Sunday, October 31, 2010

Instant Reaction: Newcastle

Anytime a team - particular one with budget options available - scores 5 goals, there is almost certainly going to be plenty of fantasy reaction. Many of us will be seriously looking at Newcastle players this week and so here are some factors to consider before pulling the trigger:
  • Newcastle have the 4th best attack this year overall and the 3rd best at home.
  • Seven players have scored for the Magpies this year but only three (Nolan, Carroll and Ameobi) have more than one.
  • Midfielders account for 10 of Newcastle's 19 goals so far.
  • Newcastle players have been awarded 38 bonus points to date, joint 3rd highest in the league behind only Chelsea (47) and United (40), and level with Arsenal.
  • Newcastle rank as the 5th best attacking team based on strength of schedule over the next 6 weeks and the next 12 weeks.
Kevin Nolan (5.7m)
I will always have a soft spot for Nolan due to the role he played during Big Sam's glorious run in charge of the Wanderers (the goal at Arsenal was particularly awesome). He is also maddeningly frustrating. His eye for goal rates him among the top midfielders in the league and he is a genuinely good finisher who can also hit them in from a distance. I don't think he always puts in enough effort for a central midfielder but he's often been deployed of a lone striker which has eased his defensive duties (interestingly though he played in a 4 man midfield today). Looking at his passing chalkboards do not inspire confidence that the goals will continue to flow but he did manage 7 shots so he's clearly willing to take a punt. Of similar priced midfielders I only really like Adam and Etherington to come close to Nolan and it would not surprise me if he led all midfielders available for under 6.5m when all is said and done.

Andy Carroll (5.5m)
Carroll burst onto the fantasy scene with a GW2 hat trick and he's been in most of minds at some point this year I'm sure. If you want to own one or two premium strikers you will need a cheap compliment and at this point I'd rank them as below, based on value for money:
  1. Carroll
  2. Odenwingie
  3. Rodallega
  4. Elmander
  5. Dembele
 I think all these players are streaky and probably only matchup plays (Carroll has 4 games of 2 points or less) but they give excellent value when the opposition is favourable. If Berbatov and/or Tevez go down hill and Torres/Rooney fail to emerge as legit options there is a reasonable argument to owning two of these boys with Drogba then picking the ultimate midfield.

Jonas Gutierrez (4.8m)
I'm a big admirer of Gutierrez's abilities and have been since his days in Mallorca. That said, I think he is a better real player than fantasy as he doesn't hit the back of the net enough (5 in 77 games for Newcastle) though he has managed 3 assists this year. He isn't a bad option but I feel Nolan is worth the extra 0.9m if you have it.

Joey Barton (4.7m)
Much like Gutierrez I think Barton is a better player in reality than for fantasy. His play makes alot of goals but he tends to play the pass that sets up the assist rather than playing the killer ball himself. The attraction of the pundits is an added bonus (sorry!) but even this has dried up of late and so I would look elsewhere for your squad men.

Shola Ameobi (4.2m)
His incredibly low price makes Ameobi worth a look, especially considering he has more goals than Torres, Rooney and Adebayor and as many as fantasy favorites like Chamakh and Anelka. However, he has just three starts on the season and Carroll didn't start one of these. I don't think Hughton will permanently go two up top and so you're starting minutes are always going to be limited. What's worse is the fact that he has seen action as a sub in every game but one, meaning your chance of a painful 1 pointer is pretty high.
If Hughton sticks with two front at home and away then Ameobi becomes a great bench guy/good rotation player but until that is proven I can't waste a bench spot on someone who might give 1 point, no matter how cheap he is.

I was fairly high on Newcastle coming into the season but I saw Nolan and Gutierrez as spot players and I wasn't on Carroll at all. I now feel that Nolan and Carroll deserve to be played every week, especially considering their favorable fixtures (even though next week's trip to Arsenal is obviously tricky). Rarely do I condone - let alone execute - an early transfer but if you want Nolan you might want to do it now as his price might rise 0.2 this week alone. My strategy is to see if either he or Lee would make my team next week anyway and if not, then I will pull the trigger on the trade I should have made 2 weeks ago.

Saturday, October 30, 2010

Commentator Sense

I am going to post a quick statistical dis-proof of all the 'common sense' nonsense the commentators spew on a weekly basis. The first is the oft-quoted rule that Drogba and Anelka can't play together. It is hard (if not impossible) to prove things like how well they interchange passes, link up and create space together but what can be proven is the ultimate result of all that work: goals.

Looking at this season only (a small but still useful sample size) we see the following data (including the Champions League and Community Shield):

Games played with Drogba and Anelka starting: 8
Goals per game in these games: 2.89
Goals by Drogba and Anelka in these games: 8 (3 for Anelka, 5 for Drogba)

Games played with either Drogba or Anelka starting: 7
Goals per game in these games: 2.29
Goals by Drogba and Anelka in these games: 7 (6 for Anelka, 1 for Drogba).

Conclusion
The pair seem to be able to play well together given that they are averaging over half a goal more per game when they do so. The only loser seems to be Anelka who has excelled when played as the focal point of the 4-5-1 this season (6 goals in 7 games). Two lessons to be learned: one, commentators rarely let statistics and facts get in the way of a convenient statement, and, two, if Drogba were to get injured Anelka might just be the most valuable fantasy player around.

Thursday, October 28, 2010

Gameweek 10 Preview

I'm putting all the design nonsense to one side to get down to some actual fantasy writing. As I'm sure most of you are chomping at the bit to get into your transfers/lineups so I will get straight down to business.

Clean Sheet Rankings
  1. Aston Villa (0.43)
  2. Arsenal (0.49)
  3. Chelsea (0.56)
  4. Man City (0.69)
  5. Everton (0.72)
  6. Bolton (0.72)
  7. Newcastle (0.72)
  8. Fulham (0.77)
  9. Man Utd (0.82)
  10. Blackburn (1.26)
  11. Wolves (1.26)
  12. Wigan (1.60)
  13. Blackpool (1.72)
  14. Liverpool (1.74)
  15. Stoke (1.93)
  16. Sunderland (1.95)
  17. Birmingham (2.05)
  18. Tottenham (2.44)
  19. West Brom (2.79)
  20. West Ham (4.53)
Captain Rankings
  1. Cesc Fabregas: Arsenal are the top rated team for the week with an expected 4.5 goals. Arsenal have scored 10 league goals with Fabregas on the pitch this season and he has scored or assisted three of them. This would suggest he should notch points this week and I wouldn't bet against him getting one of each, even after his penalty miss last week.
  2. Dimitar Berbatov: 38 points in 4 home games including 5 goals and 8 bonus points. His worst home total is 7 points. That's all I need to hear.
  3. Nani: Almost as good as Berbatov at home: 33 points in 4 games.
  4. Didier Drogba: Not a stellar matchup and he looked a bit slow last week but I can't drop him any lower than this. Played against Blackburn three times last season (twice at Ewood) and notched a goal in each game.
  5. Samir Nasri: I'm a believer. Of the 9 goals the Gunners have netted with Nasri on the pitch, he has scored or assisted 5 of them: better than Fabregas. Why is he below him? You know my unfaltering love for Cesc.
  6. Marouane Chamakh: Yeh I like Arsenal this week. All Chamakh's league goals have come at home and despite the threat of a returning Bendtner, I like him to start this one and provide a huge threat to West Ham's shaky back line (which could be without Upson too). 
  7. Florent Malouda: As with Drogba you daren't drop him too low but I don't love the fixture so I'm holding back a bit on Chelsea this week.
One week tips
These are players who I don't rank as captain options but are good sleeper picks if you need to bring someone in for a weekly play:
  • Javier Hernandez: I'm not as high on him as some but he's in hot form and appears to be a cool finisher.
  • Charlie Adam: Great fixture on paper which should see Blackpool get a couple of goals. Adam is involved directly or indirectly in most good things the Seasiders do so I'd expect bonus points with a decent chance of a goal here.
  • Kevin Davies: With Carragher pushed out to the right, the likes of Kyrgiakos and Skrtel will have to try and deal with Davies and Elmander and I'm not convinced they can. Elmander has scored more so far this year but Davies got a goal and an assist in this fixture last season and I like him here to heap more misery on poor Roy Hodgson.
I will on and off here all weekend trying to improve the usability of the site so any requests or comments are welcomed while I ring the changes. Of course, fantasy questions are also welcomed so please post away below or @plfantasy.

Please cast an eye over the proposed schedule analysis tables and let me know which one people prefer.

Defensive Schedule Tables part II

Many thanks for all the comments, I will respond to individual questions within the other post. In the meantime please take a glance at the tables below and see which one you like best. I am hoping to move the blog to a more 'website' feel with extra tables of data etc so I'm starting to see what will look good where, and of course, what will be most important. Here they are:

They all tell essentially the same story but top right and bottom left put a number (expected goals conceded) to the green/red rating. I think the top left and bottom right are cleaner so I am leaning towards including either of those as a quick look guide on the sidebar, and then put the more detailed table within a post for users to drill down into. As before, any comments/advice is welcomed.

Tuesday, October 26, 2010

Gameweek 9: Ramble On

CSIs have their crime scene. Football coaches have game film. I have some limited stats provided by premierleague.com and chalkboards. When things go wrong I like to open up the stat book, make some spreadsheets and graphs and try and desperately spot some trends or linkages that might help us predict future fantasy production better than the last 4 miserable weeks (unlike CSIs and football coaches though, I don't see CBS picking up a show focusing on me sat in my living room on my laptop anytime soon. Wait, what the hell am I saying, they picked up Shit My Dad Says. Perhaps I can get Mark Zuckerberg to play me).

Clean Sheets
Something that has frustrated and intrigued me so far this season is the lack of clean sheets and the seeming randomness in which they occur. To the first point, there have been 7 more clean sheets through GW 9 than last season. In fact, the weekly distribution is uncannily similar so far:


As for consistency, 8 teams account for 30 of the 47 clean sheets to date, though only Chelsea (7) have been truly dominant in this category. Of the teams with at least 3 clean sheets only Sunderland and Wigan look surprising (the others are AST, BIR, CHE, EVE, MNC and MNU) so perhaps our complaints about clean sheets this season are unfounded. I think it's time to just accept that defenders (away from Stamford Bridge) will never compete on a points-by-points basis with other positions and so the focal point for them should be value for money, or aPPMS as I like to use.


I have alluded to this before but I am now convinced that the best strategy (until something changes the level of scoring in the league, such as poor weather in winter) going forward is to put four budget defenders alongside a Chelsea defender (preferably a budget option like Bosingwa if he keeps his place in the first team) and start 2 of your 4 options each week in a back three. This should allow for you to have sufficient cash to pickup 5 or even 6 'every week' starters in your front 8 meaning you'll only need to pick the odd spot starter each week.


I need to research which teams' schedules gel with each other but if I can find a suitable group of players that fit together, I might use my wildcard to make it happen this week.


Bonus Points
Bonus points are somewhat of a mystery and their award can baffle you even having watched the whole game. Some players have long been 'bonus magnets' but I wondered if I could shed some light on how points are awarded which might give us some useful information when picking our squad members (this is less useful for stars as it doesn't need printing that Drogba and Fabregas will attract bonus points galore).

First let's look at the bonus points scored by each team so far this season:
  1. Chelsea (45)
  2. Arsenal (35)
  3. Sunderland (34)
  4. Man City (34)
  5. Man Utd (34)
  6. Tottenham (32)
  7. Newcastle (32)
  8. Everton (31)
  9. Bolton (28)
  10. Blackpool (27)
  11. West Brom (27)
  12. Wigan (27)
  13. Birmingham (24)
  14. Fulham (22)
  15. Aston Villa (22)
  16. Liverpool (19)
  17. Blackburn (19)
  18. West Ham (17)
  19. Stoke (17)
  20. Wolves (14)
Most of this list is as you would expect (Chelsea pretty much dominate all categories so far) but a couple of team stand out. Firstly, Sunderland stand out as excellent value given that the majority of their players fall into the cheaper half of players available. With three straight clean sheets and four on the season El Mohamady and co have made me wake up and take notice and I am officially on the bandwagon from week 14 onwards (they face @New, Sto, @Tot and @Che before then). Secondly, this table shows just how far Liverpool have fallen as not even the loyal bonus judges are handing out freebies these days (they have only received more than 3 bonus points for a game twice this season).


It's also interesting to note that when some teams win they tend to receive the full 6 bonus points while others still can't get any love. Arsenal, for example, have received 6 points on all 5 occasions they have won while Stoke were handed just 1 point when they beat Newcastle in GW6. This clearly ties into the obvious perceptions of these two teams and it is worth considering when you go to pickup a player from a less 'sexy' team. The teams who have been given the maximum 6 points for each win they have secured are as below (minimum two wins):
  • Arsenal
  • Birmingham
  • Blackpool
  • Bolton
  • Everton
  • Newcastle
  • Wigan
Perhaps this group of teams were mis-labelled as 'sexy' in the previous paragraph as Birmingham, Everton and Wigan all rank in the lower half of the table in goals scored. Perhaps it is more useful to look at the teams who seem to be unrewarded even when they win:
  1. Fulham (3.0 bonus points per win)
  2. Stoke (3.3)
  3. Aston Villa (4.0)
  4. Liverpool (4.5)
  5. West Brom (4.8)
I'm not saying that players from these teams are worthless but be careful when looking at a player like Richard Dunne who was greatly helped by bonus points last year. Red flags could also therefore be raised for Shawcross, Gerrard or Torres.


aPPMS difference

A quick update on aPPMS difference which highlights the difference between a players PPMS and aPMMS. A high value will indicate that the player has faced an unfavorable schedule to date and should (if all other things remain equal) be due for an upturn in fantasy production:
  1. Tim Cahill
  2. Mikel Arteta
  3. Scott Carson
  4. Matthew Gilks
  5. Gael Clichy
  6. Luke Varney
  7. Charlie Adam
  8. Charles N'Zogbia
  9. Ahmed El Mohamady
  10. Peter Odenwingie
And here are the worst:
  1. Barry Fergsuon
  2. Florent Malouda
  3. Michael Essien
  4. Andy Carroll
  5. Peter Cech
  6. Didier Drogba
  7. Rafael van der Vaart
  8. Sebastien Larsson
  9. Kevin Nolan
  10. Ashley Cole
I am not necessarily saying you should sell these guys (in fact I am definitely NOT saying sell Drogba) but if you are looking to differentiate your team and you don't have much cash to lose selling someone like Malouda it isn't the worst idea in the world to trade him for, say, Nani and play the fixtures a bit (if you dare).


I'll hopefully get a chance midweek to throw a few more ramblings out there, along with many more tweets now I'm up and running with my new iPhone. Let the good fantasy times role.

Thursday, October 21, 2010

Ramble On / Gameweek 9 preview

I hope my previous post wasn't too downbeat as that was not the intention. It was more an expression of frustration in being unable to quantify where the fantasy success is coming from, even with the benefit of hindsight. For now though, let's put the stats aside and have our familiar (if abbreviated) ramble around the league.

Clean Sheets
Okay so I won't totally depart from the stats. Only 5 clean sheets this week, four of which came in two 0-0 stalemates at Villa and Ewood Park. I loved Blackburn to get the clean sheet this week but Sunderland were somewhat of a surprise given their torrid away form over the past 18 months. I did think that Villa might get a draw against Chelsea but I backed the 1-1 as suggested by my prediction spreadsheet.

So far we've had 43 clean sheets, only topping 6 for a week once (in a cagey week 1). I have suggested in the past that budget defenders might be the way to go as there is little historic correlation between cost and points and this seems to hold true this season, perhaps more so than ever. If, for example I was to select a backline of budget options like Gilks, Mignolet, El Mohamady, Bramble, Alcaraz, Jones and Coleman I would have sufficient cash to pick the fairly insane front 8 of Fabregas, Malouda, Arteta, Nani, Adam, Drogba, Berbatov and Odenwingie. That gives you six every-week starters meaning you need to start 4 players from your back line, Adam and Odenwingie. If you can play the fixtures right then this could be a real recipe for success.

I am seriously considering playing my wildcard to take this strategy now and I will be posting which teams' back lines complement others' next week. I might make one slight tweak and keep Alex thus having to downgrade one of my midfielders (probably Arteta). In truth this isn't wildly different to what I have now except I am overpaying for Vidic's services (0.503 PPMS and 0.453 aPPMS) based on his current form. He faces the chop this week I think.

The Big Four
How the might have fallen. It's worrying for fantasy purposes when a traditionally great team struggles to wins games, tussles with ownership and faces the prospect of losing their best players. And things are even worse for Liverpool. Who can we trust in this United side? Nani and Berbatov are in the form of their life yet they don't have the defense nor support strikers to capitalize. Fantasy wise I feel fairly comfortable with this pair's prospects with or without Rooney but it does damage them somewhat knowing that teams may opt to stick extra men on them if the likes of Macheda, Hernandez or Owen (remember him?) are drafted in to pick up the slack.

Defensively I am terrfied for United as while they weren't outplayed by West Brom by a long shot, there were periods of that game where WBA held possession very well and looked dangerous. The manner of the conceded goals was fairly shambolic and Van der Sar looks like he needs to retire now rather than at the end of the season. I would expect O'Shea to be back in the side after Rafael failed to impress and he still represents solid value. As for his pricier teammates though, the jury is out.

Gameweek 9 Preview

Clean Sheet Rankings
(After last week's questionable rankings I have switched to using a new weighting of current and prior year data. I now base the rankings using 2/3 of this year's data and 1/3 prior year).
  1. Chelsea
  2. Tottenham
  3. West Brom
  4. Man City
  5. Sunderland
  6. Birmingham
  7. Liverpool
  8. Everton
  9. Bolton
  10. West Ham
  11. Blackpool
  12. Newcastle
  13. Fulham
  14. Stoke
  15. Man Utd
  16. Blackburn
  17. Arsenal
  18. Aston Villa
  19. Wigan
  20. Wolves
Captain Rankings: Team goals
  1. Chelsea (5.04)
  2. Bolton (2.30)
  3. Sunderland (1.92)
  4. Man City (1.76)
  5. Liverpool (1.68)
  6. Stoke (1.39)
  7. Man Utd (1.33)
  8. West Brom (1.32)
  9. West Ham (1.27)
  10. Birmingham (1.07)
  11. Newcastle (1.02)
  12. Wigan (0.93)
  13. Tottenham (0.92)
  14. Blackburn (0.78)
  15. Blackpool (0.77)
  16. Aston Villa (0.64)
  17. Arsenal (0.63)
  18. Fulham (0.57)
  19. Everton (0.37)
  20. Wolves (0.14)
Captain Picks
  1. Didier Drogba
  2. Florent Malouda
  3. Ashley Cole
  4. Carlos Tevez
  5. Darren Bent
  6. Dimitar Berbatov
  7. Nani
Not a great deal of point going beyond the top 3 this week as you simply must captain a Chelsea player and if you have him, Drogba. Drogba has netted in every home game this season in the league including an opening day hat trick against West Brom (who average less goals per game conceded than Wolves on the road if you discount that Chelsea result).

Apologies for the condensed write up this week, I hope to back on the weekend with a write up of the action as well as a couple of features next week. As always thanks for reading and I encourage you to check out my twitter feed or post comments below.

Wednesday, October 20, 2010

Where's the value?

I really don't know what to make of this season. The increased parity (by EPL standards) we are witnessing is great for competition but is wreaking havoc with the fantasy game. My top 7 ranked teams this week were about as good a clean sheet bunch as you're going to get on paper yet only Blackburn (2nd) and Everton (7th) managed to live up to their ranking with United (1st) and Newcastle (3rd) each conceding two goals at home.

At the start of the season I wrote a piece about the correlation between points and cost which was strongest among forwards and hence we should focus our attention on the big name strikers. Meanwhile bargains at the defensive positions, and to an extent the midfield, meant that value can be found more easily here and hence it was less important to spend big on premium players. 8 weeks into the new season and the correlation between cost and points per 90 minutes is as below:
  • Goalkeeper 18%
  • Defender 42%
  • Midfield 41%
  • Forward 36%
What does this mean? It's a freaking crap shoot! We are basically saying that all that glitters is not gold (goal?) and that you cannot simply plug in a player like Gerrard, Torres, Rooney or one of the countless other flops on the basis that things will even out and he'll 'come good'. With that in mind, what else might be a good indicator of success? Strength of schedule played to date?
  • Goalkeeper 24%
  • Defender 28%
  • Midfielder 19%
  • Forward 15%
As expected, this is a better measure of success for keepers as no matter how good a player you are and who you play for, you're better trying to stop Marlon Harewood scoring than Didier Drogba (groundbreaking stuff I know). For outfield players though the correlation is even weaker which is again what we'd expect as good fixtures can only help you do a degree, but this doesn't help us select our team. A good schedule merely enhances a player's prospects rather than makes them.

Well if nothing else, last's year success is at least a good starting point, right . . .
  • Goalkeepers 3%
  • Defenders 33%
  • Midfield 58%
  • Forwards 54%
Keepers are basically a coin flip and this is why I never pay for them. The midfield and forwards show at least some signs of logic but still nothing we can overly rely on.

This is where I bring it all together and tell you the magic factor to predict future value, right? Unfortunately not. There is no magic bullet. Clearly something is going right as the worldwide standings will attest but I can't tie it all together and come up with a strategy to exploit this crazy season right now. So it's back to the drawing board to try and figure out a way to explain the scores seen to date so far. Stay tuned for an attempt at an answer by the weekend.

Wednesday, October 13, 2010

Gameweek 8 Preview

Boy those international weeks drag. It seems like months ago that Alex wrapped up the week with that goal but fear not as we return with some tasty fixtures for both fantasy and real reasons. The Merseyside derby looks even more interesting than usual thanks to the takeover drama at Anfield while home games against West Brom and Birmingham for United and Arsenal suggest goals a plenty for the big boys this week. You never can be sure though so hopefully the below stats and rankings can at least give us some clues who to back this week.

Defensive Rankings
(Based on current and prior year stats. Current year only rank in parenthesis)
  1. Man Utd (4)
  2. Blackburn (6)
  3. Newcastle (5)
  4. Bolton (11)
  5. Arsenal (15)
  6. Wolves (2)
  7. Everton (10)
  8. Fulham (6)
  9. Chelsea (9)
  10. West Ham (18)
  11. Aston Villa (6)
  12. Man City (3)
  13. Stoke (15)
  14. Tottenham (12)
  15. Liverpool (17)
  16. Blackpool (13)
  17. Sunderland (14)
  18. Wigan (1)
  19. Birmingham (20)
  20. West Brom (19)
With Chelsea facing a tough trip (on paper) to Villa Park you have to love United players this week as the best bet for a clean sheet by some distance. The ranking of Wigan as 1st based on current year data is why I am still using the hybrid current/prior rankings and need to wait a few weeks before this year's data is reliable. For the record the reason for this anomaly is because they have played just two road games this year and haven't conceded in either.

Long term defensive rankings (SoS6)
If you are thinking of bringing in a new defender this week you might want to check out the below rankings for who has the fewest predicted goals conceded over the coming 6 weeks. The number of games with less than a goal predicted is in brackets:
  1. Chelsea (5)
  2. Man Utd (4)
  3. Liverpool (2)
  4. Arsenal (4)
  5. Man City (2)
  6. Aston Villa (3)
  7. Everton (3)
  8. Newcastle (4)
  9. Tottenham (3)
  10. Stoke (2)
  11. Bolton (1)
  12. Fulham (3)
  13. Birmingham (3)
  14. West Ham (0)
  15. Blackburn (2)
  16. Blackpool (0)
  17. Wigan (1)
  18. West Brom (1)
  19. Wolves (2)
  20. Sunderland (2)
Captain Rankings: Team goals
I'm changing this section up a bit this week to show the team predicted goals upon which I base my captain picks. My captain picks are generally not as good as the clean sheet rankings as I am still using some judgment rather than pure facts to make my predictions. I try and incorporate form and prior success against teams into the picks but in the end it is still just my opinion. I am therefore giving you the scoring stats so you make up your own mind too:
  1. Man Utd (4.22)
  2. Arsenal (3.60)
  3. Newcastle (3.18)
  4. Blackburn (1.86)
  5. Man City (1.70)
  6. Everton (1.48)
  7. Fulham (1.45)
  8. Bolton (1.40)
  9. Blackpool (1.13)
  10. Chelsea (1.09)
  11. Wolves (1.09)
  12. Aston Villa (0.99)
  13. Tottenham (0.79)
  14. Liverpool (0.76)
  15. West Ham (0.68)
  16. Birmingham (0.67)
  17. Stoke (0.67)
  18. Wigan (0.62)
  19. Sunderland (0.58)
  20. West Brom (0.46)
Captain Rankings: Individuals
  1. Cesc Fabregas: I know it's hasty to throw him straight in at number one but would you bet against the maestro to mark his return with a bang in a favourable fixture against Birmingham at the Emirates?
  2. Dimitar Berbatov: Gets the nod ahead of Nani based on the fact that West Brom have already conceded 6 goals to forward this year (only Wigan and Liverpool are worse)
  3. Nani: Great option against West Brom in what could be a statement game for United.
  4. Didier Drogba: Despite the tough fixture it's hard to even rank him this low never mind any lower. For the conservative players out there I would not blame you for captaining him as usual this week.
  5. Nemanja Vidic: As close to a certain clean sheet as it gets. I've also been impressed with Vidic's opportunities from set pieces this year which is why he gets the nod here.
  6. Carlos Tevez: I'd be terrified if I was a City fan after continuing talk of retirement but that won't happen before a trip to Bloomfield Road who will see probably the best player to grace the pitch since Sir Stanley hung up the boots (apologies to Pure Juice alum Chaz Adam but you're not quite there yet!)
  7. Wayne Rooney: I've probably been as critical as anyone of Rooney and he hasn't appeared in these rankings for a while but a game at home to West Brom is a great opportunity for him to start to rebuild his shattered season. Penalty duties always help the English hitman and he claims to be fully fit.
  8. Mikel Arteta: I think Arteta is ready to explode after an unspectacular start to the season. A game against a Liverpool side in crisis might be just the ticket.
  9. Florent Malouda: As with Drogba it's tough to rank him this low after his dynamite form but the fixture is tough and Malouda has laid two eggs (GW4 and GW6) on the road this year already. He did however destroy Wigan in GW2 so it's somewhat risky to drop him this low.
  10. Marouane Chamakh: Chamakh has delivered decent returns this year and makes a good fixture play this week at home to Birmingham.
I hope you're as excited as I am to get back to Premier League action this week, especially as for me it's the first uninterrupted Saturday in weeks so I can go with wall-to-wall soccer. Bliss.

Monday, October 11, 2010

PLFantasy Glossary

Given my reliance on stats I sometimes refer to various formulae I have devised to measure success in the fantasy game. In case I ever forget to define anything in an article, have a look below for a quick explanation of the jargon used:

aPoints A predictive tool to try and suggest how many points a player has earned compared with others if the schedule to date had been equal. In theory, the player leading this category should be the best player from here on if he maintains the same form.

aPPG Adjusted points per game. Calculated by taking the standard PPG calculation and adjusting it based on the SoS faced by a team to date. This is calculated by comparing the average goals per game of a team's opponents and the average goals per game throughout the league to generate a factor. This factor is then applied to the standard PPG to give a schedule adjusted number.

aPPMS Adjusted points per million spent. Calculated by taking the standard PPMS calculation and adjusting it based on the SoS faced by a team to date. This is calculated by comparing the average goals per game of a team's opponents and the average goals per game throughout the league to generate a factor. This factor is then applied to the standard PPMS to give a schedule adjusted number.

DGW Double gameweek. Occurs when a team faces two opponents in a single gameweek.

Differentiator A player who performs well despite not being widely held by other managers. Essential for any team trying to extend their lead or catch their opponent.

Green light game A game in which a player has a favorable matchup. Specifically, when a defender faces an opponent who score less than a goal a game or an attacker faces an opponent who concede more than 1.5 goals per game.

GW22 Gameweek 22

PPMS Points per million spent. Calculated as PPG divided by a player's cost at today's date. Designed to try and assess a player's relative value rather than just his points production. A rate of 0.658 is required to achieve a 2,000 season (see Value Line below).

PPMS diff The difference between standard PPMS and aPPMS. A player with a high value has played tough fixtures to date and should benefit from increased returns in the future.

PPG Points per game. Simply calculated as total points scored divided by the number of appearances made.

Red light game A game in which a player has an unfavorable matchup. Specifically, when a defender faces an opponent who score more than 1.5 goals per game or an attacker faces an opponent who concede less than 1.0 goals per game.

SoS6 The strength of schedule for a player/team over the next 6 weeks. When discussing an attacker this will be based on the opposition's defensive rankings while for a defender it will be based on the opposition's attacking prowess. The 6 refers to the number of weeks over which the strength of schedule is taken from (normally either 1, 6 or 12).

Value line The point which players need to generate sufficient points to help you to a 2,000 season. This equates to 53 points per week, 4.8 PPG for an individual player and a PPMS of 0.658 (based on the assumption that you will spend 80m on your starters and 20m on your bench).

xDF The points scored by a defender that are exclusively due to defensive performance (appearance, clean sheet and bonus points). Used to predict long term returns as assist and goals by defenders are generally hard to forecast.

Sunday, October 10, 2010

Play your card right

I'm not sure there are many things in the fantasy world I enjoy more than tinkering with my team and seeing what I can get by using my wildcard. In previous years I have been forced to use my wildcard early but this year my team is still fairly strong so I could delay if I want to. However, there are pros and cons to holding onto your wildcard:

Pros
  • You can use the wildcard at a later date to take advantage of double gameweeks or if your team falls prey to numerous injuries.
  • You can play your wildcard towards the end of the season when fantasy teams tend to get stale but real teams can hit runs of good form.
  • We are still not totally sure which teams and players are for real and so using your wildcard now might result in you bringing in a load of in-form but ultimately overpriced lemons.
  • With the other wildcard in your back pocket, you can select a team now based purely on the next 15 gameweeks and then redo your team again during the transfer window.
Cons
  • If you fall too far behind in your mini league it may be too late to make up ground later in the year.
  • You have an extra wildcard this year that can only be used between GW23 and GW27 (early January) so unless you want to use two wildcards in a close period of time it might make sense to use it now.
  • If you made a few bad choices in your original squad or have suffered a few injuries (or both!) you will chip away at your score with too many 4 point hits. You might be better just biting the bullet and using the wildcard now. In 2009/09 I managed to finish the season in 7,734th place in the competition despite using my wildcard in GW2 after picking a terrible team to start the season.
Who to target?
If you decide to roll the dice with your wildcard I have a few words of advice:
  1. Don't get too cute: Didier Drogba leads all players in fantasy points for a reason - he is awesome. Yes, every man and his dog owns him, but again, it's for good reason. It might not thrill you to pickup players owned by alot of other players but you won't find 11 differentiators out there. Play it safe to an extent.
  2. Don't chase points: This seems to contradict point one and I agree the distinction is subtle but it is one you must master if using your wildcard this early. This means caution should be exercised before loading up on the likes of Alcaraz, El Mohamady, Kalou or Elmander. I am currently doing some research to try and put a value on 'form' (ie the likelihood of a player with a good score last week having another good game) but form will only help you for 4-8 weeks or so. Class, as they say, is permanent and it is that reliability that you need if you are locking yourself into a lineup for at least 15 weeks or so.
  3. Play the fixtures: This is always key for your team but its worth repeating here. If a player has played easy games and your about to pick him up while he plays the other teams then you might get significantly worse value than you think. This is reflected in my aPPMS rating which I discuss in my weekly preview columns.
On the above point, the rankings for the schedules played to date are below (hardest to easiest):

Defensive fixture rating to date
  1. West Brom
  2. Blackpool
  3. Arsenal
  4. Sunderland
  5. Bolton
  6. West Ham
  7. Liverpool
  8. Stoke
  9. Newcastle
  10. Wigan
  11. Blackburn
  12. Fulham
  13. Man City
  14. Everton
  15. Tottenham
  16. Man Utd
  17. Chelsea
  18. Wolves
  19. Birmingham
  20. Aston Villa
 Attacking fixture rating to date
  1. Everton
  2. West Ham
  3. Wigan
  4. Bolton
  5. Wolves
  6. Man City
  7. Blackburn
  8. Sunderland
  9. Man Utd
  10. Newcastle
  11. Blackpool
  12. West Brom
  13. Aston Villa
  14. Stoke
  15. Fulham
  16. Arsenal
  17. Liverpool
  18. Birmingham
  19. Tottenham
  20. Chelsea
Note that these rankings are still somewhat distorted by wild results as if Chelsea smash a team 7-0 then suddenly their fixtures seem easier so the big score becomes a self-profiling prophecy. That said, I am still using data from both current and prior season so I am relatively comfortable relying on the numbers for guidance.

These rankings can play a role when picking transfer targets as someone like Mikel Arteta, who has perhaps underwhelmed this season, looks slightly more attractive when you consider the teams he has faced. Indeed the difference between his PPMS and his aPPMS is 0.148 which is 4th among all the players in the league that I track. In fact my new aPoints preditive tool (post to follow soon) suggests that his real total to date should be 39 points which ranks joint 5th among midfielders (Nani, Cahill, Malouda, Van der Vaat and Milner) and 12th overall.

Likewise, while Birmingham have been a bit disappointing to date, it might get a whole lot worse given the ease of their defensive fixtures to date.

I think I am going to hold off on my wildcard this week but feel free to post your proposed teams below and we can discuss the merits of using the WC this week.

Friday, October 8, 2010

Fabregas v Malouda

I'm in a dangerous position. I'm happy with my team but two back-to-back average weeks (which were an Alex wonder goal away from being poor) have me feeling antsy. It's like when you're sat in a good bar on a Saturday night. You're happy. The beer is good. And the atmosphere is decent. But, you can't help wonder what else is available. Sure you're gonna have fun, but is it going to memorable? I think my team if fine. But is is memorable?

There has been talk on the boards of wildcards but I've decided that with Drogba, Berbatov, Nani, Alex being locks to stay put and the fact I'm happy with Vidic, Odenwingie and Adam, there doesn't seem to be much point wasting my wildcard too soon. Hence I am looking to reload rather than re-build my squad.

The obvious (and likely) choice is to switch Koncheky out for Boateng who I am very high on. I could also bring Boateng in for one of my budget defenders given that I have 2.8m sat in the bank. However, I can't help but be attracted to the return of Fabregas after the international break given the way he dominates games, wins bonus points and, well, does everything. My only realistic option (without wholesale changes) to get Fabregas is to drop Malouda: surely this is a crazy move?

Fabregas v Malouda

Last season Fabregas was simply irresistible. Despite Lampard running away with the scoring title in the end, it was only through injury that Fabregas missed out (he outscored Lampard at 8.8 v 7.9 PPG and at a rate of 0.752 v 0.584 PPMS). As a comparison, Malouda (on tremendous form) is generating 7.7 PPG at a rate of 0.779 so far this season. So, Fabregas generated more points but Malouda was slightly better value. Nothing too controversial there. For reference, on a per minute basis, Fabregas is generating 10.0 PPG this season in his brief cameo role in GW2 -5.

My issue is whether Malouda can continue this rate of scoring (which is still below the level of production Fabregas showed last year). To date Chelsea have played the easiest schedule for attackers while Arsenal have faced the 5th hardest. That would suggest that Chelsea's goal record (23 in 7 games) must regress a bit to the mean while Arsenal will surely play better against weaker opponents and with Fabregas back (16 in 7 games to date). Over the next 8 games the stats show that Chelsea's opponents concede an average of just 1.18 GPG while Arsenal's opponents concede 1.36. My prediction tool suggests that over those 8 games Chelsea will score 2.5 GPG while Arsenal will add 2.25 GPG.

So it seems that Chelsea might shade the Gunners in goals scored but it is closer than it has been so far this season. But who accounts for those goals?
  • This year Malouda has scored/assisted 6 of Chelsea's 20 goals when he played (30%). 
  • Last year Fabregas accounted for 30/66 of the Arsenal goals that were scored when he played (45%).
  • Malouda accounted for 21/102 (21%) of the goals Chelsea scored when he played. 
Those numbers are quite telling as even a continuation on this pace would not see Malouda score/assist as many goals as Fabregas.

A closer look at Malouda
We all know what Fabregas can do. He is for real and a genuine star in this league (in my opinion the best player along with Drogba). The jury on Malouda however is still out to a degree. Now in his fourth season at Stamford Bridge, Malouda is really starting to look like the player Chelsea thought they were getting when they raided Lyon back in 2007 (he also showed glimpses last season). A look at his chalkboard's from the start of this season are somewhat confusing as his involvement in the game does not necessarily tie to his production. His greatest involvement in terms of touches of the ball and completed passes came against City and Stoke and he managed a combined 12 points in the two games. Meanwhile his chalkboards against Arsenal and Blackburn look terrible but he again managed 12 points over the two games.

That production is both good and bad. Good in that he can score and produce fantasy points when not playing that well but bad in that he doesn't dominate games the way Fabregas does and may drift out of form at some point this season. The longest stretch of games Fabregas went last season without scoring was four, Malouda meanwhile had a run of six games (though this only occurred once and he was generally pretty consistent all year). You do feel though that of the pair Malouda is more susceptible to off weeks as Fabregas is often bailed out by over generous bonus judges. Fabregas earned 1.69 bonus points per 90 minutes last season while Malouda managed just 0.78. Consider that Lampard's number was 1.03 and Drogba's was 1.01 and you realize just how much the pundits love Fabregas (only Rooney - remember him? - came close at 1.53).

The biggest factors playing against me dumping Malouda is his current ownership (42%) and the 0.4m I will lose by selling him. Goals for Malouda will crush my team across all leagues if Fabregas takes a while to get back to form thus making it a huge risk not to own him. Also, if FabregasMalouda back, though it's that kind of risk that makes teams champions.

As of now I am leaning towards a double switch (which I usually avoid) of Konchesky and Malouda to Boateng and Fabregas. I am justifying the 4 point hit on the basis that I will captain Fabregas and both players have good fixtures this week. I also would like to free up some cash to bring Arteta in (who I think is about to explode) and might have to move Vidic out for O'Shea to accommodate that move in the future. I see from the poll that the majority of you think Fabregas will outscore Malouda in the near future and I think I agree with that sentiment. This decision will surely take the rest of the week to settle in my head.

Thursday, October 7, 2010

I repeat: Buy Boateng

The stars are aligning in City's first team and for Boateng's fantasy prospects:

"Without flying full backs like Boateng and Kolarov, who can push forward, I’ve had to adjust the team to get results and stay in touch with the leaders. But only until everyone is back and fit.”

The positive signs here are two fold:
  1. Being specifically mentioned by your manager as an important first team player is always a big deal and this ends any worries I had that Boateng might share time at the right back spot.
  2. Being described as a 'flying' full back suggests that Boateng (and Kolarov) will have plenty of freedom and will be allowed to attack the opposition regularly. Boateng's game against Newcastle does not really suggest any more freedom than Richards had but we will see if Mancini opens up the team more when everyone is fit.
A further boon to Boateng's value is City's relatively kind fixture list which, while seeing City entertain both Arsenal and United in the next 5 weeks, also sees them face no other big five* side until week 22. There toughest away game until that point is @Fulham which isn't exactly terrifying. Based on current and prior year performances, City rank 5th in my short/mid term defensive rankings.

At 5.8m Boateng will be 0.5m cheaper than Toure, who is the best City prospect so far this year. I think Boateng will become a mainstay in this side, and provide better attacking options than Toure. I feel that he has top 5 defender potential and would put him against anyone from here on in who doesn't play for Chelsea or United.

Perhaps the strongest recommendation I can give him is that I will be buying him myself as soon as I work out how to use my transfers this week.

*For those keeping score at home the big five currently consists of Chelsea, United, Arsenal, Spurs and City. Sorry Liverpool but Theo Epstein and a juiced Big Papi can't save this team too.

Tuesday, October 5, 2010

Alex injury

It seems that Alex picked up an injury at the end of the Arsenal game and now looks set to miss 3 weeks of action. Before you press the panic button I thought I should lay down a few words of caution.

1. Alex has been sensational value this season and is third among all defenders I track in aPPMS. With the other two being El Mohamady (who is overly reliant on attacking points) and Salcido (who has only played 3 games) I would not bet against Alex leading the lot when the season is done.

2. Three weeks sounds bad but it could result in him missing just one EPL start - at Villa in GW8. This isn't a great defensive fixture anyway and so if your squad works out well you may not even lose that much. They also play Wolves at home (a very good defensive fixture) in 18 days so if Alex missed that one it would be a blow.

3. Alex has seen his price rise by 0.1m already this week after his big week so only a massive fire sale will cause his value to fall.

My advice: stick with him. Concentrate on improving your bench around him to plug the gap for one or two gameweeks and get ready to welcome him back soon as Chelsea's defensive fixtures are great until GW15.

Monday, October 4, 2010

Gameweek 7: Ramble On

We're seven weeks in and we're really starting to get a good idea on who we can rely and what trends are for real. I thought - in honor of the title of this weekly post - I would employ the help of Led Zeppelin to help explain the season so far:

"Been dazed and confused for so long it's not true"
Though the top five has a familiar look, the rest of pack appears upside down and fantasy value has been scattered throughout the league with minnows like Blackpool and West Brom providing good fantasy picks while teams like Liverpool and Everton have struggled to contribute. One of the lessons as always is to play the fixtures, which explain why Everton's midfield and forwards (hardest attacking schedule to date) have struggled while Birmingham's defense (2nd easiest defensive schedule) have provided some value. More confusing is the fact that West Brom have conceded 6 in 6 since GW2 while facing one of the hardest schedules around while Spurs have managed just 8 goals despite playing one of the easiest. To try and ease the confusion a bit, here are the rankings for the next 6 weeks of expected goals conceded by each team:

Defensive rankings
(ranking based on this year and last, this year only rank in parenthesis)
  1. Chelsea (1)
  2. Man Utd (10)
  3. Liverpool (16)
  4. Arsenal (7)
  5. Man City (2)
  6. Aston Villa (6)
  7. Everton (5)
  8. Newcastle (14)
  9. Tottenham (3)
  10. Stoke (11)
  11. Bolton (8)
  12. Fulham (9)
  13. Birmingham (19)
  14. West Ham (18)
  15. Blackburn (12)
  16. Blackpool (13)
  17. Wigan (4)
  18. West Brom (20)
  19. Wolves (17)
  20. Sunderland (15)
"I don't know what it is that I like about you, but I like it a lot"
Actually, that's a lie, I know exactly what it is I like about Didier Drogba: everything. Drogba has elevated his game to a new level to put him alongside the like of Henry, Shearer and Ronaldo as one of the all time greats of this league and the game as a whole. Drogba has been sensational for Chelsea and his fantasy owners alike and despite some red flags (he has faced the easiest attacking schedule to date and faces the 5th hardest over the next 6 weeks) he is a must own player and is close to must captain status. The only way I could see a team being successful and not owning him would be if they chose to back three other premium strikers (say, Tevez, Berbatov and Bent) and used the freed up cash to upgrade to a top midfield player who catches fire at the right time (perhaps Fabregas on his return). Still this is purely conjecture while Drogba's success and value are lodged in cold, hard facts.

"Since I've Been Loving You. I'm about to lose my worried mind, oh, yeah"
Drogba's opponent this week was Arsene Wenger's frustrating side who once again failed to deliver in a big game in which they enjoyed plenty of possession and chances. I don't know where Arsenal go from here and until Fabregas returns I don't think I want any part of them. Yes, Nasri has shown glimpses in his four appearances but his ceiling is below other similarly priced midfielders (Van der Vaat, Bale, Nani, Arteta) and his consistency is a concern. Other than Nasri no other Arsenal player has cracked the aPPMS success line (defined as 0.658 based on the value needed to achieve a 2000 point season) with Arshavin being luxury purchase not pulling his weight (0.503 aPPMS). When Fabregas returns (likely after the international break) things could change but only a almost complete reversal of cutting edge could make this team deliver fantasy value to loyal owners.

"Didn't take too long 'fore I found out, what people mean my down and out"
Even before he limped off, I had decided that this was Torres' last week to earn his place in the captain rankings each week and though it's tough to penalise him for injury, he clearly isn't playing like an elite player right now. Liverpool are pretty awful at the moment and deserve their place in the relegation zone. If you want a hint of Liverpool's collapse look no further than my defensive rankings above. Liverpool are ranked as the third best unit based on this and last year's results, but based on this year only they are a woeful 16th (three places below the Blackpool team who beat them this week.) Given that their defenders cost 6m+, they are poor value and only Konchesky (5.1m) is justifiable (though I am selling him too).
Their midfield and forwards are even worse with Gerrard (0.389 aPPMS), Kuyt (0.373) and Torres (0.309) representing some of the worst value around. How Gerrard is owned by 16% players is beyond me and is surely based on his reputation and the 2 or 3 year old strategy of holding two of the 'big three' midfielders (Lampard, Gerrard and Ronaldo). 
As I said, there is too much talent here to give up forever but this team needs drastic changes (and Joe Cole is unequivocally not the answer) before I will even look at them again.

"The Song Remains The Same"
I already touched on Drogba above but the Chelsea defense deserves some love of their own. The top three fantasy defenders all ply their trade at Stamford Bridge (and the right back would also be in there if not for the rotation of Ferreira and IvanovicVidic is bigger than the difference between Malouda and Nani (given that you can only own 3 Chelsea players and I've already said you need Drogba). As I toy with the idea of playing my wildcard, I start thinking that Cole and Alex looks like a pretty powerful combo right now. And that leads to me to the final word from Plant and Page . . .

"Yes, there are two paths you can go by, but in the long run, there's still time to change the road you're on"
When to play your wildcard, possibly one of the most contemplated fantasy questions around. For the first time in 3 seasons this decision lays in my hands having not been forced into squandering it within 5 weeks due to too many poor decision and unlucky breaks. My team have done okay so far but the last couple of weeks have left me feeling a bit underwhelmed and my team lacked the flexibility to bring in a couple of players I wanted who have gone on to have success (Van der Vaat). For those interested in using their wildcard, the two week break is a decent time as you can buy players now whose price is likely to rise (Van der Vaat) and potentially make a profit within the window even if you don't decide to keep them.

Week 7 stats
I was delighted by the positive response to the PPMS/aPPMS numbers so I've included the leaders below along with some other random trends (note that I don't track every players' data so there could be the odd anomalous player with a better PPMS due to him playing one game etc)


aPPMS rankings (non GKs)
  1. Odenwingie (1.261)
  2. Elmander (1.167)
  3. Adam (1.083)
  4. El Mohamady (1.066)
  5. Carroll (0.994)
  6. Salcido (0.899)
  7. Barton (0.871)
  8. Nani (0.853)
  9. Etherington (0.830)
  10. Albrigton (0.810)
Notable goalscoring trends
  • Only midfielders have scored for Everton this season (plus one own goal)
  • Chelsea are yet to concede a goal from open play this season while City have conceded just one.
  • Birmingham have now conceded 3 own goals on the season, 30% of all their goals given up
  • Chelsea midfielders account for 14 goals this year, more than all but Arsenal and United's team totals.
  • Defenders have scored just 19 goals this year (10% of all goals). No defender has more than one and only 3 teams have account for more than one defensive goal (Birmingham (3), Stoke (2) and West Brom (2)).
  • Chelsea have three players - Drogba (6), Malouda (6) and Kalou (4) who have scored as many or more goals than strikers from Villa, Everton, Liverpool and Spurs have combined.
As we roll into the international week it gives us time to draw a breath and I will be looking to tweak my prediction tools to give us better projections for the coming weeks. Stay tuned to twitter or keep checking back for some thoughts on the upcoming value over the coming weeks and of course updates on any injuries that occur during the break.

Friday, October 1, 2010

Gameweek 7 Team of the Week

Alright I haven't done this before but alot of other blogs do the whole 'team of the week' thing and I thought it might be fun given that my actual team is pretty boring right now (Drogba, Malouda, Berbatov, Nani and maybe Bale are all locks to stay for now). The team has to fit in with the standard rules and my budget is whatever I can afford with my current squad (currently 103.3m). Without further ado then, here are my picks for this week:

GK B.Foster
DF D.Boyata
DF A.Hutton
DF S.Carr
MD F.Malouda
MD J.Milner
MD Nani
MD R.Van der Vaat
FW D.Drogba (c)
FW F.Torres
FW P.Odenwingie

GK S.Carson
DF A.Faye
DF G.Jara
MD S.Parker

I'm pretty happy with my first stab at a team of the week and the strategy is clearly to take advantage of value spot starters in defense to free up cash for a top line front line and midfield. Still, all my starting defenders rank in the top 5 in my clean sheet rankings but only cost a combined 18.2m. That allows me to keep the undroppable Malouda and Nani combo in midfield all the while pairing them with a couple of top players with decent fixtures: Van der Vaat and Milner. This week the front line almost picks itself as Drogba and Torres have great fixtures while Odenwingie is too good to pass up at his price tag.

Fancy your chances of beating me? Post your team below and I will review the results after the GW ends.