In this most unusual of seasons, it is difficult to see what anything truly means.
The playoff race is no different.
Teams are not just a few games behind others, but entire
months late.
Adding to this that the gap between 13th and the
final playoff spot in 6th is only five points only reiterates it is still anyone’s game.
It is enthralling but incredibly confusing.
That’s why we have taken to the spreadsheets to predict – as
accurately as possible – what this playoff race really looks, and whether
Oxford have a realistic chance of getting into the coveted top six.
How does it work?
These predictions are based on each team’s underlying
performance data.
Using a side’s expected goals (xG) for and against (xGA),
coupled with those of their opponents, we can get a reasonable picture of how
many goals one side is expected to score and concede against a particular
opponent.
This is also adjusted to whether the side is playing them at
home or away.
This system has been used to calculate the games of all
teams from 6th to 13th to see how many points we can anticipate
seeing them on after 46 games.
For the purposes of this, any matches scoring within 0.10 xG
of each other have been classed as a draw.
Obviously, football games are not played on paper, but this
formula offers a valuable insight into how we can expect a team to perform
based on their remaining opposition (Along with whether their system of play
will suit their remaining matches).
Can Doncaster stay in the playoff places? Will Oxford sneak
in? Why are Portsmouth in 10th?
Nothing is for certain, but this should help to show if
Oxford fans can reasonably believe playoffs are still an option.
The prediction
Based on the current League One table, all sides above Doncaster seem relatively assured to finish in top six - but Donny’s recent form has thrown them back into the mix.
Also in there are sides right down to 13th, where
Blackpool and Accrington harvest the last realistic chances of reaching the
playoffs having played fewer games than those above them.
Its all to play for then.
Below is how that playoff mix looks after the simulation.
Only one side remains in the same position they currently in after
their final matches have been played, and there are certainly some interesting
results.
Portsmouth
Current position: 10th
Predicted Final Position: 5th
When, why and how Portsmouth find themselves in tenth position is the biggest mystery of the playoff picture.
At the beginning of March, Portsmouth sat in 4th
position, in touching distance of an automatic promotion place.
Since then its been four consecutive defeats – including a surreal
drubbing at relegation threatened Northampton.
The data, though, shows this to be a small blip.
They average the second most xG scored per 90 in League One and have
among the lowest xGA as well.
No League One side has averaged more shots per game than
Portsmouth this term and conceding only 19 goals from open play makes them difficult
to break down too.
It is no surprise, then, that the simulation has them picking up
the most points from their remaining games, with 30.
An easy fixture list - Ipswich are the highest opponent left to
play – supports their run-in data.
Only one side (Ipswich, 1.05) play sides with a lower average
points per game in their remaining matches than Portsmouth (1.24).
With the Tractor Boys up next, none of their final 12 opponents are in the top half of the table, making Danny Cowley's job a far easier one.
Consistent numbers both home and away – rare within this group of
teams – also helps them become the most likely side to claim 5th
spot, if not higher.
Oxford United
Current Position: 9th
Predicted Position: 6th
Yes, that’s right. Not only are Oxford in the playoff race, but on
this model, they are predicted to be one of the teams to win it.
With flying colours too, behind Portsmouth only on goal
difference.
It is a similar story to Portsmouth as well.
Having beaten nine teams in a row, none above mid-table, Oxford
have now played six sides in the top half in their last eight.
This was always going to be a difficult period for Oxford and the
key would be negotiating it to still be within touching distance of playoffs on
the other side.
They are not quite through it all yet – games against Lincoln and
Sunderland are still to come – but the strength of their opposition is a major
factor in why United are predicted to pick up 29 points from their last 12
games.
Given Oxford’s ruthless efficiency at putting these teams away in
the reverse fixtures, a haul like this does not seem implausible.
Having said that, Oxford are actually predicted to perform very
well against the stronger sides too.
Lincoln and Accrington both fall victim to Oxford in the
statistics world, whilst Blackpool are only able to get a point.
In fact, the only team that the data suggests will beat Oxford in
this period are the xG gods Sunderland – they top both the xG scored and xGA
tables and, in all honesty, blow away almost every side they come up against in
this simulation.
Despite being better in away matches this season – averaging over
0.30 more xG scored compared to at home – United are not hindered by having seven
of their last twelve games at the Kassam, where they pick up 19 points from a
possible 21.
They are yet to break into the top six this season, but the signs point to Robinson’s men being dark horses for a strong finish to the campaign.
Blackpool
Current Position: 12th
Final Position: 7th
It is fair to say that Blackpool certainly benefit from having
games in hand in this simulation.
It is not that they don’t deserve their predicted rise though.
They have performed admirably this season, particularly at the
back, where they have been one of the standout sides in League One.
This strong defence is enough to earn them 30 points from the 42
available, ambitious, but not undoable for a side who went on an eight game
unbeaten run earlier in the season.
They do, however, have eight games against top half sides plus
Accrington to come, though the data still suggests they will perform well.
They are also predicted to beat four out of the five sides below
them in the table, with four of these games coming in the last six match weeks,
suggesting they might be best positioned for a late charge to the
playoffs if it is not decided by then.
Having underperformed their data all season, it is within Blackpool’s power to use their numerous games in hand to push somewhere near the playoffs.
Doncaster Rovers
Current Position: 5th
Predicted Final Position: 8th
If the first three teams have been fairy-tale stories, this one
definitely isn’t.
Doncaster were top of the league on points per game at the start
of February.
Since then, Moore has departed, and Sunderland have opened up what
looks an unassailable gap between them and the promotion race.
Now, focus turns to ensuring they don’t manage to lose their
playoff spot.
This data isn’t good news then.
Their hopes are particularly damaged by eight away games from
their remaining 13, where – as Oxford can testify – they have looked incredibly
shaky in recent weeks.
All but one of their final five league matches are played against
top half sides, meaning if they continue on the bad form – which has seen them
win only twice in ten games – for much longer, it could be a steep hill to
climb if they need to creep back into the playoffs.
The simulation has them on a respectable 75 points, but that is
still two wins off both Portsmouth and Oxford, suggesting it would have to be a
real collapse for Doncaster to lose quite so much ground in reality.
As one of the streakiest sides in the division, though, anything
is possible.
Ipswich Town
Current Position: 7th
Predicted Final Position: 9th
It’s not what Ipswich fans will want to hear, but a 9th
place finish would be a fair result given how their season has gone.
Since the departure of Lambert, Ipswich have looked far from free
flowing.
They picked up a result against Plymouth recently, but poor
performances against Fleetwood and Gillingham have undermined Cook’s start at
Portman Road.
The data predicts they will struggle particularly against the
sides around them, with losses against Portsmouth, MK Dons and Charlton.
They have picked up only 18 points from their 21 league matches against sides in the top half so far this season, an average of 0.86 points per game.
These six-pointers are probably the killer blow for Ipswich’s chances in this model but they will also need to start showing more to convince us things will be different in reality.
Accrington Stanley
Current Position: 13th
Predicted Final Position: 11th
Accrington seemed to be constantly five games behind most teams
this season.
Unfortunately, they are not anymore.
It means that their impressive underlying numbers only see them
rise to 11th by the end of this campaign.
They should be exciting games though, with an average of 2.88 combined
xG in their remaining games by far the highest of the playoff chasing teams.
An 11th place finish would still be an impressive result for a club the size of Accrington, though, and their participation in
this discussion is an achievement in itself.
Fleetwood, Charlton, Gillingham
Current Positions: 11th, 6th, 9th
Predicted Final Positions: 11th, 12th, 13th
Only goal difference separates these teams in the predicted model,
with none making it close to the playoffs.
The common theme between the three sides; the data suggests they
are not actually very good.
They have all played 36 games so far, at least two more than
anyone else in the playoff race.
This advantage may well be papering over some cracks for three
sides who have not been great watches this term.
Charlton are the only interesting one here, with a predicted drop
from the final playoff position 6th down to 12th.
When you think, though, that they have claimed only 23 points from a possible 54 since the turn of the year – along with a change of manager – it becomes more plausible that a drop off is coming.
The teams finish 17 points behind the playoffs in the data,
implying they are not the ones Oxford should be too worried about.
Comments
Post a Comment