On the Uselessness of Draws

How many times does a manager or a player claim “a draw was a good result?” Whether this is justifying the result after the fact (i.e. if you’ve been 3-0 down a draw is certainly a good result), or whether the team went out to play for a draw, I decided to see how often a draw really was a good result given the pre-match predictions.

Who’s Happy With a Draw?

A draw is better than a loss and the one league point it brings is some comfort to fans, but more often it is two points lost rather than one point gained. Consider a match between evenly matched sides. In that case the home team has about a 50% chance of victory, and the away team about a 25% chance.

Who’s happy with a draw? The away team expects three points 25% of the time, and one point 25% of the time. So on average the away team should expect one point from the match: A draw meets expectations for the away team.

The home team, however, would expect 1.75 points on average from that match. If it ends in a draw they’re 0.75 points below expectations.

That simple example suggests that a draw is only a good result for weaker sides away from home, and if you’re at home you have to be much weaker than your opponent to be happy with a draw.

When is a Draw a Good Result in the Premier League?

I used Decision Technology‘s team strength model to generate expected points, home and away, for all 380 matches. I’ve highlighted the matches where a draw is good – that is, where the expected return is less than one point.

The results show that, at home, a draw is always a bad result for the top seven Premier League teams. Even for QPR and Wigan a home draw is bad news against opposition weaker than Everton. Overall, out of the 380 matches played per season, only in 49 (13%) would a draw be favourable for the home team.

Home Expected PointsAway from home things are a bit different, with a draw being a better-than-expected result in 159 out of 380 matches (42%). Even so, you have to be worse than your opposition to like the idea of an away draw: Norwich (ranked 17th) should only be happy with an away draw against Bolton (ranked 13th) or better.

The away chart shows that it’s in most teams’ interests to draw against the big clubs. That might explain why managers of lesser clubs are so defensive – it’s usually a profitable strategy. The answer for the big teams is to really go for the win.

Away Expected PointsThere is an argument that, in the risk-averse game of football, managers do not do enough to force a win. They can always claim “at least we won a point”, when the reality is that they should usually expect more than a point. Do you agree? Let us know in the comments.

5 thoughts on “On the Uselessness of Draws

  1. Hi Ian,
    I’m wondering why some 1.0 in your illustrations are highlighted and some are not. Are there any particular reason for this?

    But nice article, and I think that you are right about your assumptions on 1 pts. won contra 2 pts. lost.

    Regards
    Daniel

    • Hi Daniel,

      Thanks for the feedback!

      I should have mentioned that in the article – the highlighted cells indicate where the expected points value is less than one. So 0.99 shows up as highlighted (it gets rounded up to 1.0) but 1.01 does not get highlighted as a draw would be 0.01 points below expectations.

      Ian

  2. Hi Ian,interesting article,as ever.

    Here’s an alternative view.

    A home team with an initial Expected Points of 1.75 as you say will probably win 50% of their games and draw 25% and lose 25%.So when they draw a home game they are equalling or bettering the number of points they would expect to gain in 50% of their home games.That doesn’t paint the home draw in such a bad light.

    Expected points are derived from longterm probabilities,so it’s difficult to use them to evaluate a single game result.

    Mark

    http://thepowerofgoals.blogspot.com/

    • Hi Mark,

      Good point – in terms of probability the home team has about a 50% chance of not getting three points so indeed the draw doesn’t look as bad.

      The expected points are what we would see if the teams played a large number of games, and so in that sense they don’t apply to a single game. However, I think it’s useful for a team to have an idea of a par score – even if they will never get 1.75 points from a single game, it gives a sense of how damaging a draw is (-0.75) and how valuable a win is (+1.25).

      Loving your blog by the way!

  3. rich said:

    so Fulham are under-achieving away from home…

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Spam protection by WP Captcha-Free