David Pilgrim delivers extensive analysis of the first 50 games of the behind-closed-doors Bundesliga era to provide some clues for this weekend's betting.
Follow Dave Pilgrim on Twitter @stattobets
Since returning to behind-closed-doors action a fortnight ago, we’ve now seen exactly 50 matches played over the top two leagues, and we’re starting to build a bit of a picture about how things might play out in the closing weeks of the season, and perhaps garner some clues as to what to expect when our own Premier League resumes.
Ahead of the German resumption we mused a little about how home advantage might be eroded from a typical 46% home win rate (and 30% away win rate). I suggested we might see it settled somewhere nearer 42% home wins and 33.5% away wins.
Now is the time to investigate what we’ve seen so far and ask ourselves whether there is any strong evidence which can help guide our betting. Reacting quickly will be a key opportunity and as punters the race is on against the bookies to consider the new landscape quickly and apply what we learn – the window of opportunity tends to be short in such things.
Results
Pre-COVID we were seeing a marginal under-performance from home teams in 2019-20 compared to traditional trends, with a 43% win rate for hosts in the top flight, and marginally lower in the second tier. In Bundesliga the unusually high goal expectancy leads to a lower draw rate than might be seen in other elite European divisions.
Pre Covid win percentages
Bundesliga | Bundesliga II | Total
Home: 43.3% | 41.3% | 42.3%
Draw: 21.9% | 32% | 27%
Away: 34.8% | 26.7% | 30.7%
Since the return to action, many commentators of the game have been quick to point out how rare home wins have been. Indeed, from 27 matches, only five have been won by the hosts, but it’s hardly a robust sample, and in a game of such scarcity of goals it would be entirely wrong to jump to conclusions.
Furthermore, if we look at Bundesliga II, there has been a notable drop in away wins instead – with just four away wins from 23 matches. Combining the pair has shown a drop to 28% of home wins, a marginal increase in away wins to 32%, but most interestingly, draws have gone through the roof, with 20 of the 50 matches ending all square.
Post Covid win percentages
Bundesliga | Bundesliga II | Total
Home: 18.5% | 39.1% | 28%
Draw: 37% | 43.5% | 40%
Away: 44.5% | 17.4% | 32%
All things considered though, a sample of 50 matches simply isn’t enough. We need to look deeper.
Goals
Pre-Covid, Bundesliga games were averaging 3.25 goals per game, with Bundesliga II games averaging a more modest 2.85. Since the return, both have dropped – the top flight to 3.1 and the second tier to 2.48.
In Bundesliga, that’s been entirely down to home teams failing to score at the same rate as previously – 1.75 goals per home side in the first half of the season has plummeted to just 1.30 goals per home side, while away teams have scored 1.81 goals per game versus a historical average of 1.50.
Considering it statistically, if absolutely nothing had changed, we would expect to see 35 or fewer home goals from 27 matches in around 4% of trials – it could coincidence therefore, but the data is leaning heavily towards there being a significant shift against the home sides – to the point that its improbable to explain it away purely as variance.
Expected Goals
Taking a look at the excellent Understat.com website reveals that their Expected Goals totals for the 27 home teams thus far reads 39.34, while the away sides have created enough chances to deserve 39.87 goals so far in the top tier. Since using Expected Goals allows us to remove a good chunk of variance, we can draw two reasonably solid conclusions.
1. Since returning behind-closed-doors games, home teams have under-performed against a “fair rate” in both scoring rates and games won. Put another way, over a small sample, they have so far been unlucky. Collectively, they deserve to have scored more goals and conceded fewer.
2. Notwithstanding point (1), since the return to action, home teams have absolutely failed to create as many clear goalscoring chances per match, and have allowed more attempts on their own goal than would typically be the case under usual circumstances.
What about the draw?
Since returning behind-closed-doors, the draw rate across the two divisions in question has rocketed from 27% to 40%. That’s a move from 11/4 to 6/4 if you’re a punter, so warrants closer consideration.
Both the decrease in overall goal rates, and the closing of the gap between home and away expected goals would suggest the draw becomes a more likely outcome. A basic football model however, using a tried and tested draw-bias (to cater for the fact that chasing teams work harder) will still tell you that the draw would be comfortably north of 2/1 in a match where we expected upwards of 2.7 goals. It’s pretty likely the exceptionally high draw rate will regress to something more normal, but it’s worth keeping an eye on.
Are teams more content to play out a level outcome in the absence of fans driving them onwards perhaps? It might explain a difference between “neutral venue” and “behind closed doors” encounters. For now, a watching brief is advised.
Other Betting Markets – Corners
This is an area I’m always interested in, and the data we’ve been seeing thus far points tentatively towards a lower corner count – perhaps due to a decreased intensity of the action. We have observed reasonably significant drops in the rate of corners being awarded in both divisions – Bundesliga down from 10.2 per match to 9.4, and Bundesliga II down from 10.4 to 9.8.
Looking at it statistically, based on previous corner rates, we’d have expected something in the order of 276 to have been awarded over the first 27 matches. The observed rate of 254 is notably lower, and a quick test on that suggests there is around a 9% chance we’d see such a low rate purely by chance and variance. Possible, yes, but it’s more than possible we are seeing a genuine trend towards lower corners. It’s very much something to keep in mind this weekend heading into the fixtures.
Other Betting Markets - Cards
In the cards markets, we hypothesised before the revival of top class football that we might well see a dip in total bookings points in matches, due to the absence of a vociferous crowd to bay and jeer. So far, I have witnessed little evidence of that.
We also pondered as to whether any drop in booking points would be likely to occur primarily based on fewer visiting players being booked. In that area, there is at least some evidence to suggest that might be an emerging trend.
Before the restart, average away booking points per match stood at 24.9 in a typical German top flight match. That means we’d have expected away sides to amass roughly 670 points. Since the restart away sides have actually collected just 510 points – a full 16 yellow cards fewer than expected. Again, I need to stress, sample sizes remain small and we could be looking at little more than noise, but it's looking more and more likely that the absence of a home crowd will reduce the frequency of away cards.
Betting Angles for the Weekend?
1. Consider opposing home sides, but don’t get carried away. I’m going to be looking at a starting point in my head of closer to 6/4 or 13/8 for a home side whom I consider equal in ability to the visitors. I don’t expect draws to continue at a rate anything above 33% really. Whilst home sides definitely look to be suffering relatively speaking, the market is also well aware and may even over-react to this change.
2. Corners totals could be lower than usual, but the margins are small. We’re potentially seeing a drop of around 0.5 corners per game, so you might be finding a little extra value backing unders, but a scattergun approach isn’t advisable. Among the sides who have seen the fewest corners in their matches this season have been Dortmund, Augsburg, Fortuna Dusseldorf, Hertha and Union Berlin.
3. I’ll be focussing on away sides picking up fewer cards than usual. That could be in the match bet market, or backing Under Away Team Cards – particularly looking for games with lenient referees and disciplined team stats.