The Data

So, just what the hell is going on? How does this all work?

THE DATA

Essentially, we have a database that contains every international rugby match ever played since the first back in 1871 (Scotland v England in Edinburgh, if you’re interested).

We can query this database to pull out any statistics we (or you) want about matches played between any teams at certain grounds, at certain times, in certain competitions. You name it, we can give you the information.

We’ve spent a long time analysing this data, and hope to pull out certain aspects that could be of interest. At the moment we don’t have much (we’ve just started), but we hope to get some interesting articles up soon. Please follow to make sure you don’t miss any of these articles when they’re posted.

THE PREDICTIVE MODEL

So how do the predictions work?

Broadly speaking, our model looks at scoring patterns in historical games, both head to heads and generally, to work out an expected score for each team based on location and current ranking.

There are then a few tweaks throughout the algorithm that depend on tier, ranking differences and a few other parameters before the match is simulated 5,000 times to obtain the probabilities of certain outcomes.

More accurately speaking, this means that each predicted score is actually a metric to describe the relative historical strength of the two teams, based on their performances over previous years. However, we think it is more fun to think of them as predictions. At the very least, it certainly serves to provoke the trolls lurking on Twitter.

BETTING

So, you fancy a flutter on some matches based on our predictions? You must be mad! MAD! We have no idea what we’re doing. But, if you insist …

Our recommendations for betting are as follows.

As mentioned above, each predicted score is actually a metric to describe the relative historical strength of the two teams, based on their performances over previous years. What this means is that our predictions are generally a good place to start in forming an opinion on what may happen in upcoming matches, but you may want to plonk on top some of your own knowledge as well.

  • For example, we don’t currently take into account player selection or the impact of ‘big’ or Championship defining games, so if you think that there are a few other qualitative factors that may influence the outcome of the match then feel free to judge how much they will affect the predictions and bet accordingly.

We wouldn’t recommend straying too far from our predictions, generally speaking they are pretty accurate, around 85%, but if you think we’ve missed something, feel free to adjust for it. And please let us know after, thanks!

We’d also recommend looking at the odds relative to our predictions, and betting if it appears there is a large discrepancy.

  • For example, in the 2017 Six Nations opening match between Scotland and Ireland at Murrayfield, Ireland were tipped by the bookies as strong favourites. However, our model gave Scotland a reasonable chance of victory – we still had Ireland as favourites, but it was only by 3 or 4 points. In this situation, it may have been worth a punt on Scotland due to the mismatch in the predictions and odds. Indeed, Scotland went on to win that game.

A couple of don’ts … (or we’d really advise against)

DON’T 

  • Put a massive accumulator on all the games we’ve predicted. Some of them will be 50:50s and therefore could go either way. Have a look at the % chances we’ve given each team in each match and bet accordingly.
  • Come crying to us when your bet didn’t come off. This is sport, unpredictable stuff happens and that’s why we love it!

And the obvious one to finish…

  • Bet what you can’t afford to lose. Be sensible.

If you have any questions about the data, the model or anything else, please don’t hesitate to get in contact. There’s nothing we’d like better than a good argument about the importance of home advantage in the outcome of a rugby match!

Rugby4Cast