“Do not underestimate the muteness of a tree - Munia Khan
A grove of unsuspecting trees lurk amongst the leaders of the English Premier League.
Neither dominating their competition nor being dominated, Nottingham Forest are sitting in fourth place in the EPL table with just six matches remaining in the season.
I cannot blame those who may have underestimated Forest coming into this season. The Reds had finished in 16th and 17th place since their promotion back to the Premier League in 2022, and their 32 points from last season was the lowest total ever for a team that avoided relegation.
And this year’s squad has a lower non-penalty expected goals difference than last year’s squad!
This summary from Forest’s March 8th match against Manchester City summarizes the season well. City are one of the top five teams in the EPL and have the fourth highest payroll of any team in the world. Boasting stars like Erling Haaland, Rodri, and Kevin De Bruyne, what did Forest do well to beat them? SHRUG
Since the Premier League began tracking expected goals in the 2017-18 season, no team with a negative non-penalty expected goal difference has finished higher than sixth in the league. We should expect a team with Forest’s non-penalty expected goal difference to finish the season with somewhere between 50.34 and 51.41 points on the season (used a linear regression model to predict season points with 95% confidence interval and p < 0.0001). This would have placed them 10th in the league last season and 11th two seasons ago.
I suppose now is a good time to pause and discuss what expected goals means. When a team takes a shot at goal, there is a probability that that shot will result in a goal. A wide-open empty netter will have a probability near 1 while a prayer from the middle-third will have a probability closer to zero. Most shots have a probability somewhere in the middle. A team’s expected goal value, then, is the sum of those probabilities.
The process of determining the xG for a particular shot is dependent upon the distance to goal, the angle of the shot towards the goal, keeper positioning, whether it is strong foot/weak foot/header, and a number of other variables. If you would like to read more, Opta Analyst have a great article on the stat. For now, the simple example given above should suffice.
| Squad | Avg. Shot Distance (Yards) | Goals | Exp. Goals | G-xG Diff |
|---|---|---|---|---|
| Wolves | 18.1 | 46 | 37.8 | 8.2 |
| Nott'ham Forest | 17.7 | 50 | 39.0 | 11.0 |
| Ipswich Town | 17.6 | 32 | 30.2 | 1.8 |
| Leicester City | 17.4 | 27 | 27.7 | -0.7 |
| Everton | 17.2 | 31 | 35.0 | -4.0 |
| Manchester Utd | 17.2 | 36 | 42.1 | -6.1 |
| Crystal Palace | 17.1 | 39 | 49.2 | -10.2 |
| Brighton | 17.1 | 49 | 49.3 | -0.3 |
| Manchester City | 16.9 | 61 | 57.8 | 3.2 |
| Chelsea | 16.7 | 54 | 60.4 | -6.4 |
| Bournemouth | 16.7 | 51 | 58.3 | -7.3 |
| West Ham | 16.4 | 33 | 40.1 | -7.1 |
| Fulham | 16.4 | 46 | 42.5 | 3.5 |
| Aston Villa | 16.3 | 48 | 48.4 | -0.4 |
| Southampton | 16.2 | 22 | 28.9 | -6.9 |
| Liverpool | 16.0 | 74 | 69.7 | 4.3 |
| Newcastle Utd | 15.9 | 59 | 54.4 | 4.6 |
| Tottenham | 15.9 | 57 | 52.2 | 4.8 |
| Brentford | 15.4 | 52 | 47.6 | 4.4 |
| Arsenal | 15.0 | 55 | 49.4 | 5.6 |
So, we’ve established that shot distance is a big factor in determining the expected goal value of a particular shot. This data tells us that they generally take shots from further out than just about any other team in the EPL - only Wolverhampton take shots from further out. Yet they are still scoring ata great clip. Forest lead the league in goals over expectations, the difference between actual and expected goals.
But the offensive third is just one component of the beautiful game.
Something that was surprising to me when putting together data for this post was that Forest do not control the ball very well. They have the lowest possession percentage of any team in the Premier League.
While they do not control the ball incredibly well, their defense is stifling in their defensive third. Forest love to make attackers uncomfortable, pressing them and not allowing them to get off a clean shot. They are second in the Premier League in tackles in the defensive third of the pitch which makes life easier on keeper Matz Sels.
| Team | TklsWon | Def 3rd Tkls | Mid 3rd Tkls | Poss% | CleanSheets | Pts |
|---|---|---|---|---|---|---|
| Wolves | 393 | 354 | 255 | 47.5 | 6 | 35 |
| Nottham Forest | 356 | 345 | 182 | 40.1 | 13 | 57 |
| Manchester Utd | 413 | 331 | 262 | 52.8 | 9 | 38 |
| Crystal Palace | 395 | 327 | 261 | 43.7 | 9 | 43 |
| Everton | 369 | 319 | 225 | 40.5 | 10 | 38 |
| Leicester City | 358 | 318 | 232 | 45.6 | 1 | 18 |
| Fulham | 346 | 273 | 226 | 52.3 | 5 | 48 |
| West Ham | 323 | 265 | 205 | 47.5 | 6 | 35 |
| Southampton | 327 | 263 | 239 | 49.8 | 2 | 10 |
| Bournemouth | 357 | 257 | 273 | 47.3 | 7 | 48 |
| Aston Villa | 313 | 255 | 217 | 51.5 | 6 | 54 |
| Ipswich Town | 287 | 248 | 188 | 41.0 | 2 | 21 |
| Brentford | 319 | 243 | 185 | 47.7 | 6 | 43 |
| Tottenham | 367 | 241 | 248 | 56.6 | 6 | 37 |
| Newcastle Utd | 327 | 238 | 203 | 50.2 | 11 | 59 |
| Brighton | 346 | 232 | 246 | 52.5 | 7 | 48 |
| Liverpool | 338 | 231 | 232 | 57.7 | 13 | 76 |
| Chelsea | 307 | 208 | 199 | 58.4 | 8 | 54 |
| Arsenal | 297 | 190 | 216 | 56.3 | 11 | 63 |
| Manchester City | 260 | 161 | 183 | 61.2 | 9 | 55 |
This pressure has helped Sels notch 13 clean sheets thus far and has him as the current favorite to win the Golden Glove while tied for the most clean sheets across the five major European Leagues (RB Leipzig’s Peter Gulacsi also has 13 clean sheets across 27 matches, compared to Sels’s 32 matches played).
I won’t lie and tell you I know exactly what is causing Forest to play so far above their expectations. I’m a casual fan and I only have so much data to pull from. Which, major thanks to FBRef.com for the data I have used in this post.
I do always find it fascinating when a team’s actual results far outperform (or grossly underperform, but that’s less fun) what the numbers tell us should be happening. I think I’ve found some factors through this analysis that are likely contributing to this overperformance of expectations. There is probably no one thing we can point to and say, “yes, this is why Forest are so good despite their expected performance being average.”
I am, however, very much looking forward to watching Forest for the rest of the season.