Part 2: Using Expected Points & Goal Difference to predict a league table

In my previous post on Expected Points I discussed the differing metrics I can use in roughly trying to predict the outcome of a league season in the format of a final league table. Due to the length of part one on the EPL I decided it would be easier to do a second post rather than bore everyone to tears on the first post but thanks for a massive response all the same! For the explanations of the parameters and metrics used in compiling these tables I suggest you refer back to the first post which can be found here https://wp.me/p8dqLB-8l

EFL Championship and SkyBet Leagues 1 and 2

Having been involved in differing capacities with clubs below the EPL, alongside supporting the best team in the midlands, Aston Villa (don’t @ me!), I have not only a vested interest in the Championship and League’s 1 and 2 (I will leave the National League out of these posts for now for post length reasons) but I also compile and collect these 3 league’s data throughout the season and find that they are very interesting when it comes to using various measures when producing data work.

Although there is some work published I believe that there is nowhere near enough detailed data and stat work around given the availability of various data sources however publicly available data is few and far between in all fairness. These three leagues are all well supported of course and fans maybe don’t get enough stat work done, or publicly published anyway, on their clubs but especially fans from clubs in Leagues 1 and 2. I should say that the data I used is up until and including the 6th of October’s fixtures in all three leagues.

More on Leagues 1 and 2 later but I will start with the Championship and alongside the tables I will sum up little about what I found and then see what possibilities could end up taking place come the end of the season.

EFL Championship

Champ xGD
EFL Championship goal difference table

As I mentioned in the first post on this subject I had my doubts about using expected goal difference as a valid measure of a team’s final league position but in the process of exploring the data and using the metrics went along it kind of grew on me a little.

Dean smith
Will Dean Smith’s departure for Aston Villa radically effect Brentford’s promotion hopes?

Brentford fans will be pleased as punch to see their side at the top of the goal difference table but they might want to hold off due to the fact manager Dean Smith has now left the club to join Aston Villa and this of course will most probably have a huge impact on where their season goes from here. A +22 goal difference would have earned you a final league position of 5th or 6th position if we go, as we have done in our parameter, by last season’s final league table. Wolves won the league with a +43 GD with the nearest being another promoted club Cardiff City with +30 GD and of course it’s still possible to achieve Cardiff’s GD but unlikely anyone will touch Wolves’ 17/18 season total surely.

My doubts creep back in again when I see the high pressing Marco Bielsa’s Leeds team finishing the season with a GD of +14 when they already have a +13 GD at the time of writing but there is a long way to go of course and goal differences will sway, some considerably,  as the season progresses. Sheffield United have started very well but xG doesn’t always match up with the reality but in theory a playoff spot, if not more, is easily attainable for Chris Wilder’s men.

Sheff wed
Sheffield Wednesday bottom of a table? I have my doubts

So who should be worried? Well Hull and Ipswich (yes really) could be in trouble as most certainly will relegation favorites Rotherham United but the surprise (okay MASSIVE surprise!) is seeing Sheffield Wednesday at the bottom of this table. This is just not going to happen Wednesday are currently 6th in the table with a GD of +1 so how is it feasible they could finish bottom? They have conceded 15 goals with an xGA of 18.1 (9th highest in the Championship) so still conceding 3.1 less than predicted, a small margin granted. And they are exceeding there xG by 4.3 goals (xG is 11.7 and actual goals scored is 15) again this isn’t a great number but it is still 8th highest in the league, not bad (I am planning on publishing all leagues current xG numbers when I get a chance). The issue is essentially conceding 18 goals and scoring 19 isn’t sustainable over a 46 game season and if this continues they will find themselves slipping down the table somewhat however not to the extent that the table above shows. Wigan, Bristol City, Stoke, Villa in 12th, 13th, 14th and 15th respectively all have similar records so the although the final table is skewed somewhat it’s easy to see how, given the four teams positions noted above, Wednesday could end up much lower down the league come May next year.

Anyway let’s move on to the final expected points table to see if we can correlate anything the goal difference table tells us.

Champ xP
EFL Championship expected points table

Now this is more like it. For me this passes the eye test quite well. The top 6 sits quite well with me and the fight for the title, I expect, will be just as close as the table suggests although Sheffield United winning the title may be a stretch for me.

Sheff utd
Some numbers suggest otherwise but I think Sheffield United should be taken seriously this season

 

 

Brentford as we have discussed may have issues now Smith has departed. In general the top five are the same give or take (yes I know there are is no such thing as 77.04 points but you can see what we are trying to achieve here) 7th-18th is again very familiar and also the bottom 4. Sheffield Wednesday have managed to get out of the bottom four and finish 20th,  probably more believable but I still have my doubts they will finish that low we shall see.

Summary

The two tables do correlate in general but with complications for teams in regards to managerial changes, injuries and schedules etc there can always be a massive swing either way on how things can change especially with two or three big wins changing goal differences up and down the table. I like the expected points tables in general and although being massively swayed by the teams xG numbers these are consistently more reliable than using current points per game. I would expect Leeds and Middlesborough to better their positions in my table but I’m interested to see how Brentford, Sheffield United and Derby progress and also how the Dean Smith effect at Aston Villa pans out.

SkyBet League 1

SkyBet L1 xP
SkyBet League 1 goal difference and expected points tables

Four out of the last five seasons have seen the League 1 title won by a team with a points total of 90 or more and Barnsley look on course, even at this early stage, to achieve this, well maybe.

Image result for barnsley fc v luton town
Barnsley earned another 3 points against Luton on Saturday

With the lowest xGA conceded (12.5 actual conceded is 9) and the highest xG at 23.1 (they exceed this by 2.9 with 26 goals scored) and scoring at 2.1 goals per game the numbers are looking really good for the Oakwell outfit so far who are in 3rd.  Portsmouth and Peterborough lead the way currently and both of those sides are looking good and with the latter’s expected goal difference of -5 in my final table this again questions my thinking but 19 goals conceded with an xGA of 21.2 kind of skews the table given that Posh have currently scored 30 goals and exceed their xG by a whopping 8.8 this then answers the question of my table somewhat in terms of goal differences but again slightly questions the sustainability of Steve Evans’ teams goal scoring and goal conceding ratios over the course of a season.

Luton Town are another side that has real potential in at least achieving a playoff spot. In their first season back in League 1 and in both the goal difference tables and the expected points table the Hatters are predicted to end up in the higher echelons. With a predicted GD of +17 (last season this would get you 4th-5th position and the predicted 77 points would get you the same position wise) Nathan Jones has a real team spirit and huge work ethic ingrained into his side which I believe will all help in them in achieving a somewhat surprising top 5 finish.

_91527152_nathan_jones-4
Luton Town manager Nathan Jones

Sunderland,  Southend and a resurgent Blackpool all look to have decent chances of a playoff berth each come the end of the season. At the other end of the table Plymouth, who finished 7th last season but currently sit bottom of the table this season, already look worryingly frail at the back with a rather large 2.18 xGA per match (actually conceding 1.83) and with a huge total xGA of 26.3 after 12 games played they have been very lucky to concede just the 19 goals maybe they have a kind of alright defensive policy after all? Or does it say more about the shooting ability of the teams they have faced? You can decide (Hint: it’s not the first one) anyway it doesn’t make for optimistic reading and I predict them to finish bottom, and some way short of 23rd, in both of my predictive tables.

These results and metrics used are probably shoved to the back of any clubs data driven stat work at this level of football and is probably paid no attention to at all when in reality this could be massively important and relevant in a clubs performance now and in a predictive nature.

SkyBet League 2

SkyBet L2 xP
SkyBet League 2 goal difference and expected points tables

So to the last league being covered in this article SkyBet League 2 and one of the most competitive with a standard being very similar throughout the league. There are some interesting outcomes here but let’s start at the top as it stands.

cowley brothers
Nick (left) and Danny Cowley have got Lincoln firing again

One thing is for sure so far this season. The Cowley brothers and Lincoln City side have found their promotion winning form when winning the National League two years ago. Finishing 7th last season, The Imps already have just over half the points they achieved in the whole of last season. With an xG total of 14.3 this season Lincoln have scored 20 goals at the time of writing (games inc. 6th of October remember) a plus of 5.7 xG means they sit pretty at the top of League 2 and also predicted to keep that top spot come the end of the season with a points total of 83.68 (or 84 if you like your numbers rounded up) a points total that last year would have got your team a 4th place finish stretching to 3rd (Wycombe finished in 4th with 84 points last season).

I can’t quite get my head around a +18 GD when winning a title in this league but this is what MK Dons are predicted to achieve in my goal difference table and its odd I agree especially given Lincoln already have +17 currently but in achieving fourth last season Wycombe did finish with the same goal difference so again although doubt creeps in it is still feasible, maybe. Speaking of MK Dons I have them coming very close in the expected points table with a predicted total of 78 points which last season would see you end up in either 5th or 4th position so comparing my table with last season’s it won’t be far off.

Looking up and down there isn’t much that sticks out between the two tables in terms of positioning and GD is pretty good as well with Grimsby predicting to have a -18 GD (23rd and 24th ended with -13 and -15 respectively last season) and could Mansfield end up with the same amount of points as last year (72) as predicted in my table? Can Notts County recover from a disastrous start to the season and end up escaping to 20th position with 51 points by the end of the season? in any instance it is more likely than them achieving 77 points as they did last year.

In the last two seasons 46 points or above would see you safe from relegation but using my measures 50-52 will be the probable cut off this year and the likes of Macclesfield (currently on 4 points), Cambridge (8), Cheltenham (9) and the aforementioned Notts County (12) will not be wanting to hear that I’m sure.

So there we have it I hope I have cleared up a few things and maybe reaffirmed some peoples thinking and hopes for the season ahead or maybe I have opened a few eyes to what is possible when using these metrics alongside the parameters I have used when compiling these tables.

Probably worth putting a few quid on anyway, maybe.

Gareth Cooper

GC Analytics

If there are any queries or questions in regards to the past two articles on this subject or others involving xG, xA or any predictive measures you have seen and if you think they may help when scouting teams or players or even if you think I could help your club or department in any of these areas feel free to get in touch.

 

Expected points & goal difference- How accurate can predicting a final EPL league table be?

Points per game is boring

Trying to predict the outcome of a team’s position at the end of the season is always going to be tricky and probably, in most people’s eyes, near on impossible and in some people’s opinions pointless however there are ways in which we can get reasonably close in achieving this. Given it’s the international break (yawn!) I thought this was a good time to publish what I have been working on.

Goals per game and goals against per game, points per game have all been used in the past to calculate the final league positions of clubs across various leagues but the future is nigh and when expected goals came along in giving us a better measure of a team’s performance across not only a single game but also longer term across a whole season we also gained a significant and more reliable way of assessing the way a league table may look at the end of the season.

I am going to be using two metrics. Expected goal difference and xP (expected points) which I believe are sustainable as they have been used in articles published before and also they are fairly accurate. In calculating expected goal difference I will use the difference between xG and xGA multiplied by the remaining games to get our goal difference totals and league position finishes and for the xP component I will use a more intricate method detailed below.

For this article I will be focusing on the English Premier League and in a few days publish another article covering the Championship and Leagues 1 and 2.

Expected Goals (make a) Difference

Using goal difference is probably not the most thought about way of trying to predict a final league table and work has been published in the past using simple goal difference and it has been proven to be a good indicator of the longer term sustainability of a team’s results but in this article I will not only put forward the case for what I believe are the benefits of using this measure but also why it may not be the best indicator.

So let’s start with the methodology used in creating the tables within the goal difference using some examples from a previous season.

The best teams earn the most points and it’s fair to say they will end up scoring more goals and in theory will end up conceding the least amount of goals over the course of the season but will this dictate their final league position? In the mind’s eye of course but in reality is it the same?

In an excellent, albeit brief, article published in 2012 on 5addedminutes.com Newcastle were used as an example of this so rather than trying to search through seasons and seasons worth of data to try and find something similar I will use them as an example.

The EPL does not seem to change in regards to competitiveness from year to year and it is often spoken about that there are three ‘mini leagues’ within the EPL. In the 2011-12 season Newcastle United finished the season on a goal difference of +5 and 5th in the table on 65 points (normally this would equate to 55 points and only be good enough for around 7th or 8th position) so in theory they overachieved by 11 points and with only eight wins by a single goal and a trio of big defeats. These slender one goal margin victories can inflate a team’s final league position and although results fluctuate this may in part explain the Magpie’s final position in the table and of course there is no reason to believe this is sustainable winning by these small margins. The following year Newcastle finished in 16th position with a -23 goal difference with a 41 points total proving that it certainly wasn’t a sustainable way to play over the course of a season.

The model above, as I explained above is merely using standard goal difference so what happens if we use expected goal difference to produce an ‘alternate’ final goal difference table.

The graphic below shows the EPL table at the end of the season when we apply this model.

EPL xG diff
EPL 18/19 final league table using Goal difference as the measure.

The method for this is simply xGF per game – xGA per game = goal diff / 36 (games remaining) = final goal difference. Then we can assign a final league position based on the given clubs goal difference.

This is where I have some doubts about the reliability of this method with some of the numbers a little low on the eye test but we shall see in time. When referring and compiling anything along these types of lines looking at previous data sets is vital and it is imperative to go back and look at a previous season or seasons. For simplicity, we will look back to last season’s final EPL table.

Manchester City topped the goal difference charts with a +79 goal difference and finished as champions and in my table above this year they could finish with a +69 GD, a Liverpool (4th last year with +46 GD) are expected to have a +40 at the end of this year so all looking pretty rosy so far in comparison.

Things get a tad shady when we go a bit further down my table. Watford are expected to end up with a GD of +26 and a possibility of finishing 4th when last year they finished on -20 and ended up finishing in 13th  on 41 points, quite a large margin and unlikely I think it is fair to say. Bournemouth, with an expected GD of +19 and a final league position of 5th   (it should be noted I have placed them above Spurs for alphabetical order purposes only in this table), performed significantly worse last season with a -16 GD and a points finish of 44 points,  one would assume they are unlikely-ish to finish in the Europa League positions this year.

Watford celebration.JPG
Watford- Can Troy Deeney and co beat the usual suspects into a top 5 finish?

But is it so inconceivable that Watford could finish above the likes of Arsenal, Spurs and Manchester United? In truth it probably is but it is interestingly comparable in that last season The Cherries finished only 2 places and 3 points in front of The Hornets with only a -4 GD in the favor of Eddie Howe’s men. Judging by my table they will be closely matched again this year only higher up the table. However unlikely that maybe this is a useful comparison when identifying which teams in the league are consistently  going to be your closest rivals in terms of gaining points throughout the season and I’ve only compared one season to another of which one is predicted but the trend is already (slightly) evident.

The bottom four look a fair bit more ‘believable’ in regards to their final league positions but I still think this table is slightly skewed somewhat given Cardiff and Fulham are playing in a different league after being promoted last year so their data is less reliable even though Neil Warnock’s side look nailed on in finding it tough going getting out of the relegation places.

Now we have identified a kind of mini trend in regards to comparing two team’s possible similarities in scoring and conceding goals to end up with similar goal differences and final league positions come the end of the season I now want to look at my alternative method of calculating a points finish.

xP (Expected Points)

I will start with the methodology again to briefly get a gist of what to expect and to give some clarity.

I will use Burnley as a quick example. Using an adapted Poisson distribution calculation we input Burnley’s xG per game in the first column and their xGA into the second column this then spits out the expected points per game they should achieve we then multiply that by the remaining games (36) to give us Burnley’s expected points finish from there we can assign them a final league position. This is similar to what betting agencies would use in a type of Monte-Carlo type calculation. Simple, right? My math symbols are poor at best so I will leave them up to smarter people than me but the methodology is the same regardless. On to the table.

EPL xP finish 1819

Although this is my personal preferred method of calculating a team’s expected points finish it, like any predictive model, is never going to be 100% perfect as Manchester City’s 103 xP finish in my table may show… but hang on a minute didn’t they did finish on 100 points in winning the title last year? So perhaps I am being too harsh on the model, myself and Pep Guardiola’s rampant City squad. And as with the Bournemouth/Watford goal difference situation over the last two seasons we see a bit of consistency edging in, as we should of course if the workings are correct.

Throughout the process of compiling these two sets of equation’s my mind swung back and forth in regards to one being a better method than the other and its clear to see why. There are many similarities throughout when comparing the two tables. The top three are the same and the top five in general look very familiar, 7th to 11th are fairly similar as are the bottom four.

Nuno Santo
Wolves manager Nuno Esporito Santo

One team sticks out above the rest though. Wolverhampton Wanderers.

Nuno Esporito  Santo has carried on the form and style of play that has seen his Wolves team comfortably go up as champions in the Championship last season with their infamous 3-4-3 shape. Is it possible that they could continue as they have started this season and end up in 4th spot come the end of the season? Of course if we go by the goal difference table most definitely not as they would finish 7 places below in 11th, something to ponder whichever way your opinion might sway.

Arsenal are starting to come round to Unai Emery’s way of thinking so I would expect them to finish higher than the predicted 8th but my confidence in that happening isn’t helped by the fact in both tables they are only a place or two apart and this would also, obviously, depend a fair bit on Bournemouth, Watford and Wolves faltering which they may well end up doing of course but in my opinion a top four, or even top five finish for the Gunners is not looking good either way.

unai emry
Unai Emery’s Arsenal side are showing signs of improvement but can they finish high enough come the end of the season? 

How many teams will be involved in the fight for the last of (in my opinion) the automatic champions league spot after City, Liverpool and Chelsea remains to be seen but The Gunners have a fight on their hands from some unlikely teams at this point in the season and the end result will be interesting come the 19th of May next year.

Take it or leave it (depending on who you support)

So there we have it and make of it what you will and take from it what you want but you will probably want to disregard all of it if you are a Huddersfield United, Cardiff City or a Newcastle United fan and you may want to think about crossing your fingers and toes if you’re a supporter of the teams in the red half of Manchester and North London. Only kidding (maybe).

Gareth Cooper

GC Analytics

Aston Villa- How long can the current situation last?

Villa manager Steve Bruce

In recent weeks Aston Villa manager Steve Bruce has come under increasing pressure from the Villa fans especially given that his main critics may have thought, or had hoped, he be replaced by Brentfords Dean Smith or more optimisticly former Arsenal and Barcelona striker Thierry Henry to name but two names linked with Bruce’s job.

However new owners Nassef Sawiris and Wes Edens didnt see it that way and decided that Bruce was the man to try once more to get Villa back into the Premier League after last seasons last ditch failure against Fulham in the play off final at Wembley.

So he can do it again right?

Well for us Villa fans its not looked so likely this season, so far.

First things first it shouldn’t be underestimated the impact the loanees who have not returned to Villa Park have affected the squad and Bruce’s planning.

Snodgrass returned to West Ham and has been playing regularly for the Hammers this season. John Terry has not returned for a second year and no matter how hard the club have tried it has been made clear to the former Chelsea captain that should he sign a short term contract he would not be returning on the same wages and terms as he was the year before mainly due to Villa’s on going financial issues but also this being the last year Villa will receive any parachute payments. Sam Johnstone, much to the annoyance that he could easily have stayed with Villa for another year, went to a promotion rival in West Brom, Lewis Grabban went on to pastures new and Josh Onomah returned to Spurs.

So blaming a squad lacking the quality of last year from a squad that mainly Bruce himself has built since being appointed in October 2016 is not the kind of thing fans want to hear. Neither is ‘the results will come’ or ‘the players are working hard’ or even ‘we are a big club and know where we should be’ or the most presumtious of them all, ‘we will be there and there abouts at the end of the season’, how can he be so sure? See @Myoldmansaid on Twitter for his amusing quotes for ‘Brucie Bingo’.

But the main issue that seems to get the fans up in arms, including me, is the consistency in playing some players out of position.

The benefit of being involved in football on a daily basis, especially doing tactical analysis and data work is being able to see certain things that some others may not. Tactically speaking. On the other hand what we have seen this season its pretty clear to everyone what has been the issues with team selection.

We were all thinking to ourselves that after last year we should at least be challenging for the playoffs even should we lose Snodgrass and Terry. And now given that Villa have aquired the loan signings of Chelsea striker Tammy Abraham who scored 26 goals for Bristol City the last time he played in the Championship and Yannick Bolasie Evertons long term injury returnee, Villa really should be aiming for the play offs and maybe beyond.

Bruce has a 45% win at Villa, more than anywhere else in his managerial career it should be noted.

Players out of positions

We are all familiar with Mile Jedinak being played in the centre half position alongside the ridiculous decision to let an orthadox centre back Tommy Elphick go out on loan. But other shapes and systems have been employed over the previous weeks and I wrote a post detailing what should be done a while back regarding what would be best given the squad Bruce has.

Tuanzebe constantly out at right back (he is a centre half), El Mohamady is not a right back, Grealish isnt suited out on the left and when playing centrally he has had his best games for Villa. I could go on of course but I think we all get my drift. And this is what fans just dont get about team selection and if the average fan can see its not working why cant Bruce?

Solutions?

Villa in a 3-5-2
Villa in a 3-1-4-2

I have advocated both of the shapes above in the past as i truly believe this is the best Bruce can get out of what Villa have, yes i know its three at the back but Bruce used it in pre season and then totally abandoned it when the league campaign started without any real reason as it did seem to work ok. Granted Jedinak is still one of those three centre backs but…well it is what it is.

Yannick Bolasie

Now i know that Bolasie and El Ghazi arent wing backs but if the team want to start going at their opponents from the off the 3-1-4-2 is probably the way to go. Whelan or Bjarnason will screen the back four and this shape also gets the best out of Grealish which is vital.

Of course Bruce has so many options and with Hourihane playing well he has another issue in selecting a competent but attacking midfield.

Villa in a midfield diamond

I feel like im starting to repeat myself. I would still like to see the diamond above against a lesser team (if their is such a thing) just to see how it could work. To be honest, its a stretch I know and the personnel have to be spot on but at least we have defenders in the corrct positions and two centre forwards plus a decent midfield with options to go to a 4-4-2 when defending.

Creating relationships

Villa pass map v Bristol City

Slow and sidways would be how I would describe the performance against Bristol City on Friday. As my pass map shows their was very little passing between players and the focus seemed to be get the ball up to Abraham and see what happens.

Grealish, Abraham and Kodija are shown as central here but this means the three players movement could have been across the front line especially Grealish and Kodija but overall it just didnt work did it?!

Tuanzebe isnt a ambitious full back at all and simply gave the ball to Hourihane in midfield. A full back should be an attacking option. Period.

Their was no width as Bruce abandoned the 4-1-4-1 that had seen Grealish produce his best form in previous games especially last season.

Villa in a 4-4-2 v Sheffield Wednesday

Against Sheffield Wednesday although ending up being beat 1-2 Bruce nearly got this right with one exception. Four at the back. Or the player selection of that back four (have a game of Brucie Bingo im not going there again)

At home Villa should be the dominant, attacking side and the 3-1-4-2 is the way to go given the players available for that game. This 4-4-2 is disjointed and looks nothing like a 4-4-2 should look like in any game let alone when playing at home. Fans want to see ambition and see their team put on an attacking performance especially at home but this doesnt seem to happening at the moment.

These pass maps are critical in showing the lack of tactical shape in a game and I wonder how or what some players are being told to play and what their specific roles are in a game.

Villa’s xG was 2.37 in the Wednesday game meaning they had chances that warranted at least 2 goals but again they fell short. Goals win games and Villa arent scoring enough.

The future

However long Steve Bruce remains in charge Villa fans can expect multiple shape changes and personnel in various positions and its such a shame that this is the case as their are at least 2 other set ups that would suit the squad Bruce has and thats apart from the ones mentioned above.

The indecisiveness of team selection and playing players out of position smacks of a little desperation and not knowing what is the best shape let alone the best starting eleven is worrying even at this early stage.

A new director of football has been appointed so we may see some changes now that the position has been filled and with Preston coming to Villa Park tonight I cant help but feel Bruce is on thin ice with the fans anyway so 3 points is more than a must.

Gareth Cooper

GC Analytics