player’s importance to his team.

Secondary Assists

With this in mind, another unique facet of StrataData is the capturing of secondary assists, a key pass (or action) that occurs directly before the assist. This plays an important part in evaluating an effective attack, as it allows us to capture the build-up to a goal or chance. Secondary assists are quite rare but do provide a nice level of insight into those players who play “the pass before the pass” and I have used this metric to rank the 30 nominees once more, again adjusting for minutes played:

secondaryassists_expsecondaryassists_per90

De Bruyne leads the secondary assists ranking, which highlights his importance to the strength of Manchester City’s attack, being the creative force early in the earlier moments of chance creation. Next in line for secondary assists are Neymar, Gareth Bale and Messi. The Barcelona effect could be at play again here, but it’s worthy of note that Suarez hasn’t yet registered a secondary assist in this calendar year.

Please let me take a moment to give some appreciation to Hugo Lloris now too, as he is the only goalkeeper to register a secondary assist (or to appear on any of the graphs to date for that matter!)

Discipline

At Stratagem we also believe that a sound disciplinary record should be taken into account when assessing the potential Ballon d’Or winner, given their status as role models and ambassadors of the beautiful game. As such, here I’ve simply counted the number of yellow and red cards each player has received during the calendar year so far (across the competitions outlined above) and again adjusted for minutes played. Here is the ranking:

cards

Ah, Sergio Ramos. Could it be anybody else? Real Madrid’s centre-back tops the list for both yellow and red cards per 90, with Arturo Vidal and Luis Suarez in second and third place. Suarez was building himself a very solid case for consideration to this point, but it appears his indiscipline on the field is a still a factor that will count against the Uruguayan when it comes to awards of this nature. Finally, current Ballon d’Or holder Messi is ninth on the list with previous winner Ronaldo in 17th, which is the first real edge the Portuguese has had over the Argentinian so far.

Initial Conclusions

Although I have endeavoured to analyse data beyond goals scored and assists made, it’s clear to see that there is a core group of players who sit around the top of each key category. This core group includes the Barcelona trio of Messi, Suarez and Neymar, whose attacking potency clearly allows them to rack up the numbers compared to players from less dominant teams. These players are clearly performing at an incredibly high level in a team with less exceptional teammates, which is arguably more impressive.

For these reasons Messi, Ronaldo, Suarez, Neymar and Payet make up the players who have most impressed me from this initial analysis. Has anyone else stood out for you? If so, please let me know by contacting me on Twitter (username below!)

Finally, given that we’ve talked about positional play and impact, I do think it’s unfair to focus only on attacking prowess as I seek to answer the question of who should win the Ballon d’Or, especially when we have four goalkeepers in the fold. For that reason part two of this series will focus primarily on Gianluigi Buffon, Manuel Neuer, Rui Patricio and Hugo Lloris. Let’s see if any of them can make a jump onto my five-man shortlist.

Until next time!

Sagar Jilka (@DrSagarJilka)

PS: If you’re wondering what game Lloris registered his secondary assist in, it was Tottenham’s 3-0 home win over Bournemouth on the 20th March.

Who is the most inefficient attacker?

With the Ballon d’Or shortlist just published the debate around the best players in the world has surfaced once again. Lionel Messi and Cristiano Ronaldo are expected to continue their domination of the award and you would get long odds on anybody outside of those two winning it for the first time since 2007, when Kaka lifted the trophy.

My colleague Sagar Jilka is following me with an in-depth series looking at the underlying metrics of the players on the Ballon d’Or shortlist later this week, but before that I just wanted to have a bit of fun and look at who the most inefficient attacker in football is (well, football covered by StrataBet’s 22 competitions at least.)

I had no preconceived ideas about who this might be before diving into the data, though the social media derision aimed at Edinson Cavani of PSG gave me inkling over the type of player I might expect to see! A high profile miss against Arsenal in the Champions League and an even worse one in the Ligue 1 grudge match with Marseille do little to fight the Uruguayan’s cause, and I was keen to see where he ranked on the list of the most inefficient attackers in football.

To begin with I just wanted to see which attackers had missed the most Great Chances (~40% conversion on average) or Superb Chances (~75% conversion on average) in all games between 1st July 2016 and 31st October 2016. During this period a total of 2,617 Great Chances or Superb Chances were missed, while 2,322 were scored. This gives an overall conversion rate across both chance types of ~47%, which is about right given the numbers of each.

Indeed, only 128 of the 2,617 missed chances were Superb and of these only five players across all leagues missed more than one (all missing two each). Unsurprisingly, given what we know about the conversion rates of the league in general, two of the players are in the MLS (Jozy Altidore and Juan Agudelo) and both have spent unsuccessful times at Premier League clubs:

graphic1_v4_figure

Of the attackers that missed Greats one man stood out on his own in front of the pack and that man was our old friend Edinson Cavani. This means that for once it appears that the eye test is backed up by the data. He has missed 15 Great Chances in the 1,135 minutes he has played in Ligue 1 and the Champions League this season. That averages out as a Great Chance missed every 75 minutes, so it’s no wonder he looks like he misses a lot…

graphic2_v4_figure

There are some other notable players on the list, however, Ballon d’Or contender Cristiano Ronaldo and Bayern Munich’s star striker Robert Lewandowski to name just a couple! This shows that everybody misses chances and that simply getting into positions to have them is also a real skill – although missing near enough every chance will naturally not see the player in question start a lot of games.

To continue I also wanted to look at a few other things thrown up by the data, which brings me on to conversion rates:

graphic3_v4_figure

As you might expect, the higher profile attackers have now disappeared from the list, leaving just those players who continuously miss very presentable opportunities. It is again no surprise to see that two MLS players head the graphic. I must make special mention of Will Bruin of Houston Dynamo, who has missed all seven Great Chances to fall his way and is one of the reasons that they failed to make the play-offs.

However, let’s leave poor Will alone for a moment and look at the actual conversion rates of the higher profile attackers to see how much difference there is:

graphic4_v3

While Cavani, Lewandowski and Ronaldo have relatively similar conversion rates Messi continues to show he is from another planet by posting a ridiculous 100%, albeit off fewer chances. Interestingly on Cavani this table shows that he has more Great and Superb Chances than Lewandowski and Ronaldo combined, which naturally owes to PSG’s strength in Ligue 1 and the way they are set up to play to serve the lone central striker. However, it also highlights why he has become so known for missing opportunities.

To expand the investigation I also wanted to have a look at which teams and leagues were more notorious for missing high quality chances:

graphic5

Swiss Super League side Sion are out in front as the side with the most Great or Superb chances missed per game, averaging an enormous 2.23. Although currently 12 points behind Basel and in second place, if they can bring that rate down and finish just one of those chances on average per game they may at least make the title race more interesting.

Real Madrid and PSG are the other teams of note to feature here, but it is key to state that all the sides missing the highest number of Great or Superb Chances per game are actually doing quite well. Indeed, all of them sit in the top three in their respective leagues, which shows just how powerful chance creation metrics are in terms of predicting future performance.

As Sion and Young Boys both feature on the above graphic it is no surprise to see that the Swiss Super League is the StrataBet competition that has the most Great or Superb Chances missed per game (0.60 on average). The English Premier League has the lowest at 0.34, which breaks down to roughly an average of one every three games being missed. With just less than one being created per game this is comparatively low, with only the Swedish Allsvenskan seeing fewer clear chances on a per game basis.

graphic6

To conclude, I feel that a proviso is needed to confirm that this is just a bit of fun and I am in no way saying the players and teams mentioned within this piece are “bad”. All the data indicates is that they have missed some very presentable chances over the last four months, but as the saying goes “you have to be in it to win it.” As conversion rates are fairly uniform across all leagues, players who have missed a higher-than-normal number of Great or Superb Chances could well revert to the mean and go on a hot streak if the opportunities keep coming their way.

I remember that people used to talk negatively about Andy Cole, saying that he needed five chances to score a goal, but if a player is still scoring 30 a season that’s probably a heck of a lot of chances he’s getting! Of course at the time those numbers were plucked from thin air, but these days using StrataData we can easily see which players have the most high quality chances and the most frequent chances.

This leads me to a reminder to finish off…

The offer remains for any budding data scientists out there who are frustrated with a lack of football data access. We are looking to expand upon the partnerships we have with @AnalyticsFC and @zorba138 in order to expose more people to our unique StrataData offering.

We can offer full API access in return for regular written content using elements of the data itself, which can be posted on external websites or on any form of social media. If you are unable to interrogate our database using Python, R or are unfamiliar with MongoDB then we would also be open to providing subsets of league, team and player data in Excel spreadsheets or any other desirable format.

So if you would be interested in working with us, then please get in touch with me on Twitter (username below) or via davew@stratagem.co.

Thanks for reading!

Dave Willoughby (@donceno)

Weekend Review – 08/11/16

Highlights

English Premier League
Arsenal 1-1 Tottenham Hotspur (Sunday 12pm)
This North London derby swung back and forth nicely in terms of momentum but was most notable for the excellent tactical approach adopted by visiting coach Pochettino. The Argentinian showed bravery to switch to a three-man defensive line, bringing in Wimmer alongside Dier and Vertonghen, while releasing his full-backs to play much more attacking roles. The rest of the side had a familiar look to it, with Dembele continuing to provide a talismanic presence in central midfield. The Belgian had an excellent game, but there were not too many a white shirt who didn’t.

Arsenal were almost at full strength but lacked rhythm for much of the contest, with the omission of Giroud raising some eyebrows after his last display from the bench in the league. From a trading perspective it was quite a difficult one to work in terms of Asian Handicaps and Total Goals due to the tightness and overall lack of chances, but a dearth of corners in the first half an hour meant that an opportunity opened up there following the first goal. The lesson here is that there is always something to trade on, even if you can’t find any joy on the biggest markets.

Spanish Primera Division
Sevilla 1-2 Barcelona (Sunday 7.45pm)
Sevilla were unable to repeat their trick of beating a title contender at their own game, although they played a full part in a very entertaining contest here. It was our belief that it would be tight and likely to be decided by one goal, which meant that backing Sevilla +1 returned stake, while our liking for the under 3.5 line actually paid off despite there being enough chances in the game for more than three goals.

The pattern of the contest was set due to Vitolo’s early opener, which meant that Barcelona had no choice but to throw caution to the wind at some stage. Without Pique and Alba they were always liable to concede chances at the other end, but a piece of finishing brilliance from Messi and a Suarez goal pulled them through and they probably should have scored one more on the balance of play. That is not to say that Sevilla did not have openings too, however, with N’Zonzi heading a corner wide from six yards with the goal gaping.

Portuguese Primeira Liga
Porto 1-1 Benfica (Sunday 6pm)
Our opinion on this one was that it was just too close to call, given the visitors’ better pedigree but poor teams news and the hosts being at full strength but also being named favourites by the market. Ultimately the outcome was as close as we had expected it to be, although our analyst fair score had the home side winning 1-0 on virtue of the opportunities they were able to fashion in the first half.

In the end it ended up being a half win for Benfica and a half loss on the suggestion of over 2.25 being worth a look due to the possible volatility of an early goal, although when Diogo Jota gave Porto the lead shortly after half time it did look to be on the cards. In the end Benfica left it until stoppage time to equalise from a corner, with substitute Lisandro Lopez scoring.

Norwegian Tippeligaen
Lillestrom 1-0 Molde / Rosenborg 2-1 Bodo/Glimt / Tromso 3-1 Odd / Stabaek 3-0 Start (Sunday 5pm)
The final day in Norway was a little less hectic than the final day in Sweden in terms of goals, but it was still packed with drama at the top and bottom of the table. Lillestrom and Tromso both survived thanks to home wins over sides much higher in the league who had apparently switched off for the break a week too early, while Stabaek did what was necessary against bottom club Start and champions Rosenborg held up their end of the bargain to consign Bodo/Glimt to automatic relegation. Our view was that not much would change at in the relegation fight, as we expected that Bodo/Glimt might take a point, but at least we were right to believe both Lillestrom and Tromso would be safe.

English Premier League
Chelsea 5-0 Everton (Saturday 5.30pm)
Chelsea went to the top of the table on Saturday evening after a thrilling win over Everton, keeping their fifth clean sheet in succession and looking every inch a title challenger. Conte has happened upon a great blend and was able to name an unchanged team yet again, while Koeman’s visitors definitely missed the energy and industry of Gueye in midfield. We liked the home team at -1 and were comfortable after twenty minutes, with Everton’s decision to switch to a five-man defence backfiring. Truthfully they would have struggled to cope with this version of Chelsea in any shape, however.

Ones to Watch

Swedish Allsvenskan
Helsingborg 2-0 Jonkopings Sodra / Ostersunds FK 2-4 Gefle (Sunday 2pm)
Helsingborg and Gefle went head-to-head to make the relegation play-off instead of going down automatically and both continued their excellent late season form with wins. Gefle in particular caught the eye with their performance, but still drop out of Allsvenskan.

Dutch Eredivisie
AZ 2-2 Ajax (Sunday 1.30pm)
This always looked a tough ask for Ajax off the back of their Europa League exploits, especially with AZ at full strength for the first time in a while. Our suggestion of backing the hosts at +0.5 deservedly paid off.

English Championship
Bristol City 0-2 Brighton & Hove Albion (Saturday 5.30pm)
Chris Hughton’s team produced a very professional performance to take maximum points from one of the toughest away trips in the Championship. Their early goals meant that our suggestion of backing them early at DNB never looked in any doubt.

German Bundesliga 2
Eintracht Braunschweig 2-2 Hannover 96 (Sunday 12.30pm)
We thought the timing of this game suited the hosts and they really should have beaten the -0.25 Asian Handicap, creating much higher quality chances than Hannover throughout and ultimately deserving a 2-1 victory according to our analyst fair score.

Turkish Super Lig
Galatasaray 1-2 Istanbul Basaksehir (Friday 5pm)
Sneijder, Balta and Inan did all pass fit and the line did move to -0.5, which made Basaksehir even more of a value bet at kick-off. The surprise league leaders continued their immense start with a hard-fought and deserved win.