Make no mistake—I am enjoying this season very much. This is the best chance to get into the
Champion’s League Southampton has ever had.
It may be the best chance we ever have.
That does not mean that we will succeed.
To understand the rational basis of my pessimism, it is
important to understand the concept of reversion or regression to the
mean. See here and here.
As translated to a sports context, teams that are doing
poorly are often not as bad as they appear and teams that are doing well are
often not as good as they appear. (For
that matter, teams that are doing average are often not as average as they
appear.) This means is that there is a
tendency for clubs that have done well to regress and for clubs that have done
poorly to improve. This is very likely
the main reason clubs seems to improve when they fire their manager. The manager only gets fired after a string of
bad results when the club is playing below its true ability. An improvement is virtually inevitable no
matter who is the manager.
We can see that has happened in the Premier League this
year. Chelsea earned 2.6 points per game
in their first ten games but only 2.1 points per game since for an average of 2.3
points per game. We earned 2.2 points
per game in our first ten games but only 1.5 points per game since for an
average of 1.9 points per game. On the other
hand, Burnley earned 0.4 points per game in their first ten games, but 1.5
since then for an average of one point per game. QPR earned 0.7 points per game in their
first ten games, but 1.1 per game since for an average of .9 points per game.
The mean to which clubs regress is an individual thing. Clearly, Chelsea, Southampton, and Burnley
are clubs of different quality who are moving toward their own mean performances—not
a purely average performance for the league as a whole. (Note that I am
simplifying things here by assuming that there are enough games in the Premier
League season to be statistically significant for this purpose. In many ways that is not true.) Manchester City is a good example of
this. After ten games there were averaging
2.0 points per game, but they are now averaging 2.2 points per game. Very likely, Manchester City was slightly
underperforming at the start of the season.
My doom and gloom, such as it is, is based upon the undetermined (and indeterminable)
true level of quality of our club. Are
we a 1.9 points per game club which should end its season with approximately
71 points? Or are we a 1.5 points per
game club, which means that we should expect to earn 25 or 26 more points this
year end the season with 64 or 65 points?
Or have we played above our heads the whole season and should we expect
to get just a point per game and end up at 56 points again? Keep in mind that even if we really are a 1.9
point per game caliber club, we could still underperform (or over-perform) for
the rest of the season.
All our optimism and hope is based upon a belief that we
really might be as good as or better than we have played so far and that we are
good enough and lucky enough to keep it going for the rest of the season. No one can say for sure if this is true. The history of the Premier League probably
suggests that it is not.
Consider what it has taken to get into European Football qualifying
spots in the past. (I am assuming that
the winner of the League Cup will finish in the top seven of the Premier
League. If this turns out not to be
true, sixth place may not be a European Qualifying spot.)
Season 1st 2nd 3rd 4th 5th 6th
2013-14 86 84 82 79 73 69
2012-13 89 78 75 73 72 63
2011-12 89 89 70 68 62 58
2010-11 80 71 71 68 62 58
2009-10 86 85 75 70 67 64
2008-09 90 86 83 72 63 62
2007-08 87 85 83 76 65 60
2006-07 89 83 68 68 60 58
2005-06 91 83 82 67 65 63
2004-05 95 83 77 61 58 58
2003-04 90 79 75 60 56 56
2002-03 83 78 69 67 64 60
2001-02 87 80 77 71 66 64
2000-01 80 70 69 68 66 61
1999-00 91 73 69 67 65 58
1998-99 79 78 75 67 57 55
1997-98 78 77 65 63 59 58
1996-97 75 68 68 68 61 59
1995-96 82 78 71 63 63 61
Average 86.6 79.4 73.9 68.3 63.5 60.6
Range 75-95 70-89 65-83 60-79 56-73 55-69
The first thing that jumped out as me is how the final table
last year exaggerated how far away we were from competing for Europe. Compare last year’s results to a straight
line extrapolation of this year’s point totals:
2014-15 2013-2014
Chelsea 89 1st 86
Man City 85 2nd 84
Southampton 71 3rd 82
Man U 67 4th 79
Arsenal 65 5th 72
Spurs 62 6th 69
69 points was the highest sixth place total since the
Premier League dropped to 20 teams. 73 points was the highest fifth place
total. 79 points was the highest fourth
place total. 82 points was only one less
than the highest third place total.
Getting into Europe was hard last year.
Every indication is that it will be easier this year—although winning
the League itself will be more difficult.
Historically, our extrapolated total of 71 points puts us
somewhere between third and fourth on average and no team with that score has
finished lower than fifth. However, the
extrapolated point totals are lower than average this year for both third and
fourth (but higher for fifth and sixth).
If we rule out finishing first or second, what do we need to
do to qualify for the other European spots?
Specifically, how many points per game must we average to finish third
through sixth?
Most Avg. Least
Third 2.6 2.1 1.5
Fourth 2.4 1.7 1.2
Fifth 2.0 1.4 1.0
Sixth 1.8 1.3 0.9
Since we can be relatively confident that this will not be a
year with an all-time high (or low) point total for third or fourth place, it
certainly looks doable. Two points a
game should be good enough and 1.8 might be good enough for the Champions
League. That is just 31 to 34 more
points. If we just beat all the bad
teams that is 24 points. Add in beating
Swansea and Newcastle in our next two games and we are almost there.
Apparently, I have trouble doing doom and gloom.
No comments:
Post a Comment