Talking Tactics: How did Sunderland manage to sacrifice a lead twice against Accrington Stanley?

After an improved performance, if not three points, against Hull City in midweek Lee Johnson made three changes to the side for Saturday’s game against Accrington Stanley. Denver Hume replaced Callum McFadzean in a straight swap at left-back, with Aiden McGeady replacing Jordan Jones on the left-wing and Aiden O’Brien coming in for Grant Leadbitter, a change which caused the 4-3-3 formation from midweek to become a 4-2-2-2 formation at the weekend.

John Coleman’s Accrington Stanley started the game in a 3-5-2 formation before changing to a 4-3-1-2 after half an hour. Savin started in goal, with Barclay, Nottingham and Burgess making up the back three. Rodgers and McConville started as wing backs, with Conneely, Phillips and Butcher starting in midfield behind the two strikers Dion Charles and Colby Bishop.

In what was an open first half, with a number of chances at both ends, Sunderland will have been happy to have managed to keep a clean sheet, whilst also putting away a couple of their chances.

I actually think that the xG graph below fails to show just how high-quality the chances were that Sunderland created in the opening half-an-hour, and Charlie Wyke’s headed opening goal especially is, in my opinion, underscored by an xG model which doesn’t take into account the player taking each shot.

Whilst Sunderland’s creativity in the opening half-an-hour maybe underappreciated by the xG timeline, what is not missing is the shocking drop off from Lee Johnson’s side at the beginning of the second half. The long red horizontal line shows that we went approximately 30 second-half minutes without creating a since chance and in that time Accrington found the back of our net twice.

Sunderland deserve credit for coming back once again to make the score 3-2 with Power‘s goal, but McConville’s brilliant freekick meant that Lee Johnson’s side were made to rue the mistakes which led to them sacrificing their two-goal lead at the beginning of the second half.

Check the Gallery at the bottom of the article for full data visualisations, courtesy of @markrstats

---

The Return of the 4-2-2-2

Despite an improved performance last time out, and Lee Johnson stating in his post-match press conference that the 4-3-3 formation used against Hull was his preferred shape, when facing up against a side in Accrington who used three central defenders, it's quite easy to see why the Sunderland manager decided to push Aiden O’Brien up alongside Charlie Wyke in an attempt to ensure his side were able to press the Accrington defence effectively.

Sunderland’s first goal of the afternoon showed exactly why there was merit in Lee Johnson’s decision to switch formations, as the combination of O’Brien and Winchester managed to dispossess the Accrington midfielder and showed the advantages of the formation off the ball, before the three players in the box for Lynden Gooch’s cross showed the strengths of the formation in attack.

A similar situation occurred in the buildup to Wyke, and Sunderland’s, second goal of the afternoon. This time Josh Scowen managed to win the ball in the middle of the pitch before releasing Aiden McGeady down the left who, like Gooch for the first goal, had three players in the box looking to attack the ball.

---

What went wrong?

Perhaps unsurprisingly, given Sunderland’s starting shape was designed to take advantage of the visitor’s 3-5-2 formation, the effectiveness of Lee Johnson’s game plan started to disappear when John Coleman reverted to a back four in a 4-3-1-2 shape as left-wing-back McConville went into midfield, left-centre-back Burgess shuffled across to left-back and centre-midfielder Butcher moved into a number ten position.

In truth, Sunderland actually performed quite well for the first five minutes of the second half as McGeady and Gooch both looked to create from central areas after coming inside from the wings. However, Sunderland’s second-half struggles mainly came as they looked to control the game more than they did in the first half a strategy which, in itself, wasn’t a bad idea.

The problem with Sunderland’s attempts to control the game as they held initially a two-goal, lead was that they continued with the 4-2-2-2 formation that has caused the first half to be so open and for Sunderland’s high pressing and attacking talents to out-gun their opposition.

As the second image in the gallery below shows, Sunderland’s pressing levels dropped off in the opening 15 minutes of the second period and, with the front four still operating in the same way as the first half, gaps began to appear between the attacking and defensive midfielders and out double-pivot began to be look exposed in a way which has been all-too-common in the last six games.

With this in mind, Johnson’s decision to bring Grant Leadbitter on in place of Lynden Gooch as he switched to the 4-3-3 formation we saw at Hull was a welcome one and a quick look at the xG timeline shows that the switch helped to nullify Accrington’s threat. That was until a mix up between Luke O’Nien and his goalkeeper meant Sunderland were forced to change their game plan once again in search of a winning goal.

---

The final roll of the dice

After the switch to 4-3-3 shored up the defence at the expense of our chance creation, Lee Johnson took the logical decision to sacrifice a midfielder for an extra forward by bringing Jordan Jones and Jack Diamond on for Carl Winchester and Aiden O’Brien and moving Aiden McGeady into the number ten position in a 4-2-3-1 formation.

The pace which was injected into the game by Sunderland’s attacking substitutions played a direct role in Max Power’s goal which put Lee Johnson’s side into the lead after Diamond picked the ball up in his own half after an Accrington corner and ran the length of the pitch before squaring the ball to Power whose shot deflected past the Accrington ‘keeper.

The final score suggests that Sunderland switched off after getting what they thought was the winning goal, but in truth McConville’s freekick was worthy of a point even if it’s Sunderland’s mistakes from earlier in the game that made it possible.

Knowing the result before I watched the game perhaps made this an easier conclusion to come to, but I am finding it difficult to get too angry over Saturday’s result. Of course, sacrificing a two-goal lead is inexcusable but when you consider that this was essentially a dead rubber game and that we conceded from a freakish own goal and a direct freekick, it isn’t the time to sharpen our pitchforks just yet.

Perhaps most reassuringly, an alternative xG model (below) gives Sunderland a 78% chance of winning Saturday’s game. If we’re going to be unlucky in games, I’d rather it was now than in the playoffs next month.

---