Sunday, 9 June 2019

Serie B - Draw Central

Last night's results from Spain mean that for the third consecutive season, the league with the highest Draw percentage is Serie B - or at least, it is top of the leagues that I cover.

Spain's Segunda DivisiĆ³n is messed up this year, with the decision in January to expel Reus Deportiu, but award 1-0 wins to their opponents in all subsequent matches. 

These unplayed matches might count in the table, but they all need to be ignored from any analysis, as was the awarded Bolton Wanderers v Reading match in the Championship this season.

Only one draw yesterday means that even if all three remaining fixtures finish all-square today, Serie B's 32.75% strike rate cannot be surpassed. 

Serie B has a long history with the Draw, so no one should be surprised. 

Of course, a high percentage of Draws doesn't mean that backing them blindly is automatically a profitable strategy, but it does mean that it might be a good area in which to start looking. 

In 2016-17, fewer than 7% of matches closed with the Draw priced at 4.0 or higher. For comparison, 41% of English Premier League matches see the Draw at this price.

Backing the Draw in 'close' games in Serie B, resulted in an ROI of 9.7% from 107 selections.

For 'close', I'm using matches where the 'true' difference between the two teams is 25% or less.

In 2017-18, this number climbed to a little under 12% of matches, and backing the Draw in 'close' games, resulted in an ROI of 24.8% from 105 selections.

Did this profitability continue in 2018-19? In short, yes. This past season saw fewer than 6% of matches with the Draw priced at 4.0 or higher, and in 'close' games, the ROI was 11.4% from 94 selections.   

In the seven years for which we have Pinnacle data, it does appear that the market now shows a clear bias in favor of the Home team. From 2012-16, the edge was on backing the Draw when the Away team was favoured, but in the past three seasons, it's better to back the Draw when the Home team is favoured, a 29.4% ROI.

No comments: