How Often Do Football Matches End in a Draw?

Draw Stats

If you want a really simple baseline figure to use, however, by looking at various draw stats over various leagues and going back as far as possible, it seems that a proportion of 25% is about right. So around a quarter, one in four, of all games will end in a stalemate. Odds of 3/1 would be fair for an outcome that has a 25% chance of success, such as two heads in a row coming up with coin tosses. That means that in the crudest sense possible, if the draw is priced at odds of more than 3/1, it can – historically speaking – be viewed as good value.

We will look in more detail at whether the draw offers value shortly. But for now let us return to answering the title question of this piece. Whilst 25% is a reasonable figure to have in mind, if we look more specifically at English football, analysis in 2014 showed that in 188,060 games between 1888 and 2014 in the top four tiers produced 47,412 draws. That’s 25.21%, so very much in keeping with the general figure we initially suggested.

If we look at a more specific data set, for example recent Premier League clashes, we see a rather similar picture. At the time of writing, with the 2020-21 season having recently begun, the average draw percentage for the last six completed seasons (picking up from where our previous data left off) is almost 24%.

It is interesting that despite some fairly significant variance even within a short six-year timeframe (from a low of 18.7% in 2019-19 to a high of over 28% in 2015-16), the average still stays very close to that historic 25% mark. In seasons such as these, pundits and media outlets are quick to try and draw conclusions and find reasons for the lack of or abundance of draws. They will say it shows how defences are getting on top, or managers are adapting a win at all costs approach, or how the game is becoming more or less tactical.

The bigger picture suggests this is just natural variance from one season to the next and that, in fact, these so-called experts are simply trying to find some explanation that broadly fits the numbers without bothering to think whether it really is a valid causal relation. That said, we are not arguing that the level of draws is fixed at 25% and that it always has been the case that around one in four games ends with no winner.

Changes to Draw Ratio Over Time

In general, the more goals that are scored in any given game, the less likely it is that that game will end in a draw. In the earliest years of professional, organised football in England, games were crazily high scoring. The first few seasons saw well over four goals per game on average and at this time the draw ratio was much lower. In 1890 fans saw a draw once every eight or nine games (12%) and the figure largely stayed under 20% during the 19th century.

However, the average goals per game rapidly dropped for a range of reasons including rule changes, improvements in tactics and fitness and a more professional approach to the sport. This caused a corresponding increase in the number of draws. Indeed, for the vast majority of seasons in the 20th and 21st centuries, the English football pyramid has seen a draw rate of between 20% and 30%. Only incredibly rarely has it gone over 30% and then only marginally (31% in 1977) whilst it has only dipped under 20% once since the early 1930s (18.7% in 2018/19 as per the table above).

Fluctuations, such as there have been, may simply be natural variance, or they may be down to rule changes and the other factors discussed above. One of the best examples of the latter was when the offside rule was changed in 1925. The relaxation of the law led to a huge jump in the goals average and draws dropped correspondingly.

However, it would seem that players, managers, formations and tactics soon adjusted and the stats for both metrics soon returned to more normal levels. We have perhaps seen this in microcosm at the start of the 2020-21 season with the use of VAR and the handball rule.

Handball Rule Reduces Draws

Stricter handball interpretations, along with the use of the video referee, led to a glut of spot kicks being awarded in the early days of the season. The previous high (for a Premier League campaign) was 106 in 2009-10 but following 20 penalties awarded in 28 games it was suggested we were on target to see a whopping 280 during the season.

These penalties helped fuel a torrent of goals at the start of the delayed season, with 144 in the first 38 games. That is an average of almost 3.8 goals per game, around a goal a game more than we would expect. Unsurprisingly we saw very few draws in this period, just three, equating to a mere 8%.

However, in the final round of 10 matches we saw teams begin to adjust to the new rules and, more importantly, referees were told to be more lenient in their interpretation of the handball rule. Whilst we still saw an incredible number of goals (helped by Villa’s 7-2 mauling of Liverpool!) we did see fewer penalties.

We would expect that referees will do their best to avoid giving too many penalties and players will become more comfortable with the rules and will learn how to avoid conceding them. Of course, in these crazy times it isn’t just rules that have changed. It is possible that games without fans mean less pressure on players and less urgent pressing, whilst fitness in this compressed season is also sure to be an issue. It will be interesting to see what happens with the draw ratio this term but we could well see a record low.

How Often Do European Sides Draw?

As said, we feel the 25% draw ratio is a reasonable ballpark figure to use over most leagues but how often do sides in the other big European competitions actually draw? Well, we looked at the last 10 seasons of La Liga, Serie A, Ligue 1 and the Bundesliga and our analysis supports our earlier theory.

In the decade we looked at, Ligue 1 and the Italian top flight saw the most draws, with La Liga seeing the fewest. However, all four, plus the Premier League itself, stayed within a fairly tight range between about 24% and 30%.

If we cast the net further afield to smaller leagues, unsurprisingly, you see a wider range of results. There are some anomalous draw stats when we look at obscure competitions or isolate a single season but, once again, we have to say that in general around one in four matches will end with neither team taking victory.

Is 3/1 Or Above Good Value on the Draw?

Earlier in this article, we said that based on the rough long-term, global trend, of around one in four games being drawn, odds of 3/1 or above could be viewed as decent value in a sense. Of course, this is only based on historic averages. It does not mean, for example, that odds of 33/1 on a draw between England and San Marino are amazing value. Equally, it does not mean that a draw between two sides who need a point to qualify from a group is bad value if it is only priced at evens.

Each game must be assessed individually on its merits having considered all factors. However, what sort of odds are typical for a draw and how often will you see the outcome priced at 3/1 or more? In general, you can expect to find average odds of between about 9/4 (3.24) and 12/5 (3.4) for two teams to draw a game, some way below the 3/1 you might seek.

To provide more concrete numbers from the real world, we looked at just over 100 games spread over a huge range of competitions taking place on a given day and the results of that analysis can be seen below.

• Highest draw odds – 25/1
• Lowest draw odds – 7/4
• Percentage of games priced at over 3/1 – 24.8%
• Percentage of games priced at under 3/1 – 66.7%
• Percentage of games priced at exactly 3/1 – 8.6%

NB: Total percentage slightly above 100 due to rounding

Note that the odds are based on the best available price from a selection of top UK betting sites. In addition, we should point out that the highest draw odds of 25/1 were something of an outlier. This was a German Cup game between Bayern Munich and minnows Duren. The next highest draw odds were more typical, at 23/5.

Whilst our sample is admittedly is not the largest, we feel confident that these results give a very accurate indication of the sort of prices you can expect on the draw. The vast majority of games from our selection saw the stalemate priced at between 3/1 and 4/1 and this is certainly very typical.