Monday, 5 May 2014

Winning the lottery of the play-offs

So, after the usual last minute drama, the play-off line up is complete. In the Championship, late entrants Brighton face a tough tie with Derby County, who achieved their record points tally for a season. In the other game, inconsistent QPR play Wembley specialists Wigan for a place in the final. League One sees an all southern-clash in the Peterborough v Orient matchup, while the north provide Rotherham and Preston. Finally, in the bottom division, Fleetwood face York and Southend take on Burton.

So, who's going to win?

(I've put a couple of notes about the statistics used in this post at the bottom. In particular, whenever I say 3rd, read 4th for League Two)

It's a cliche as old as the play-offs themselves that they are unfair, and that finishing 3rd means you are doomed to failure. Yet surely we would expect the highest-placed league side to actually perform better over the years. Or are the play-offs just a straight lottery, where league position has no impact? To find out, we ran the numbers. Let's start with reaching the final:


So across all three divisions, it is clear that the 3rd-placed team does tend to beat 6th place in the semi-final, on the whole at least. Across all three divisions, we would expect the 3rd placed team (i.e. Derby, Orient and Fleetwood) to reach the final 60% of the time. So on average, we can predict two of the three aforementioned sides will win their semi-final.

However, the gap between 4th and 5th is less distinct. In League One, the average remains the same, making it most likely that we'll be seeing Leyton Orient play Rotherham United at Wembley. In League Two, it appears to be essentially a toss-up statistically speaking, while in the Championship the numbers suggest we should actually be backing Wigan to beat QPR. Apparently, 4th place is almost as bad as 6th in that division...

Something which is backed up by looking at who will actually win the play-offs. According to our statistics, covering the 25 years of matches thus far, finishing third does give an advantage, making a given team twice as likely to win promotion than those finishing below them. So, chances are that at least two of our three third-placed teams will be playing in the division above next season. But is that an over-simplification?


Yes it is. Once we split the results by division, suddenly it's much less clear. League Two has seen a massive 13 out of 25 winners come from top spot (actually 4th in today's four up/four down system). So it is more likely than not that in any given year the team finishing there will move up to League One.

Yet in that middle tier, the picture is entirely different. Finishing 3rd has proven slightly more likely to end in play-off success, but 4th and 6th are almost as good. On the basis of these numbers, it looks like Preston might as well concede now! 

And then in the Championship, the statistics reverse again. Now it's 3rd and 5th who look most likely to gain a win, while 4th and 6th trail behind. We would only expect a winner from either of those positions put together every 3 years.  Then again, Brighton and QPR fans can take solace in the fact that the last such winner was 6th-placed Blackpool in 2010, so are those spots due a winner?



Looking at the last five years, an even stranger pattern emerges. Apparently since 2008, the cliches are true. The team in 6th, usually seen as a "form side" who have come from the chasing pack in the last few weeks, have won a staggering 33% of all 15 finals in this time period. In fact, 2008 was the only year that the highest finisher in all three leagues (Hull City, Doncaster Rovers and Stockport County) won promotion through the play-offs.

So, onto the official Stoppage Time predictions:

Championship

4th has done worst both across the history of the playoffs and in recent years. So we start by predicting a Wigan victory over QPR. To meet them in the final, recent form suggests Brighton, but we think the overall picture is a more reliable indicator, and so will pick Derby County to reach Wembley. Interestingly, every year Arsenal have won the FA Cup (5 times in the play-off era), the 3rd placed Championship club has NOT been promoted. Then again, the one year they were beaten finalists, 3rd placed Bolton did go up from the Championship. So if Hull win a week on Saturday, Rams fans should be confident! We've said that 2 out of 3 top-placed competitors should win, so we'll stick our necks out and say Derby.

Prediction: Derby to beat Wigan in the final. 


League One

Probably the most balanced historically, League One's playoff final is tough to predict. We're going to follow the historical trend by expecting the higher placed sides to contest it, namely Orient and Rotherham. 4th place has done better from this division than any other, so we'll go for the Millers

Prediction: Rotherham to beat Leyton Orient in the final


League One

We can't look past the historical dominance of the top side in this league. Third highest have done slightly better than second, so we'll send Burton through, but our money is on Fleetwood to triumph.

Prediction: Fleetwood to beat Burton Albion in the final

Do you agree? Fancy a 28/1 treble on our three picks? Or have we got it all wrong? Let us know in the comments below, or via twitter and Facebook.


Oh, and 12% of play-off finals have gone to penalties...




N.B In 1987 and 1988 play-offs were used, but they involved 3 clubs from the lower division and the lowest finisher not automatically relegated from the league above. As such, I've left them out of the analysis.

N.B.2 I've used the current names for the divisions, but obviously they were known under different names in earlier years. I've also used 3rd-6th for the Championship and League One, and 4th-7th for League Two, although some years (such as 1991 and 1995) these numbers were different. So where e.g. the top 3 went up automatically as in that year, I've simply counted 4th as equivalent to 3rd for that season. 


No comments:

Post a Comment