Relegation at the Riverside: Where has it gone wrong for Middlesbrough?

Josh Taylor – @weeklykickabout

Middlesbrough are in trouble. Big trouble. Sitting in 19th place, they have the fewest wins in the league this season. And there are some big reasons behind that. Boro have the least shots in the Premier League, 8.71 per game, and are one of two clubs with less than 10, along with fellow bottom feeders Sunderland. To have less shots than a Sunderland side is almost an achievement in itself considering more than a quarter of their shots have come from Jermain Defoe, and full back Patrick Van Aanholt has taken the next most shots for the black cats, despite moving to Crystal Palace in January.

When your formation resembles a back 6 rather than a traditional 4 or 3 at the back that seems to be sweeping the Premier League, goals are never going to be the aim. Karanka has always placed a focus on the defensive side of the game, much like his mentor while assistant manager at Real Madrid, Jose Mourinho. Boro’s 4231 formation transitions to a 631 with Negredo being the outlet up front, and Clayton and De Roon dropping into deep defensive positions alongside the central defenders. This defensive approach means that ‘boro have the most draws in the premier league, joint with Jose’s Manchester United. But how effective is this when it comes to avoiding relegation?

Defending a draw in the modern game is much less frequent than it once was. In 1980/81, there were 236 draws and a similar number the following season. That second season, 1981/82, was the season that 3 points for a win was introduced. The system was championed by Jimmy Hill, who said it would increase the incentives for a win, and therefore would provide more attacking football. 10 years later, the number of draws cut down to 135. Hill was right, and there was some mathematical reasoning for it.

Let’s make an example by taking odds of each result when ‘Boro come across a stronger opposition. If they attack, the probability of a win increases, but so does the chance of a loss. If they defend, the chance of a win decreases, but so does the chance of an opposition win. We can set this out in a table in the format W/D/L:

Stronger Side Attack Stronger Side Defends
Middlesbrough Attack 32/20/48 16/60/24
Middlesbrough Defend 16/60/24 8/80/12


So, we have our odds of each result based on how each team plays. Now we can apply the two systems, 2 and 3 points for a win to find the expected points.

Stronger Side Attack Stronger Side Defends
Middlesbrough Attack (0.32*2)+(0.20*1)= 0.84

(0.32*3)+(0.20*1)= 1.16



Middlesbrough Defend (0.16*2)+(0.60*1)= 0.92

(0.16*3)+(0.60*1)= 1.08



As you can see, in the two point model, if a stronger side attack, you would expect to take more points if you defend. In the three point model, the expected points is higher for when you attack, so that both teams are attacking, generally creating a more entertaining game. This holds true up until the point where the stronger side is twice as likely to win, where it becomes more effective to defend your point.

From this, we can conclude that either the management at ‘boro have not run models like this, or that they believe they are half as good as every other team in the Premier League. If the second is true, not being bottom of the league is an achievement in itself, but the first is the more likely.

So first we know that the concept of defending a draw is less efficient, but if you do need to defend a point, how should it be done? Well the best method for preventing the opposition from scoring is to prevent them from shooting. The best team at doing this so far this season has been Hull City, who have defended 25% of the chances against them.


Interestingly, Arsenal are right down at the bottom, but that’s a story for another time. Middlesbrough came out 4th, a fair showing considering they’re only .1% lower than Burnley, who defend like this (from @unfitforpurpose).

The Boro backline aren’t afraid to get stuck in either. They have by far the most tackles in the Premier League, and have blocked the most crosses. The opposition has taken the ball past the ‘Boro 18 yard line more than any other team except Hull and Sunderland, but they are middle of the table when it comes to shots conceded.

There’s only one statistic that really matters for a defense; Goals conceded. Middlesbrough have conceded 33 goals up to now, which is 30 less than Swansea, 5 less than West Brom and 1 less than Arsenal. The lowest placed team to have concede less goals than the Teesiders are Everton in 7th, and only 5 teams have let in less overall. If the Premier League were judged solely on defensive records, the board at Middlesborough would have been ecstatic with Karanka’s efforts.

The issue is that it isn’t. Unfortunately for Aitor, to win games you need to score goals, and goalscoring has been the issue up to this point. Middlesbrough have hit the back of the net 20 times this campaign, one less than Romelu Lukaku. If we look at the expected goals, it tells an even grimmer story. Despite the measly goals total, ‘Boro have actually outperformed their expected goals total by 8%, with their total xG up to week 25 sitting at 17.7, with 19 goals scored. In the three games since, the only goal was scored in the previous game, a 3-1 defeat to Manchester United. That was the first time since the end of January that the expected goals tally was above 1, the previous being another 3-1 defeat, this time to West Ham. In total, Middlesbrough have exceeded one expected goal just 5 times. On a game by game basis, of the 15 games that Middlesbrough have scored in, 12 have featured an over-achievement in xG, with their biggest win, a 3-0 win against Swansea, gave them an xG of only 0.7, while their biggest over-achievement was in a 1-1 draw with Manchester City, Marten de Roon equalising with a header in the 91st minute, one of four shots leading to total xG of just 0.15.

Negredo has been more and more isolated as the season, as shown by the timelapse of passmaps (from @11tegen11) below. The first frame is the 3-0 win over Swansea, where Negredo scored twice, the second is the 1-0 defeat to Burnley, the third a 1-0 defeat to Spurs and the final frame is the 2-0 loss to Manchester City.

This slideshow requires JavaScript.

  1. In the Swansea game, Negredo dropped deep and linked up play, something he’s used to from his time playing at Sevilla and Manchester City. It worked spectacularly, having their key attacker in a centralised system gave other players, such as Ramirez and Downing, the service and freedom to create attacks.
  2. The second frame, the loss to Burnley, was a similar story, but as usual with a Burnley away game, the attack was frustrated by solid defensive organisation.
  3. In the Spurs game, you can see Negredo beginning to become more isolated. Less than 3 passes were made with both wingers in both directions, and Adama Traore was the only player close to Negredo in terms of attacking positioning.
  4. The final frame, the City game, Negredo came on as a substitute in the 25th minute, but in the 65 that followed, he had 23 touches, and completed only 9 passes. In the opposite direction, the only player to make more than 3 passes to the Striker was Brad Guzan.

The individual frames are available here, or on @11tegen11’s twitter timeline.

According to StrataBet data, Middlesbrough have had 200 chances so far this season, ranging from Poor Chances to Superb, with a poor chance being converted around 2% of the time and a superb chance converted 75% of the time. This ranges through fairly good (8%), good (15%), very good (25%) and great (40%) chances. 84 of Middlesbrough’s 200 chances were classified as poor chances, the highest percentage in the league, even above Spurs, who have taken 50 more shots from outside the box than anyone else, and more than double the amount that Middlesbrough have taken themselves. This shows that Middlesbrough do not get into positions, even in the box, that provide good scoring opportunities.

In terms of judging the quality of Middlesbrough’s attacks, they again fare very poorly. Only Tottenham, Chelsea and Arsenal have a lower average number of attacking players in an attack. The difference is that Chelsea and Arsenal also have the lowest number of defensive players in front of the ball, while Middlesbrough are 11th. To refer back to the isolation of Negredo, of his 35 chances, 30 came with no attacking players between him and the goal.

One ray of light this season has been the emergence of Adama Traore, the La Masia product purchased from Villa in the summer. Only Eden Hazard has completed more dribbles than the Spanish/Guinean winger this season, despite playing only 21 games, 5 less than Hazard and his next closest rival, Wilfried Zaha. These three players are the only players with over 100 successful dribbles, with Sanchez the next highest on 73. However,


every dribble needs an end product, and unfortunately, Traore fares more similarly in this regard to Zaha than he does Hazard. Still without an assist, and with only 8 shots to his name, Traore has played only 12 key passes, less than a quarter of Hazard’s total and half of Zaha’s. However, aged only 21, there is more than enough time for his decision making to improve, and with a good coach, there’s no ceiling for his potential.

That ray of light might be one for the next few years, but Middlesbrough’s future in the shorter term looks increasingly bleak. Of the clubs in the danger zone, Sunderland have hope in Defoe, Hull have hope in Marco Silva, Swansea have Paul Clement, Crystal Palace have Allardyce, Leicester have Shakespeare, but Middlesbrough? What do they have hope in? Their long time manager is gone, and the likelihood is that Middlesbrough will be following suit, saying goodbye to the Premier League in the very near future.

This article was written with the aid of StrataData, which is property of Stratagem Technologies. StrataData powers the StrataBet Sports Trading Platform , in addition to StrataBet Premium Recommendations.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s