Much has been made of Leeds’ great start to the season, but as supporters know, good runs soon came to an end under Garry Monk and in particular Thomas Christiansen last season. What can the stats tell us about this season so far, and are Leeds genuine contenders for promotion this time around?

Using expected goals we can look at exactly how Leeds have performed in games, decipher whether the apparent improvement is down to luck, and if not, how much exactly have they improved relative to the second half of last season.

The data tells us that Leeds have the 8th best expected goals per game in the division with 1.16, along with the second lowest expected goals conceded, with just 0.615. This is only bettered, unsurprisingly, by Middlesbrough.

These stats can’t be looked at in isolation however. The best offence in the division by expected goals is West Brom, however they also have the worst defense by this metric, conceding 1.61 per game.

The above gives a picture of each team’s defensive and offensive strength. Leeds, in the bottom right hand corner, clearly have performed well so far this season and deserve to be where they are. While Leeds are a better than average team in attack, their defense has been their main strength. For supporters, there is evidence to suggest this is not a false dawn.

A metric that can be used to rate teams comes in the form of the expected goals ratio. This takes the sum of all expected goals for a team and divides it by the sum of expected goals for and against the team. So a team’s rating is bumped up by creating chances, and bumped down for conceding them. This metric has been shown to be a good predictor of future team performance, detailed here: http://11tegen11.net/2014/02/11/putting-expected-goals-to-the-test/

Leeds ranking third in this metric is yet more evidence that their good start is far from lucky, and their results have been matched by their performances. Brentford meanwhile look like the team to beat.

If we look at the same metrics for the second half of last season, we can quantify the improvement Bielsa and his staff have induced in what is largely the same group of players as last year.

Looking at the data from Leeds’ Boxing Day win over Burton Albion onwards, there is an enormous improvement this season relative to last. Leeds’ expected goals conceded per game for this period was 1.46, well over double what we’ve seen at the start of this season. Interestingly, their xG per game is almost the same, coming in at 1.21. Leeds have benefitted in the early part of this season from some excellent finishing. The goals scored by Roofe and Klich away to Derby for example.

Comparing this season’s xG ratio to that of the second half of last season shows how each team’s rating has changed. Not surprisingly, by this measure Leeds are the most improved side in the division.

With the squad used to this point made up almost entirely of players at the club last season, the improvement that Bielsa has brought about is remarkable.

All in all, while Leeds may have benefitted so far from some excellent finishing, which may or may not continue throughout the season, there is a lot of evidence to suggest that Leeds have been good value for the results they have put together so far, and this is not yet another false dawn.